CA2134494C - Explicit synchronization for self-clocking glyph codes - Google Patents
Explicit synchronization for self-clocking glyph codesInfo
- Publication number
- CA2134494C CA2134494C CA002134494A CA2134494A CA2134494C CA 2134494 C CA2134494 C CA 2134494C CA 002134494 A CA002134494 A CA 002134494A CA 2134494 A CA2134494 A CA 2134494A CA 2134494 C CA2134494 C CA 2134494C
- Authority
- CA
- Canada
- Prior art keywords
- glyphs
- code
- data
- synchronization
- glyph
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1408—Methods for optical code recognition the method being specifically adapted for the type of code
- G06K7/143—Glyph-codes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K19/00—Record carriers for use with machines and with at least a part designed to carry digital markings
- G06K19/06—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code
- G06K19/06009—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code with optically detectable marking
- G06K19/06037—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code with optically detectable marking multi-dimensional coding
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Electromagnetism (AREA)
- General Health & Medical Sciences (AREA)
- Toxicology (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing For Digital Recording And Reproducing (AREA)
Abstract
A self-clocking glyph code is composed of data glyphs which encode logically ordered data values, together with one or more multi-glyph synchronization code patterns. Each of the synchronization code patterns is distinguishable from the data glyphs, but the synchronization code pattern or patterns provide an explicit spatial reference for the data glyphs. To this end, the glyphs of the synchronization code pattern or patterns are written on the recording medium in accordance with the same spatial formatting rule as the data glyphs, plus at least one additional formatting rule that establishes a predetermined geometric relationship between the glyphs of the synchronization code pattern or patterns and corresponding ones of the spatially formatted data glyphs, such that the logical order of at least a significant number ( ~ 1) of the data glyphs is inferable from the synchronization code pattern or patterns. Preferably, each of the synchronization code patterns is composed of glyphs that are selected from the same finite symbol set as the data glyphs, so the synchronization code pattern or patterns typically are defined by glyph patterns that have a suitablyhigh statistical probability of being unique with respect to any pattern of dataglyphs.
Description
~ 21 34494 EXPLICIT SYNCHRONIZATION FOR
SELF-CLOCKING GLYPH CODES
Field of the Invention This invention relates to self-clocking glyph codes for recording digital information on graphic recording media, such as plain paper. Even more specifically, this invention pertains to techniques for increasing the tolerance of such glyph codes to geometric distortion and physical damage, while preserving their homogeneous visual appearance and e~h~n~ing their suitability for applications requiring random access to the information they encode.
Background of the Invention In keeping with prior proposals, a basic self-clocking glyph code, such as shown in Fig. 1, typically is composed of an array 21 of elongated, slash-like symbols or "glyphs" 22 and 23 that are written with their longitudinal axes tilted at angles of a~ .lately -45~ and +45~ with respect to the vertical axis 24 of a recording medium 25 to encode bit values of "1" and "O", respectively, or vice-versa, in each of the glyphs. These codes are "self-clocking" because they include an optically detectable symbol (a "glyph" for each of the values they encode. This means that the detection of the glyphs implicitly synchronizes the coding process to the code once the decoding process has been properly spatially oriented with respect to the code. As will be evident, this implicit synchronization is valid so long as all of the glyphs are detected or otherwise accounted for in correct logical order. If, 21 344q4 however, synchronization is lost, there is nothing to restore the implicit synchronization, so a loss of synchroni7~tion generally causes an implicitly synchronized decoding process to experience fatal error. For a more detailed discussion of these glyph codes and of techniques for decoding them, reference may be made to the following commonly ~ igned U.S. patent docllments:
Bloomberg U.S. Patent No. 5,168,147 which issued December 1, 1992 on "Binary Image Processing for Decoding Self-Clocking Glyph Shape Codes"; Bloomberg et al. U.S. Patent No. 5,091,966, which issued February 25, 1992 on "Adaptive Scaling for Decoding Spatially Periodic SeLf-Clocking Glyph Shape Codes"; and Stearns et al, U.S. Patent No. 5,128,525, which issued July 7, 1992 on "Convolution Filt~ring for Decoding Self-Clocking Glyph Shape Codes".
As will be seen, these prior proposals not only describe the use of two discrimin~hle glyph shapes for the encoding of single bit binary values, but also teach that "glyph shape encoding" is extensible to the encoding of digital values of any given bit length, n, in each glyph by lltili7.ing a code having 2n dis~rimin~hl~ glyph shapes (where the discriminS~hility between glyphs representing different values is based upon the distinctive rotational orient,~ti--n.c~ and/or the distinctive geometric configurations of the different glyphs). Thus, even though a str~ightforward self-clocking glyph code of the type shown in Fig. 1 is featured in this disclosure to simplify the description, it will be evident that the broader aspects of this invention are applicable to other glyph codes, including more complex ones.In practice, each of the glyphs 22 and ~A
~ 21 34494 23 usually is defined by writing a recognizable pattern of "on" and "ofP' pixelsinto a two dimensional, array of pixel positions (i.e., a "symbol cell"). As a general rule, the pixels that form the body of the glyph (say, the "on" pixels) are written into contiguous pixel ~., 213~9~
positions that are more or less centered within the symbol cell, and the other or "off" pixels are written into the reln~ining pixel positions of the symbol cell to provide a contrasting surround that demarks the edges of the glyph and separates it from its neighbors.
6 The symbol cells, in turn, ordinarily are tiled onto the recording medium in accordance with a preselected spatial formatting rule, so the logical order of the data values that the glyphs encode is preserved by the spatial order in which the glyphs are mapped onto the recording medium. For example, the symbol cells may be written on the recording medium in accordance with a 10 regular and repeating spatial formatting rule that is selected to map the glyph encodings into a two dimensional, rectangular array of logical data blocks of predetermined size, such as data blocks having a 16 symbol cell x 16 symbol cell format. These data blocks suitably are organized on the recording medium in left-to-right, top-to-bottom logical order.
The size of the symbol cells that are used for the glyphs depends on a variety of factors, including the spatial resolution of the printing process that is employed to write the glyphs, the type and extent of the degeneration that the printed glyph code is required to tolerate, and the spatial resolution of the 20 lowest resolution sc~nning process that is expected to be able to recover thecode. In view of these constraints, a 300 s.p.i. printer suitably centers the glyphs in 5 pixel x 5 pixel or 7 pixel x 7 pixel symbol cells. Even larger symbol cells can be employed for the glyphs if the increased granularity of the textured appearance of the printed glyph code is tolerable or unavoidable. As a 25 general rule, of course, the smallest practical symbol cell size is esthetically most pleasing because the visual texturing appearance of the glyph code 213~4~
gradually blends as the symbol cell size is reduced. Indeed, codes composed of smaller symbol cells tend to have generally uniform gray scale appearances when they are viewed under ordinary lighting at normal viewing dist~nces.
The existing techniques for decoding self clocking glyph codes are designe~ to 5 be initi~ e-l at or near the center of a reference glyph that occupies a knownspatial position relative to the remainder of the glyph code (for example, the glyph in the upper, lefthand corner of a rectangular array of glyphs). Thus, accurately locating this reference glyph clearly is a key to spatially synchronizing such a decoding process with the glyph code.
While synchronous initialization of the decoding process is a necessary condition for orderly decoding of a self-clocking glyph code, it may not be a sufficient condition to ensure accurate decoding of the code because the scanned-in image of the glyph code pattern often is distorted by skew and/or 15 scaling errors. For this reason, prior decoding processes generally have attempted to determine the relative spatial positions of three or more referenceglyphs in recognizeable positions that have a known nominal, non-colinear relationship with respect to each other (such as the corner glyphs of a rectangular array of glyphs) with sufficient precision to compute skew and 20 scaling correction factors. These correction factors then are used to adjust the angle and magnitude of a vector that dictates the direction and magnitude of the jumps the decoding process makes while it is advancing from glyph-to-glyph.
25 As will be appreciated, anything that prevents the decoding process from establiching and maint~ining proper spatial synchronization with the glyph 2~3~43~
code threatens to defeat the decoding. Indeed, as a general rule, any glyphs that are isolated from synchronously readable glyphs may be unsynchronizable, especially in the interior of a glyph code pattern. Spatial synchronization is, of course, required to enable the decoding process to preserve the logical ordering of the data that is encoded in the glyphs. Minor 5 synchronization errors affecting a limited number of glyphs may be tolerable if the code has sufficient error correcting capacity (i. e., error correction code protection) to correct for the resulting decode errors, but more extended synchronization errors generally result in incomplete reading of the encoded information. Ultimately, this performance degradation can become 10 sufficiently severe to be unacceptable.
Accordingly, it will be evident that one of the more significant disadvantages of spatially synchronizing a glyph decoding process through the use of a global imaginal reference that is computed from the spatial locations of certain key 15 glyphs, such as the corner glyphs of a rectangular glyph pattern, is that various factors may prevent the spatial location of one or more of those key glyphs from being determined with sufficient precision to compute a valid reference, thereby preventing the decoding process from achieving and/or maint~ining spatial synchronism with the glyph code. Moreover, even when a 20 valid glo~al reference can be computed, there is an appreciable risk that thedecoding process will lose synchronization. For example, the scanned-in image of the glyph code might be sufficiently degraded to cause the decoding process to erroneously branch from one row of glyphs to an adjacent row, or to skip overa glyph without accounting for it, or to miscount the number of unreadable 25 glyphs in a damaged region of the image. Unfortunately, none of the known self-clocking glyph codes or related decoding processes enable the decoding 2~3~49~
process to re-establish proper synchronization once it has been lost, so a loss of synchronization usually causes these existing decoding processes to degenerate into an irreversible failure mode.
The problems that have been encountered with establi~hing and m~int,~inin~
5 proper spatial synchronization during the decoding of self-clocking glyph codes have been compounded by the lack of reliable techniques for detecting asynchronous operation of the decoder. Glyph centers have been counted during the decoding of these codes on the theory that a miscount over any predefined segment of the formatted code will signal a loss of synchroni~qtion.
10 This is a weak synchronization validation process if part of the glyph code pattern (e. g., a portion of its periphery) is missing or otherwise undetectable.
there may be an ambiguity with respect to the number of glyphs that are present in the image and with respect to what part of the glyph pattern is present in the image. Thus, the decoding process may provide invalid results, 15 without providing any warning that the results are invalid.
Image rotation of 90~, 180~, 270~ and reflection result in rotational and mirrorimage transformations of the glyph code pattern, and these transformations can alter the apparent spatial order of the glyphs and/or the apparent 20 rotational orientation of the individual glyphs. Indeed, this is an especially significant issue when the glyphs encode the data values in their rotational orientations, such as when the glyphs are slash-like symbols that are oriented at +45~ with respect to the longitudinal axis of the recording medium to encode high ("1") and low (~0") logic level bits, respectively, because such 25 glyphs often are written in patterns that have rotational symmetry.
Unfortunately, the orientation of scanned-in glyph code patterns frequently ~ 213~
involves some ambiguity. For example, there typically is a four-fold rotational ambiguity in the case of a glyph code that is composed of a square array of slash-like glyphs of the above-described type that are written into square symbol cells7 and a two-fold ambiguity in the case of a glyph code that is composed of a rectangular array of such glyphs.
Some of the commercially available two dimensional data codes, such as VericodeTM by Veritex and DatacodeTM by Data Code International, have distinctive reference marks along one or more of their borders that can be used for scaling and for explicitly spatially synchronizing the decoding of those 10 codes However, these reference marks are visually distinctive so they tend to increase the visual obtrusiveness of those codes. One of the principal advantages of self-clocking glyph codes is that they tend to be esthetically pleasing because of their non-obtrusive visual appearance, so it would be counterproductive to supplement them with reference patterns that have such 15 a strong visual impact.
Summary of the Invention In accordance with the present invention, a self-clocking glyph code is 20 composed of data glyphs which encode logically ordered data values, together with one or more multi-glyph synchronization code patterns. Each of the synchronization code patterns is distinguishable from the data glyphs, but the synchronization code pattern or patterns provide an explicit spatial reference for the data glyphs. To this end, the glyphs of the synchronization code 25 pattern or patterns are written on the recording medium in accordance with the same spatial formatting rule as the data glyphs, plus at least one 21 34~94 Additional form~t+ing rule that establishes a predetermined geometric relationship between the glyphs of the synchronization code pattern or patterns and corresponding ones of the spatially formatted data glyphs, such that the logical order of at least a .qignifi~nt. number (~ 1) of the data glyphs is inferable from the synchronization code pattern or patterns. Preferably, each of the synchronization code patterns is composed of glyphs that are selected from the same finite symbol set as the data glyphs, so the synchronization code pattern or patterns typically are defined by glyph patterns that have a suitably high statistical probability of being unique with respect to any pattern of data glyphs.
Other aspects of this invention are as follows:
A process for spatially referencing a self-clocking glyph code that is composed of logically ordered data glyphs that are selected from a predetermined symbol set for encoding information in said data glyphs; said process comprising writing said logic~lly ordered data glyphs in a two dimen~ l field on a recor~ing medium in accordance with a predetermined spatial formatting rule;
writing a syncl~rolli~ation code in said field, said syncl~rol~i;~tion code being composed of logically ordered glyphs that are distinguishable from said data glyphs, said synchrol~i;~tion code being written in said field in accordance ~ 21 34494 with said spatial formatting rule plus at least one additional rule that establishes a predetermined geometric relationship among the glyphs of said synchronization code and said data glyphs, whereby the logical ordering of at least some of the data glyphs in said field is inferable from the glyphs of saidsynchronization code based upon said geometric relationship.
An optically readable record for storing encoded information; said record comprising a recording medium;
a self-clocking data code for encoding said information; said data code being composed of data glyphs that are written on said recording medium on centers that are spatially distributed in nominal accordance with a two dimensional spatial formatting rule; said data glyphs being ~lçfine-l by respective symbols that are selected from a finite set of optically discrimin~hle symbols to encode said information; and at least one self-clocking synchronization code for establi~hing a persistent spatial reference for said data code; said synchronization code beingdiscrimin~ble from said data code and being composed of additional glyphs that are written on said recording medium in nominal accordance with said spatial formatting rule plus at least one additional rule that establishes a predetermined geometric relationship among the glyphs of the synchronization code and said data glyphs, whereby the logical ordering of at 9a r~
, ~
lesst some of the data glyphs is inferable from their geometric relationship to the glyphs of the synchronization code.
An optically readable record for storing encoded information; said record comprising a recording medium;
a self-clocking glyph code for storing said information on said recording medium; said glyph code being composed of a data code and at least one a synchronization pattern for providing a persistent spatial reference for said data code said data code being composed of a first logically ordered set of discrete data glyphs that are written on said recording medium on spatially offset centers in nominal accordance with a two dimensional spatial formatting rule;
said data glyphs being defined by respective symbols that are selected from a finite set of optically discrimin~ble symbols to encode said information; and said synchronization pattern being discrimin~hle from said data code and being composed of a second logically ordered set of glyphs that are written on said recording medium in nominal accordance with said spatial formS~tting rule plus at least one additional rule that establishes a predetermined geometric relationship among the glyphs of the synchronization pattern and said data glyphs, whereby the logical ordering of at least some of the data 9b ~ 2 1 3~494 glyphs is inferable from their geometric relationship to the glyphs of the synchronization pattern.
An optically readable record for storing encoded information; said record comprlsmg a recording medium;
a self-clocking data code for encoding said information; said data code being composed of data glyphs that are written on said recording medium on centers that are spatially distributed in nominal accordance with a two dimensional spatial formatting rule; said data glyphs being defined by respective symbols that are selected from a finite set of optically discrimin~hle symbols to encode said information; and a synchronization code pattern composed of at least two intersecting nonparallel synchronization code subpatterns for establi~hing a two dimensional, persistent spatial reference for said data code, each of said synchronization patterns being discrimin~hle from said data code and being composed of glyphs that are written on said recording medium in accordance with said spatial formatting rule plus at least one additional rule that establishes a predetermined geometric relationship between the glyphs of the said synchronization patterns and said data glyphs, whereby the logical ordering of at least some of the data glyphs is inferable from their geometric relationship to the glyphs of the synchronization patterns.
gc ~' 2~ 34494 Brief Description of the Drawings Still further objects and advantages of this invention will become apparent when the following detailed description is read in conjunction with the attached drawings, in which:
Fig. 1 is a simplified, enlarged illustration of a standard self-clocking glyph code that is written in accordance with a basic two dimensional block formatting rule;
Fig. 2 is a simplified, enlarged illustration of a block formatted self-clockingglyph code that includes synchronization code patterns in accordance with this invention;
~A
~ 2~3~4~
Fig. 3 is a simplified, enlarged illustration of a block formatted self-clockingglyph code that contains non-colinear synchronization code patterns in accordance with a more detailed aspect of this invention;
5 Fig. 4 is a simplified, enlarged illustration of a glyph code pattern that has a lattice-like framework of two dimensional sync code sequences in accordance with an even more detailed aspect of this invention;
Fig. 5 is a simplified, enlarged illustration of a glyph code pattern that has a10 lattice-like framework of interlocking, two dimensional sync codes;
Figs. 6 and 7 are full size illustrations of the sync code and data code components, respectively, of a glyph code pattern of the type shown in Figs. 4 and 5;
Fig. 8 is a full size illustration of a composite glyph code pattern that is composed of the sync code and data code components that are shown in Figs. 6 and 7, respectively;
20 Fig. 9 is a simplified, enlarged illustration of a multiple logical data block formatted glyph code pattern that has a lattice-like framework of two dimensional sync codes and internal block labels for extended addressability;
Fig. 10 diagrammatically illustrates the use of asymmetric sync code 25 sequences for symmetry disambiguation;
~ ~ 2~34~
Fig. 11 illustrates the use of a lattice-like sync code pattern for navigating around damage in glyph code patterns; and Fig. 12 illustrates the use of a lattice-like sync code pattern for explicitly 5 synchronizing a distorted glyph code pattern.
Detailed Description of the Illustrated Embodiments While the invention is described in some detail hereinbelow with reference to 10 certain illustrated embodiments, it is to be understood that there is no intent to limit it to those embodiments. On the contrary, the intent is to cover all modifications, alternatives and equivalents ~alling within the spirit and scope of the invention as defined by the appended claims.
15 In accordance with this invention, as shown in Fig. 2, a self clocking glyph code 31 includes a data code 32, together with a two dimensional, distributed sync code pattern 33. For illustrative clarity, the data code 32 is shown as being composed of slash-like symbols or "glyphs" 34 that are tilted at -45~ with respect to the longitudinal axis of the recording medium 35 (see the process 20 direction arrow 36), while the sync code pattern 33 is shown as being composed of glyphs 37 that are tilted at +45~ with respect to that axis. In practice, of course the angular orientation of the data glyphs 34 is modulated in accordance with the respective data values they encode. Further, the angular orientation of the sync code glyphs 37 might also be modulated in accordance 25 with a cyclic or acyclic sequence of values, as more fully described hereinbelow.
Consequently, there may not be any symbolic distinction between a given data 213~94 glyph 34 and a given sync glyph 37. Indeed, in keeping with one of the more detailed features of this invention, the data glyphs 34 and the sync glyphs 37 preferably are potentially identical members of the same finite symbol set because this causes the sync code pattern 33 to visually blend with the data code 32 (see Fig. 8) effectively. Moreover, to even further enhance the visual 5 homogeneity of the composite code 31, the sync code pattern 33 advantageously is selected to avoid visually distinctive glyph patterns, such aslong runs of the same symbol.
More particularly, in keeping with this invention, the sync code encoded glyph 10 pattern 33 (sometimes referred to herein as the "sync code pattern") is composed of a predefined, cyclical or acyclical pattern of glyphs that is sufficiently distinctive to be disambiguated from the data code 32 with acceptably high statistical reliability. This sync code pattern 33 is written onthe recording medium 35 in accordance with the same spatial formatting rule 15 as the data code 32, but the formatting of the sync code pattern 33 is further governed by at least one additional rule that establishes a predetermined localized geometric relationship between the spatial positioning of the glyphs 37 of the sync code pattern 33 and the spatial positioning of nearby data glyphs34. The logical ordering of the sync code values that are encoded by the glyphs 20 of the sync code pattern 33 is predetermined. Thus, the sync code pattern 33 defines localized spatial references for determining the logical ordering of thedata glyphs 34.
As will be appreciated, the replicability of the sync code pattern 33 ensures 25 that the spatial references it provides survive replication of the composite code 31 (such as when the composite code 31 is replicated by photocopying or 213~94 .
farsimile tr~n~mi~ion to provide a hardcopy replica or by bitmap input sc~nnin~ to provide an electronic representation) subject only to the losses that might occur if the original image and/or the copy of their composite code 31 is ~l~m~ged. Advantageously, the sync code pattern 33 is distributed throughout the composite code 31 and is selected so that the density of the spatial 5 references it provides is sufficient to locally reference all of the data glyphs 34.
In that event, any non-linear distortions the data code 32 might experience while being replicated are tracked by corresponding localized non-linear distortions of the sync code pattern 33.
10 In theory, the sync code pattern 33 may take many different forms. In practice, however, it advantageously is written as a linear pattern (or, for a nonlinear implementation, in areas of adjacent glyph cells) because these type~ of subpatterns are easier to identify than nonlinear patterns of spatially scattered glyphs. For example, the sync code pattern 33 may be composed of 15 linear subpatterns of glyphs 37 that are written on the same center-to-centerspacing as the data glyphs 34 to span the spatially formatted glyph code 31 (or logical subsets thereof- see Fig. 4) in one direction (see Fig. 2) or in a plurality of different directions (see Fig. 3 et seq.).
20 Focusing on linear sync code glyph patterns (i. e., patterns of sync code encode glyphs that are written in linear spatial sequences), it will be understood thatsuch a sync code pattern may be defined somewhat more generally by a two dimensional array of linear sync code subpatterns, where each of the subpatterns is composed of every mth glyph (where m is an integer) along a 25 substantially straight line that runs parallel to a predetermined synchronization direction. The degenerative case, where m = 1, provides fine ~ 213~94 grain explicit synchronization because the sync code pattern is composed of adjacent glyphs. On occasion, however, a coarser explicit synchronization may suffice because the implicit synchronization that is provided by the self-clocking properties of the data glyphs 34 is valid whenever the presence of the data glyphs is detectable reliably, regardless or whether the detected data 5 glyphs can be accurately decoded or not. In any event, when linear sync code subpatterns of a given granularity, m, are employed, each set of glyphs that is (1) aligned in a synchronization direction and (2) composed of glyphs that are separated from each other by m - 1 intermediate glyphs, implicitly includes either all sync glyphs in proper sequence or no sync glyphs, except where 10 desynchronizing faults (i. e., synchronization errors) are encountered.
Sync code patterns composed of parallel, longitudinally separated, linear subpatterns, such as the pattern 33 in Fig. 2, are an interesting example of how the inclusion of a pattern of sync code encoded glyphs facilitates 15 establi~hing and maint~ining synchronization during the decoding of a self-clocking code 32. In this embodiment, the sync code subpatterns 33a -33d are written in parallel alignment with a major axes of the data code 32 (i. e., its lateral or x -axis in a standard orthogonal coordinate space), and these subpatterns 33a -33d are composed of sync code encoded glyphs 37 that are 20 written on the same center-to-center spacing along the x-axis as the data glyphs 34. Further, the sync code subpatterns 33a -33d each is of sufficient length to span the data code 32 along its x-axis. And, these sync code subpatterns 33a -33d are spatially displaced from each other along the orthogonal axis of the composite code pattern 31 (i. e., its longitudinal axis or 25 y-axis) by a predetermined constant number of lines of data code 32.
- ~ 213~4~
Given the x, y coordinates on which the glyphs 37 of any two neighboring sync code subpatterns (such as the subpatterns 33a and 33b) are centered, prospective center locations for the data glyphs 34 that reside therebetween can be identified by pairing the glyphs 37 that occupy logically corresponding 6 positions in the two sync code subpatterns with each other and by then linearly interpolating between the x, y coordinates of these logically paired glyphs.
This process provides strong explicit synchronization axially of the sync code subpatterns 33a - 33d because it is sensitive to nonline~r distortion along thataxis. Weaker synchronization is provided orthogonally to the subpatterns 33a 10 - 33d because orthogonal positioning is determined by interpolation from the spacing of the neighboring sync code subpatterns 33a - 33d..
Fortunately, the implicit synchronization that is provided by the self-clocking property of the data glyphs 34 often is sufficient for maint~ining the 15 synchronization of the decode process while the data glyphs 34 that are disposed between the parallel sync code subpatterns 33a-33d are being read.
For example, the data glyphs 34 that are logically aligned between sync code glyphs 37 that occupy logically corresponding positions in two neighboring sync code subpatterns, suitably are identified on a "best efforts basis," (1) by20 explicitly resynchronizing the decode process at one of the logically paired sync code glyphs 37, and (2) by then advancing the decode process, data glyph-by-data glyph, toward the second of the paired sync code glyphs 37. This process isrepeated for one after another of the logical pairings of the sync code glyphs 37.
It, of course, will be understood that this is a "best efforts" process because 26 local damage can frustrate the attempt to identify some of the data glyphs 34with the precision that is required to reliably determine their logical ordering.
~ 2~3~
However, synchronization losses are easily detected. In that connection, it willbe noted that the number of data glyphs 34 that are read while the decode process is searching for data glyphs 34 that are logically aligned between any two logically paired sync code glyphs 37 is a reliable intlir<qtion of whether the anticipated number of data glyphs 34 was found or not. Moreover, the success 5 or failure of the decode process in tracking from one to the other of the logically paired sync code glyphs 37 is a strong indication of whether synchronization was preserved during the reading of the intervening data glyphs 34. Thus, if synchronization errors occur, suitable remedial and/or corrective action can be taken. This action may be transparent to the user, such as when there is 10 sufficient error correction capacity to correct the errors or may involve more drastic measures, such as notifying the user that the decode results are invalidand/or incomplete.
As shown in Fig. 3, the computational uncertainty of relying on data glyphs 34 15 as implicit spatial synchronization references can be avoided by employing a sync code pattern 43 that is composed of a plurality of nonparallel linear sync code subpatterns 43a - 43b that jointly span the composite code pattern 31( (e.g., along its x-axis and its y-axis). Two such subpatterns 43a and 43b are sufficient for reducing the synchronization errors that may tend to occur; for 20 example, as a result of one or two dimensional distortion of the image of thecomposite code pattern 31. Moreover, if the sync codes that are employed provide unique addresses over some definite span of each of the sync code subpatterns, such as is provided by the some of the digital codes that are discussed hereinbelow, a single pair of nonparallel sync lines can further 2~ reduce synchronization errors caused by dropouts of individual data glyphs 34, groups of neighboring data glyphs, and/or full lines of data glyphs.
213~4~4 .
Linearly extrapolating from the x and y coordinates of the glyphs 37 of the sync code subpatterns 43a and 43b in directions parallel to the y-axis and x-axis, respectively, of the composite code pattern 31 yields lines that intersect at prospective center locations for the data glyphs 34, thereby providing explicit 5 spatial synchronization references for the data glyphs 34. In practice, the scanned in image of the composite code pattern 31 may be skewed somewhat, but any slight skew error may be elimin.qted if the angular orientations of the sync code subpatterns 43a and 43b relative to the x and y axes of the composite code pattern 31 is known. Specifically, that information can be used to compute 10 a global or local skew correction factor that brings the sync code strings 43a and 43b into the proper angular orientation with the x and y axes of the composite code pattern 31. While any two nonparallel sync code strings that jointly span the composite code pattern 31 along its major axes will suffice, two perpendicular sync code strings 43a and 43b that span the composite code 15 pattern 31 along its x-axis and its y-axis, respectively, are shown to provide a straightforward example.
Still further, as shown in Figs 4 - 12, two non-colinear sets of parallel linearsync code subpatterns can be used to form a lattice-like sync code pattern 53 20 that is composed of a two dimensional tiled array of sync code cells 54a - 54i that have rectangular or parallelogram-like configurations. The x, y coordinates on which the glyphs 37 of such a sync code pattern are centered may be identified, and the glyphs 37 may be decoded, systematically throughout the composite code pattern 31 by following the sync code lattice.
25 Since there are four paths to most vertices of this lattice, almost any givenglyph 37 of the sync code pattern 53 can be reached by following a path of ~ 2134~9~
logically ordered sync code glyphs ~this notion of imposing a predetermined logical order on the values of neighboring glyphs in the sync code pattern 53 isdiscussed in more detail hereinbelow), even when damage frustrates any attempt to determine the logical ordering of the glyphs 37 along one or more of the alternative paths. For example, see Fig.8 where the overwrites 81,82, and 6 83 partially obliterate portions of the lattice-like sync code pattern 53. The closed line geometry of the overwrite 81 may prevent the sync glyphs within its interior from being reached, but it is possible to navigate around the open line overwrites 82 and 83 on sync code glyphs.
As will be appreciated, the use a two dimensional sync code lattice, such as thesync code pattern 53, permits positive verification of the number of data glyphs34 that are read in a linear array of glyphs, such as a row or column of data glyphs 34 that has both ends terminated by sync code glyphs 37. This, in turn, means that such a sync code pattern permits substantial verification that branching to an adjacent row of glyphs has not occurred during the reading of the data glyphs 34 that are fully confined by the sync code lattice.
Advantageously, the dimensions of the lattice cells 54a - 54i (as measured by sync code glyph count) are uniform because this permits the periodicity of the cyclical synchronization codes that are discussed below to be matched to the periodicity of the sync code pattern 53. While Fig 4 shows that each cell 54a -54i of the sync code pattern 53 contains a 6 x 6 array of data glyphs 34, it will be understood that any M x N lattice may be employed in practice. For example, each of the cells 54a -54i of the lattice pattern 53 suitably frames a logical data block composed of a 16 x 16 array of data glyphs 34, thereby reducing the overhead cost of including the sync code pattern 53 to provide the ~ 213~4 spatial references that are required to explicitly synchronize the decoding of the data glyphs 34.
DIGITAL CODES FOR SYNC GLYPH CODES.
5 As is known, digital codes of various descriptions have been extensively utilized to synchronize communication ~ign~l~. This usually is a one dimensional synchronization problem. The present invention, on the other hand, deals with a two dimensional synchronization problem. As will be seen, however, it has been found that very effective two dimensional 10 synchronization code patterns can be implemented by using combinations of one dimensional digital codes. Moreover, it has been further found that some of these two dimensional sync code patterns have attractive attributes for more detailed applications, including absolute and relative logical addressing of thestructure and/or information content of self-clocking glyph codes, orientational15 disambiguation of the information content of such codes, and enh~nced tolerance to damage and/or distortion.
There clearly is a fundamental advantage to associating digital sync codes with glyph patterns that are spatially well ordered and relatively easily 20 identified. The above-described linear glyph code patterns are good examples of patterns that satisfy these general criteria .
Indeed, as will be recalled, even parallel line sync code patterns (see Fig. 2) can provide synchronization information with respect to two dimensions: i. e., 25 along and transverse to the lines. Furthermore, it will be evident that the positive confirmation that a sequence of glyphs on a line encode a predefined 213~4 digital sync code makes the line a registration mark with respect to transverse position; e.g. the edge of a glyph subregion. Simple repetitive digital codes such as 11111111.. or 10101010... can be used for this purpose. Theyclearly are simple to detect, but they provide limited addressing information along the line and minirn~l absolute addressing information. Moreover, the high symmetry of these repetitive codes provides little aid in resolving orientation ambiguity (this subject is discussed in more detail hereinbelow).
It also will be understood that the intersections of sync code patterns that arecomposed of nonparallel linear strings of sync code (Figs. 3 - ~ can serve as 10 references for synchronization, even if the digital codes themselves lack unique addressing, such as in the simple code sequences mentioned hereinabove. For example two sets of intersecting linear subpatterns of glyphs that encode sync codes, as in Fig. 4, can provide some two dimensional absolute registration capability. Speçific~lly, the confirmation that each of these linear 15 glyph strings encode sync code establishes linear boundaries for a glyph region or cell that is framed by sync code glyphs, and that framed region is known (because the spatial formatting rule is known) to have a predefined pattern of interior glyph positions.
20 Cyclic digital code sequences that conform to so-called m~im~l length sequences (MLS) can be generated by appropriately configured digital shift registers or by reference to a bit-level description of the desired MLS (see, Shift Register Sequences by Solomon W. Golomb, Holden-Day, Inc. 1967). It now has been found that these MLS digital sequences have properties that make them 25 very useful for more sophisticated proce~qing of two dimensional glyph codes.
213~4~
More particularly, it is known that an n-stage shift register that produces a MLS will generate a binary sequence of length L=2n-1, in which every subsequence of length n is unique. Thus any subsequence of at least length 1 =n, llefine.s position along the sequence and, therefore, a local 5 synchronization reference and a local address with a periodic ambiguity of L.
Continuation of such a shift register sequence is cyclically periodic with period L. For example, if n = 3, then L = 7 and a corresponding MLS shift register bit sequence is S= 00111010011101.... Thus, it will be evident that n may be easily chosen, if desired, to be large enough so that there is no repeat of the bit 10 sequence S within a bit string of any given finite length. More generally, itwill be observed that when a spatially ordered array of glyphs encode the values of one after another of the bits of a non-repeating shift register subsequence in logical order, the bit values that are encoded by the glyphs provide absolute one dimensional addressing transversely of the array of 15 glyphs. As will be appreciated, this absolute addressing property follows from the mapping of the logically ordered, non-repeating shift register subsequence on to the spatially ordered pattern of glyphs, so it is not limited to linear patterns of sync code glyphs.
20 As a general rule, MLS sequences are not symmetric for n>2. Although an address d~fining MLS subsequence of length n may be symmetric in the forward and reverse directions (e.g., 101), extending the sequence to no more than length 2n will break the symmetry (e. g., 101001). As will be seen, this property of MLS codes can aid in glyph pattern symmetry r~ nbiguation.
213~494 .
As previously pointed out, an n-stage shift register m~xim~l length sequence has a cyclic period of bit length L=2n-1,which is exactly one less than a "round" binary number. Thus, for n= 1, 2,3, 4, 5,6,7, ...; L= 1,3,7,15, 51,63, 127,.... These bit lengths do not naturally match the round binary units in 5 which digital data is typically assembled; e.g., nibbles(4), bytes(8), blocks(512x8). This is me~nin~ful because it implies that MLS codes have the drawback of not precisely sp~nning data blocks that are arrayed in convenient binary units.
10 Fortunately, however, it has been shown that for an n- stage shift register with a m~qxim~ql length sequence of 2n- 1 that every cycle period between 1 and 2n can be obtained from it by one of 2n- 1 simple st~nfl~qrd modifications to a nonlinear shift register [Golomb, op. ct. p. 192 - 194]. In many cases, the resulting sequence is a subsequence of a m~ximal length sequence. For 15 example:
~xim~l n=4, L=15, S=111000100110101111000....
Truncated L' =8, S= 11100010.111000....
M~xim~l n=5, L=31, S=0000100101100111110001101110101.
TruncatedL'=17, S=00001001011001111.00001... (Note, each 5 bit subsequence is unique) In general, these cyclic codes, including the cyclic truncated sequences, provide 25 unique n bit subsequence addressing within each cycle (i. e., within each local region of the printed glyph code).
21~9~
An n-stage shift register with a m~im~l length sequence of L=2n- 1 generally contains a subsequence of (n - 1) O's adjacent to (n) 1's. Thus, such a sequence can be padded by inserting one or more (i. e., "r") additional O's adjacent to the (n - 1) O's. Adding one 0, yields a sequence of period length 5 (L' =2n), a round binary value. Thus, padding of this type can be used in a glyph code to align each cycle of sync code with a block of data code that is written in a convenient binary format. Inserting two or more O's enables alignment with (1) a data block having an ~'even" binary format, together with (2) one or more regions of additional glyphs which may be used for separate 10 purposes, such as transverse lines of sync code glyphs. As will be seen, thiszero padding has the useful property of preserving the uniqueness of all n-bit subsequences, except for the string of O's that is immediately followed by and unambiguously identified by the (n) 1's, as unambiguous addresses along the code.
Examples:
n=2, r=0, L=3, S=011011 (Cautions: symmetric) n=2,r=l,L'=4,S=00110011 (Cautions: bit inversion= reverse of sequence; symmetric) n=2, r=2, L' =5, S =0001100011.
n=3, r=0, L' =7, S =00111010011101 n=3,r=1,L'=8,S=0001110100011101 [Caution: bit inversion =sequence reverse]
n=3,r=2, L'=9, S=000011101000011101.
2~3~
.
n=4, r=0, L' = 16, S=000111101011001000111101011001 n=4, r= 1, L' = 16, S=00001111010110010000111101011001 n=4, r=2, L' = 17, S=0000011110101100100000111101011001 5 It also has been recognized that a periodic sync code that provides unique addressing within a subsequence of predetermined length (such as an MLS
code) can be employed to provide directional steering to the glyph location for an intersecting transverse pattern of sync glyphs. Thus, when a periodic sync code of the foregoing type is employed, it is useful to match the spacing of 10 transverse lines of sync glyphs to the period of the periodic sync code so that there is consistent transverse referencing between the periodic sync code and the transverse sync glyphs.
The maintenance of consistent transverse references between intersecting 15 strings of sync code is, of course, extensible to some of the more interesting glyph code patterns, including the lattice-like glyph code patterns shown in Figs. 4 - 8 . More particularly, it will be evident that two intersecting sets of parallel glyph code subpatterns that encode periodic sync codes may be spaced at periodic intervals fitting both sync codes. Moreover, it can be shown that a 20 spacing that matches the periodic intervals of both codes yields an interlocking periodic grid of sync with the same bit patterns in the neighborhood of each intersection. Indeed, this is illustrated in Fig. 5 for the code 0011101 in a square lattice. And, it also applies to Figs. 6 - 8 where the sync code pattern is a square lattice of glyph strings that encode the 17 bit cycle code with two zero 25 bit ps3~1rling discussed above. Figs. 6 and 7 show the sync lattice and the data glyphs printed separately, while Fig. 8 illustrates the combined or composite 213~94 glyph code which shows that the sync code visually blends uniformly with the variable data encoded glyphs.
On a somewhat more general level, if intersecting strings glyphs encode linear address information and share common glyphs, the values encoded in the 6 shared glyphs preferably are consistent with both of the encoded sequences.
On the other hand where the sync codes do not otherwise provide appreciable unique addressing, as in simple codes(ll1 or 1010), it may be advantageous to arrange the intersection glyphs to be discordant with one or both of the sync sequences to facilitate the identification of the intersection location; e.g., 1111101111111, or 101010001010.
As shown in Fig. 9, the translational ambiguity of a repeating periodic s.ync sequence can be overcome by writing additional predefined data in "reserved"
glyphs 58a-58i which have a known order and which reside at predefined 15 positions with respect to the encoded sync code glyphs. For robustness, the glyphs that are reserved for encoding this additional information preferably are spatially adjacent to the sync code glyphs 37. The additional information that might be encoded in these reserved glyphs includes physical and/or logical data block identification. For example, the cyclic sync codes that are 20 contemplated by this invention can be supplemented by encoding appropriate data block specific information in these reserved glyphs 58a - 68i to enable thesync code to be used for synchronizing the of data glyphs, on a data block basisand the reserved glyphs 58a - 58i to be employed for identifying the data blocksphysically with respect to the overall glyph code pattern. To this end, in Fig. 9 25 there are nine data blocks 54a - 54i which are surrounded by a cyclic MLS code [n=4, L' = 16 (padded with a single zero)] and labeled with a byte of data in --2~--213~4~
.
the upper left corner giving block coordinates in the form of two hex code numbers; e.g. 00100011 corresponds to 23, which is a suitable identifier for the data block in the second column and third row of the tiled format.
Physical addressing of the data blocks 54a - 54i provides a powerful in-le~in~
5 mec~ni~m for random access to the data recorded in any of the data blocks 54a - 54i. Furthermore, if desired, this inde2~ing may be e~tenfle~
hierarchically to include, for example, a page index value, a volume index value and similar logical correlators in the data that is written into the reserved glyphs 58a - 58i. Standard random access techniques can then be 10 employed for advancing the glyph reading process from one data block to another data block within the same glyph code pattern or a different glyph code pattern. As will be appreciated, the ability to index reliably and efficiently from one glyph code pattern to another is dependent on having a substantially stable and/or easily reconstructable relationship between those 15 patterns, so that the indexing can be performed using straight forward mathematical functions (i. e., addition and subtraction). Image distortions and other factors may perturb the physical relationship among well-linked glyph code patterns slightly, so it sometimes may be best to perform random access seeks using a convergence strategy; whereby each seek starts with a coarse 20 indexing to at least roughly approach the target location, and then continueswith one or more progressively finer seeks until the target is located precisely.
As will be appreciated, the indexing data in the reserved glyphs 58a-58i can be used to control the coarse seeks, while the finer seeks can be controlled using the precision provided by the above-described addressing codes.
213~g4 .
Considering the issues involved in f~i~Ambiguating the sync code glyphs 37 from the data glyphs 34 in some additional detail, the general rule is that the ability to distinguish a predefined ordered subset of k glyphs that encode sync code from glyphs that encode variable data with a r2qn-1om occurrence of bits is5 governed by the probability of random match of all bits, 1/2k, which decreasesexponentially (geometrically) with k. Even for nonrandom data, matching to a pseudorandom synchronization digital code, such as the m~im~l length sequences, will be ~imil~rly improbable. Thus, the first key test for disambiguating sync glyphs from data glyphs focuses on determininE~ whether 10 a given spatially ordered sequence of glyphs encodes values of the known sync code in valid logical order.
As will be appreciated, the linear sync glyph patterns that are proposed hereinabove facilitate this logically ordered sorting of glyphs, but there still is 15 an uncertainty of translation by an integer number of glyphs, together with apossible gross rotational ambiguity as more fully described hereinbelow.
Therefore, when a cyclic sync code having a period of L bits is employed, the testing to identify the glyphs that are encoded with this sync code suitably begins with a search for glyphs that encode a logically ordered subsequence of 20 k bits that matches on the known sync code, where k is long enough to uniquely define synchronized address within a single period of the sync code.
Then the probability of this subsequence randomly matching a translation phase of the sync code is L / 2k of course, if k + L is equal to or greater than m, where m is the total number of sync glyphs in a full line of sync code, then the25 number of valid sequences is reduced from L to (m - k), yielding a random match probability, of (m - k)/ 2k.
2134~94 Once a tentatively valid sync code subsequence is found, a follow-on test typically is performed to confirm that the particular subsequence continues to match on the known sync code as it is being extended by the decode values for one after another of the glyphs that logically extend the subsequence in 5 accordance with the spatial formatting rules. In other words, the apparent sync code pattern is further delineated and verified by confirming that the decode values it yields for subsequences of increasing k continue to match on the known sync code. ~enerally, adequate sync distinguishment is achieved when the probability of mistakenly identifying data as sync is deemed to be 10 acceptable. For most applications, this condition can be satisfied while using a moderately sized k for sync verification because of the fast growth of the exponential probability dependence on k.
Verification of known cyclic sync patterns is further facilitated by the 15 bidirectional (i. e., forward and backward), bit-by-bit verifiability of the logical extensions of such patterns. Furthermore, the use of two dimensional lattices of sync glyphs that encode interlocking cyclic sync codes, as described hereinabove, facilitates robust distinguishment of the sync pattern by providing connectivity among linear strings of sync glyph. As shown in Figs.
20 11 and 12, the additional sync verification tools that are provided by the bidirectional and interlocking logical attributes of this class of sync code areespecially valuable for reestablishing explicit spatial synchronization for the decoding of glyph code when the sync code pattern is disrupted (i. e., logicallybroken erroneously) by local damage and/or degradation of the glyph code.
2134~
As will be understood, absolute verification of synchronization may also be achieved on finite area bounded glyph code patterns, such as shown in Fig. 3, by pre~efining and detecting a geometrically unique subpattern of sync code glyphs for which there is no congruent pattern of glyph centers. In the 5 absence of glyph "dropouts", such a sync code pattern can be verified in the image domain, without having to perform any pattern matching at the bit level (i. e., in the digital domain). For e~ mple, if the glyph code is written on a rectangular array of glyph centers, a sync code pattern that is written into a predetermined column and row of glyphs to span the array completely (i. e., 10 including the border glyphs, as in Figure 3) is geometrically identifiable assync code. Such a geometrically identifiable sync code pattern also provides a degree of absolute 2D addressing, referenced by counting glyphs off this sync pattern, rather than by reference to a boundary feature.
15 It has been found that sync code encoded glyph patterns also can be used to identify and correct ambiguities in glyph codes due to symmetries that exist in the pattern of glyph centers with respect to operations such as image rotations (e.g., 90~, 180~, 270~) and/or image reflections. When these orientational ambiguities transform individual glyphs into permissible members of the 20 glyph code, disambiguation is not trivial.
For example, slash-like glyphs that are tilted to the left or right at 45~ ( i. e., +45~ and -45~, respectively) to encode bit values of "0" or "1," respectively, might be written in accordance with a spatial formatting rule that maps their 25 centers into a square lattice array. As previously pointed out such a glyph pattern has a four-fold rotational ambiguity. Specifically, a 180~ rotation of 2134~9~
. ~
the glyph code pattern m~int~in~ the identities of the individual glyphs because each of the glyphs is transformed into the same, orient~tionally matched member of the glyph set. In other words, the values encoded in the individual glyphs are preserved. However, this 180~ rotation inverts the apparent ordering of linearly aligned glyphs. Fortunately, there is a 5 straightforward solution to this problem because the glyphs still are linearlyaligned, and the direction in which they are aligned is llnch~n~ed. In particular, if the glyph code pattern includes a linear pattern of sync glyphs, as at 61 in Fig. 10, that is encoded in accordance with an asymmetrical sync code sequence (e.g., a m~im~l length shift register sequence),testing for the sync 10 code sequence and its reverse resolves the ambiguity. On the other hand, 90~or 270~ rotations transform each of the two permitted glyphs into the other, thereby interchanging their apparent encoded values. Furthermore, the apparent directions of orthogonal lines (e.g., rows and columns, or 45~
diagonals) also are interchanged. And to further complicate the problem, the 15 apparent ordering of a linear pattern of sync code encoded glyphs may be normal (in the event of a 90~ rotation) or reversed (in the event of a 270~
rotation) with respect to the sync code sequence. Thus, to resolve this ambiguity through the use of a n-stage shift register m~im~l length sequence sync code, a test sequence of at least 2n is employed to avoid inadvertent match20 to a subsequence that is reversed and/or bit-inverted. Hence, it will be understood that four tests for the sync sequence, deciding bit inversion and sequence reversal, will resolve the ambiguity for four rotation states. [0~, 90~, 180~, 270~].
25 Mirror symmetry, which might occur in connection with the use of transparencies, is an additional ambiguity that might be encountered, so ~ 2~3~
additional encoding precautions may be taken to assure ~ qmhiguation For example, to resolve the combination of possible ambiguities, row and column sync codes suitably are selected to be distinct or to at least provide a different phase shift along the code as referenced from sync line intersections.
5 An alternate approach to ~ mbiguation would be to use an asymmetric spatial patterns for the sync code. This requires r~ mhiguation search in image space of glyphs rather than digital space of codes.
For some applications, sync code encoded glyph patterns may be used in 10 combination with coarser synchronization information, such as image reference marks or features, to support a dual phase, coarse/fine synchronization process) In decoding glyph arrays cont~ining glyphs encoding a predefined sync code 15 and variable data glyphs, the predefined sync code glyphs generally are decoded first. For example, in the case of a lattice-like sync code pattern, thedecoding can be initiated at any desired location. To locate nearby sync code encode glyphs possible sync subsequences of glyphs are tested in a local region that is large enough to definitely include a section of sync pattern. If there is 20 concern for orientation ambiguity, these possible sync subsequences are evaluated for each of the possible orientations until a satisfactory result is found. When an initial section of the known sync pattern is found, the sync pattern then is extended glyph-by-glyph. If the glyph code sequence is blocked by damage, alternate sync pattern paths may be followed. When required 2~ jumps to adjacent regions may be logically connected via digital sync code addressing properties. After the sync glyphs have been located as completely ~ 2 1 ~
as possible, variable data may read with logical order referenced and confirmed by local sync glyph code. Optionally the lattice of sync glyph locations may be used to interpolate the expected image positions of vaAable data glyphs.
6 Alternatively, the sync code may be used only after the data code has been transformed into a two dimensional array of decode values. In this event, the sync code is used in a feed forward mode to reorder and correct the decode values as may be required to compensate for decode errors, such as additions, dropouts and shearing of data.
SELF-CLOCKING GLYPH CODES
Field of the Invention This invention relates to self-clocking glyph codes for recording digital information on graphic recording media, such as plain paper. Even more specifically, this invention pertains to techniques for increasing the tolerance of such glyph codes to geometric distortion and physical damage, while preserving their homogeneous visual appearance and e~h~n~ing their suitability for applications requiring random access to the information they encode.
Background of the Invention In keeping with prior proposals, a basic self-clocking glyph code, such as shown in Fig. 1, typically is composed of an array 21 of elongated, slash-like symbols or "glyphs" 22 and 23 that are written with their longitudinal axes tilted at angles of a~ .lately -45~ and +45~ with respect to the vertical axis 24 of a recording medium 25 to encode bit values of "1" and "O", respectively, or vice-versa, in each of the glyphs. These codes are "self-clocking" because they include an optically detectable symbol (a "glyph" for each of the values they encode. This means that the detection of the glyphs implicitly synchronizes the coding process to the code once the decoding process has been properly spatially oriented with respect to the code. As will be evident, this implicit synchronization is valid so long as all of the glyphs are detected or otherwise accounted for in correct logical order. If, 21 344q4 however, synchronization is lost, there is nothing to restore the implicit synchronization, so a loss of synchroni7~tion generally causes an implicitly synchronized decoding process to experience fatal error. For a more detailed discussion of these glyph codes and of techniques for decoding them, reference may be made to the following commonly ~ igned U.S. patent docllments:
Bloomberg U.S. Patent No. 5,168,147 which issued December 1, 1992 on "Binary Image Processing for Decoding Self-Clocking Glyph Shape Codes"; Bloomberg et al. U.S. Patent No. 5,091,966, which issued February 25, 1992 on "Adaptive Scaling for Decoding Spatially Periodic SeLf-Clocking Glyph Shape Codes"; and Stearns et al, U.S. Patent No. 5,128,525, which issued July 7, 1992 on "Convolution Filt~ring for Decoding Self-Clocking Glyph Shape Codes".
As will be seen, these prior proposals not only describe the use of two discrimin~hle glyph shapes for the encoding of single bit binary values, but also teach that "glyph shape encoding" is extensible to the encoding of digital values of any given bit length, n, in each glyph by lltili7.ing a code having 2n dis~rimin~hl~ glyph shapes (where the discriminS~hility between glyphs representing different values is based upon the distinctive rotational orient,~ti--n.c~ and/or the distinctive geometric configurations of the different glyphs). Thus, even though a str~ightforward self-clocking glyph code of the type shown in Fig. 1 is featured in this disclosure to simplify the description, it will be evident that the broader aspects of this invention are applicable to other glyph codes, including more complex ones.In practice, each of the glyphs 22 and ~A
~ 21 34494 23 usually is defined by writing a recognizable pattern of "on" and "ofP' pixelsinto a two dimensional, array of pixel positions (i.e., a "symbol cell"). As a general rule, the pixels that form the body of the glyph (say, the "on" pixels) are written into contiguous pixel ~., 213~9~
positions that are more or less centered within the symbol cell, and the other or "off" pixels are written into the reln~ining pixel positions of the symbol cell to provide a contrasting surround that demarks the edges of the glyph and separates it from its neighbors.
6 The symbol cells, in turn, ordinarily are tiled onto the recording medium in accordance with a preselected spatial formatting rule, so the logical order of the data values that the glyphs encode is preserved by the spatial order in which the glyphs are mapped onto the recording medium. For example, the symbol cells may be written on the recording medium in accordance with a 10 regular and repeating spatial formatting rule that is selected to map the glyph encodings into a two dimensional, rectangular array of logical data blocks of predetermined size, such as data blocks having a 16 symbol cell x 16 symbol cell format. These data blocks suitably are organized on the recording medium in left-to-right, top-to-bottom logical order.
The size of the symbol cells that are used for the glyphs depends on a variety of factors, including the spatial resolution of the printing process that is employed to write the glyphs, the type and extent of the degeneration that the printed glyph code is required to tolerate, and the spatial resolution of the 20 lowest resolution sc~nning process that is expected to be able to recover thecode. In view of these constraints, a 300 s.p.i. printer suitably centers the glyphs in 5 pixel x 5 pixel or 7 pixel x 7 pixel symbol cells. Even larger symbol cells can be employed for the glyphs if the increased granularity of the textured appearance of the printed glyph code is tolerable or unavoidable. As a 25 general rule, of course, the smallest practical symbol cell size is esthetically most pleasing because the visual texturing appearance of the glyph code 213~4~
gradually blends as the symbol cell size is reduced. Indeed, codes composed of smaller symbol cells tend to have generally uniform gray scale appearances when they are viewed under ordinary lighting at normal viewing dist~nces.
The existing techniques for decoding self clocking glyph codes are designe~ to 5 be initi~ e-l at or near the center of a reference glyph that occupies a knownspatial position relative to the remainder of the glyph code (for example, the glyph in the upper, lefthand corner of a rectangular array of glyphs). Thus, accurately locating this reference glyph clearly is a key to spatially synchronizing such a decoding process with the glyph code.
While synchronous initialization of the decoding process is a necessary condition for orderly decoding of a self-clocking glyph code, it may not be a sufficient condition to ensure accurate decoding of the code because the scanned-in image of the glyph code pattern often is distorted by skew and/or 15 scaling errors. For this reason, prior decoding processes generally have attempted to determine the relative spatial positions of three or more referenceglyphs in recognizeable positions that have a known nominal, non-colinear relationship with respect to each other (such as the corner glyphs of a rectangular array of glyphs) with sufficient precision to compute skew and 20 scaling correction factors. These correction factors then are used to adjust the angle and magnitude of a vector that dictates the direction and magnitude of the jumps the decoding process makes while it is advancing from glyph-to-glyph.
25 As will be appreciated, anything that prevents the decoding process from establiching and maint~ining proper spatial synchronization with the glyph 2~3~43~
code threatens to defeat the decoding. Indeed, as a general rule, any glyphs that are isolated from synchronously readable glyphs may be unsynchronizable, especially in the interior of a glyph code pattern. Spatial synchronization is, of course, required to enable the decoding process to preserve the logical ordering of the data that is encoded in the glyphs. Minor 5 synchronization errors affecting a limited number of glyphs may be tolerable if the code has sufficient error correcting capacity (i. e., error correction code protection) to correct for the resulting decode errors, but more extended synchronization errors generally result in incomplete reading of the encoded information. Ultimately, this performance degradation can become 10 sufficiently severe to be unacceptable.
Accordingly, it will be evident that one of the more significant disadvantages of spatially synchronizing a glyph decoding process through the use of a global imaginal reference that is computed from the spatial locations of certain key 15 glyphs, such as the corner glyphs of a rectangular glyph pattern, is that various factors may prevent the spatial location of one or more of those key glyphs from being determined with sufficient precision to compute a valid reference, thereby preventing the decoding process from achieving and/or maint~ining spatial synchronism with the glyph code. Moreover, even when a 20 valid glo~al reference can be computed, there is an appreciable risk that thedecoding process will lose synchronization. For example, the scanned-in image of the glyph code might be sufficiently degraded to cause the decoding process to erroneously branch from one row of glyphs to an adjacent row, or to skip overa glyph without accounting for it, or to miscount the number of unreadable 25 glyphs in a damaged region of the image. Unfortunately, none of the known self-clocking glyph codes or related decoding processes enable the decoding 2~3~49~
process to re-establish proper synchronization once it has been lost, so a loss of synchronization usually causes these existing decoding processes to degenerate into an irreversible failure mode.
The problems that have been encountered with establi~hing and m~int,~inin~
5 proper spatial synchronization during the decoding of self-clocking glyph codes have been compounded by the lack of reliable techniques for detecting asynchronous operation of the decoder. Glyph centers have been counted during the decoding of these codes on the theory that a miscount over any predefined segment of the formatted code will signal a loss of synchroni~qtion.
10 This is a weak synchronization validation process if part of the glyph code pattern (e. g., a portion of its periphery) is missing or otherwise undetectable.
there may be an ambiguity with respect to the number of glyphs that are present in the image and with respect to what part of the glyph pattern is present in the image. Thus, the decoding process may provide invalid results, 15 without providing any warning that the results are invalid.
Image rotation of 90~, 180~, 270~ and reflection result in rotational and mirrorimage transformations of the glyph code pattern, and these transformations can alter the apparent spatial order of the glyphs and/or the apparent 20 rotational orientation of the individual glyphs. Indeed, this is an especially significant issue when the glyphs encode the data values in their rotational orientations, such as when the glyphs are slash-like symbols that are oriented at +45~ with respect to the longitudinal axis of the recording medium to encode high ("1") and low (~0") logic level bits, respectively, because such 25 glyphs often are written in patterns that have rotational symmetry.
Unfortunately, the orientation of scanned-in glyph code patterns frequently ~ 213~
involves some ambiguity. For example, there typically is a four-fold rotational ambiguity in the case of a glyph code that is composed of a square array of slash-like glyphs of the above-described type that are written into square symbol cells7 and a two-fold ambiguity in the case of a glyph code that is composed of a rectangular array of such glyphs.
Some of the commercially available two dimensional data codes, such as VericodeTM by Veritex and DatacodeTM by Data Code International, have distinctive reference marks along one or more of their borders that can be used for scaling and for explicitly spatially synchronizing the decoding of those 10 codes However, these reference marks are visually distinctive so they tend to increase the visual obtrusiveness of those codes. One of the principal advantages of self-clocking glyph codes is that they tend to be esthetically pleasing because of their non-obtrusive visual appearance, so it would be counterproductive to supplement them with reference patterns that have such 15 a strong visual impact.
Summary of the Invention In accordance with the present invention, a self-clocking glyph code is 20 composed of data glyphs which encode logically ordered data values, together with one or more multi-glyph synchronization code patterns. Each of the synchronization code patterns is distinguishable from the data glyphs, but the synchronization code pattern or patterns provide an explicit spatial reference for the data glyphs. To this end, the glyphs of the synchronization code 25 pattern or patterns are written on the recording medium in accordance with the same spatial formatting rule as the data glyphs, plus at least one 21 34~94 Additional form~t+ing rule that establishes a predetermined geometric relationship between the glyphs of the synchronization code pattern or patterns and corresponding ones of the spatially formatted data glyphs, such that the logical order of at least a .qignifi~nt. number (~ 1) of the data glyphs is inferable from the synchronization code pattern or patterns. Preferably, each of the synchronization code patterns is composed of glyphs that are selected from the same finite symbol set as the data glyphs, so the synchronization code pattern or patterns typically are defined by glyph patterns that have a suitably high statistical probability of being unique with respect to any pattern of data glyphs.
Other aspects of this invention are as follows:
A process for spatially referencing a self-clocking glyph code that is composed of logically ordered data glyphs that are selected from a predetermined symbol set for encoding information in said data glyphs; said process comprising writing said logic~lly ordered data glyphs in a two dimen~ l field on a recor~ing medium in accordance with a predetermined spatial formatting rule;
writing a syncl~rolli~ation code in said field, said syncl~rol~i;~tion code being composed of logically ordered glyphs that are distinguishable from said data glyphs, said synchrol~i;~tion code being written in said field in accordance ~ 21 34494 with said spatial formatting rule plus at least one additional rule that establishes a predetermined geometric relationship among the glyphs of said synchronization code and said data glyphs, whereby the logical ordering of at least some of the data glyphs in said field is inferable from the glyphs of saidsynchronization code based upon said geometric relationship.
An optically readable record for storing encoded information; said record comprising a recording medium;
a self-clocking data code for encoding said information; said data code being composed of data glyphs that are written on said recording medium on centers that are spatially distributed in nominal accordance with a two dimensional spatial formatting rule; said data glyphs being ~lçfine-l by respective symbols that are selected from a finite set of optically discrimin~hle symbols to encode said information; and at least one self-clocking synchronization code for establi~hing a persistent spatial reference for said data code; said synchronization code beingdiscrimin~ble from said data code and being composed of additional glyphs that are written on said recording medium in nominal accordance with said spatial formatting rule plus at least one additional rule that establishes a predetermined geometric relationship among the glyphs of the synchronization code and said data glyphs, whereby the logical ordering of at 9a r~
, ~
lesst some of the data glyphs is inferable from their geometric relationship to the glyphs of the synchronization code.
An optically readable record for storing encoded information; said record comprising a recording medium;
a self-clocking glyph code for storing said information on said recording medium; said glyph code being composed of a data code and at least one a synchronization pattern for providing a persistent spatial reference for said data code said data code being composed of a first logically ordered set of discrete data glyphs that are written on said recording medium on spatially offset centers in nominal accordance with a two dimensional spatial formatting rule;
said data glyphs being defined by respective symbols that are selected from a finite set of optically discrimin~ble symbols to encode said information; and said synchronization pattern being discrimin~hle from said data code and being composed of a second logically ordered set of glyphs that are written on said recording medium in nominal accordance with said spatial formS~tting rule plus at least one additional rule that establishes a predetermined geometric relationship among the glyphs of the synchronization pattern and said data glyphs, whereby the logical ordering of at least some of the data 9b ~ 2 1 3~494 glyphs is inferable from their geometric relationship to the glyphs of the synchronization pattern.
An optically readable record for storing encoded information; said record comprlsmg a recording medium;
a self-clocking data code for encoding said information; said data code being composed of data glyphs that are written on said recording medium on centers that are spatially distributed in nominal accordance with a two dimensional spatial formatting rule; said data glyphs being defined by respective symbols that are selected from a finite set of optically discrimin~hle symbols to encode said information; and a synchronization code pattern composed of at least two intersecting nonparallel synchronization code subpatterns for establi~hing a two dimensional, persistent spatial reference for said data code, each of said synchronization patterns being discrimin~hle from said data code and being composed of glyphs that are written on said recording medium in accordance with said spatial formatting rule plus at least one additional rule that establishes a predetermined geometric relationship between the glyphs of the said synchronization patterns and said data glyphs, whereby the logical ordering of at least some of the data glyphs is inferable from their geometric relationship to the glyphs of the synchronization patterns.
gc ~' 2~ 34494 Brief Description of the Drawings Still further objects and advantages of this invention will become apparent when the following detailed description is read in conjunction with the attached drawings, in which:
Fig. 1 is a simplified, enlarged illustration of a standard self-clocking glyph code that is written in accordance with a basic two dimensional block formatting rule;
Fig. 2 is a simplified, enlarged illustration of a block formatted self-clockingglyph code that includes synchronization code patterns in accordance with this invention;
~A
~ 2~3~4~
Fig. 3 is a simplified, enlarged illustration of a block formatted self-clockingglyph code that contains non-colinear synchronization code patterns in accordance with a more detailed aspect of this invention;
5 Fig. 4 is a simplified, enlarged illustration of a glyph code pattern that has a lattice-like framework of two dimensional sync code sequences in accordance with an even more detailed aspect of this invention;
Fig. 5 is a simplified, enlarged illustration of a glyph code pattern that has a10 lattice-like framework of interlocking, two dimensional sync codes;
Figs. 6 and 7 are full size illustrations of the sync code and data code components, respectively, of a glyph code pattern of the type shown in Figs. 4 and 5;
Fig. 8 is a full size illustration of a composite glyph code pattern that is composed of the sync code and data code components that are shown in Figs. 6 and 7, respectively;
20 Fig. 9 is a simplified, enlarged illustration of a multiple logical data block formatted glyph code pattern that has a lattice-like framework of two dimensional sync codes and internal block labels for extended addressability;
Fig. 10 diagrammatically illustrates the use of asymmetric sync code 25 sequences for symmetry disambiguation;
~ ~ 2~34~
Fig. 11 illustrates the use of a lattice-like sync code pattern for navigating around damage in glyph code patterns; and Fig. 12 illustrates the use of a lattice-like sync code pattern for explicitly 5 synchronizing a distorted glyph code pattern.
Detailed Description of the Illustrated Embodiments While the invention is described in some detail hereinbelow with reference to 10 certain illustrated embodiments, it is to be understood that there is no intent to limit it to those embodiments. On the contrary, the intent is to cover all modifications, alternatives and equivalents ~alling within the spirit and scope of the invention as defined by the appended claims.
15 In accordance with this invention, as shown in Fig. 2, a self clocking glyph code 31 includes a data code 32, together with a two dimensional, distributed sync code pattern 33. For illustrative clarity, the data code 32 is shown as being composed of slash-like symbols or "glyphs" 34 that are tilted at -45~ with respect to the longitudinal axis of the recording medium 35 (see the process 20 direction arrow 36), while the sync code pattern 33 is shown as being composed of glyphs 37 that are tilted at +45~ with respect to that axis. In practice, of course the angular orientation of the data glyphs 34 is modulated in accordance with the respective data values they encode. Further, the angular orientation of the sync code glyphs 37 might also be modulated in accordance 25 with a cyclic or acyclic sequence of values, as more fully described hereinbelow.
Consequently, there may not be any symbolic distinction between a given data 213~94 glyph 34 and a given sync glyph 37. Indeed, in keeping with one of the more detailed features of this invention, the data glyphs 34 and the sync glyphs 37 preferably are potentially identical members of the same finite symbol set because this causes the sync code pattern 33 to visually blend with the data code 32 (see Fig. 8) effectively. Moreover, to even further enhance the visual 5 homogeneity of the composite code 31, the sync code pattern 33 advantageously is selected to avoid visually distinctive glyph patterns, such aslong runs of the same symbol.
More particularly, in keeping with this invention, the sync code encoded glyph 10 pattern 33 (sometimes referred to herein as the "sync code pattern") is composed of a predefined, cyclical or acyclical pattern of glyphs that is sufficiently distinctive to be disambiguated from the data code 32 with acceptably high statistical reliability. This sync code pattern 33 is written onthe recording medium 35 in accordance with the same spatial formatting rule 15 as the data code 32, but the formatting of the sync code pattern 33 is further governed by at least one additional rule that establishes a predetermined localized geometric relationship between the spatial positioning of the glyphs 37 of the sync code pattern 33 and the spatial positioning of nearby data glyphs34. The logical ordering of the sync code values that are encoded by the glyphs 20 of the sync code pattern 33 is predetermined. Thus, the sync code pattern 33 defines localized spatial references for determining the logical ordering of thedata glyphs 34.
As will be appreciated, the replicability of the sync code pattern 33 ensures 25 that the spatial references it provides survive replication of the composite code 31 (such as when the composite code 31 is replicated by photocopying or 213~94 .
farsimile tr~n~mi~ion to provide a hardcopy replica or by bitmap input sc~nnin~ to provide an electronic representation) subject only to the losses that might occur if the original image and/or the copy of their composite code 31 is ~l~m~ged. Advantageously, the sync code pattern 33 is distributed throughout the composite code 31 and is selected so that the density of the spatial 5 references it provides is sufficient to locally reference all of the data glyphs 34.
In that event, any non-linear distortions the data code 32 might experience while being replicated are tracked by corresponding localized non-linear distortions of the sync code pattern 33.
10 In theory, the sync code pattern 33 may take many different forms. In practice, however, it advantageously is written as a linear pattern (or, for a nonlinear implementation, in areas of adjacent glyph cells) because these type~ of subpatterns are easier to identify than nonlinear patterns of spatially scattered glyphs. For example, the sync code pattern 33 may be composed of 15 linear subpatterns of glyphs 37 that are written on the same center-to-centerspacing as the data glyphs 34 to span the spatially formatted glyph code 31 (or logical subsets thereof- see Fig. 4) in one direction (see Fig. 2) or in a plurality of different directions (see Fig. 3 et seq.).
20 Focusing on linear sync code glyph patterns (i. e., patterns of sync code encode glyphs that are written in linear spatial sequences), it will be understood thatsuch a sync code pattern may be defined somewhat more generally by a two dimensional array of linear sync code subpatterns, where each of the subpatterns is composed of every mth glyph (where m is an integer) along a 25 substantially straight line that runs parallel to a predetermined synchronization direction. The degenerative case, where m = 1, provides fine ~ 213~94 grain explicit synchronization because the sync code pattern is composed of adjacent glyphs. On occasion, however, a coarser explicit synchronization may suffice because the implicit synchronization that is provided by the self-clocking properties of the data glyphs 34 is valid whenever the presence of the data glyphs is detectable reliably, regardless or whether the detected data 5 glyphs can be accurately decoded or not. In any event, when linear sync code subpatterns of a given granularity, m, are employed, each set of glyphs that is (1) aligned in a synchronization direction and (2) composed of glyphs that are separated from each other by m - 1 intermediate glyphs, implicitly includes either all sync glyphs in proper sequence or no sync glyphs, except where 10 desynchronizing faults (i. e., synchronization errors) are encountered.
Sync code patterns composed of parallel, longitudinally separated, linear subpatterns, such as the pattern 33 in Fig. 2, are an interesting example of how the inclusion of a pattern of sync code encoded glyphs facilitates 15 establi~hing and maint~ining synchronization during the decoding of a self-clocking code 32. In this embodiment, the sync code subpatterns 33a -33d are written in parallel alignment with a major axes of the data code 32 (i. e., its lateral or x -axis in a standard orthogonal coordinate space), and these subpatterns 33a -33d are composed of sync code encoded glyphs 37 that are 20 written on the same center-to-center spacing along the x-axis as the data glyphs 34. Further, the sync code subpatterns 33a -33d each is of sufficient length to span the data code 32 along its x-axis. And, these sync code subpatterns 33a -33d are spatially displaced from each other along the orthogonal axis of the composite code pattern 31 (i. e., its longitudinal axis or 25 y-axis) by a predetermined constant number of lines of data code 32.
- ~ 213~4~
Given the x, y coordinates on which the glyphs 37 of any two neighboring sync code subpatterns (such as the subpatterns 33a and 33b) are centered, prospective center locations for the data glyphs 34 that reside therebetween can be identified by pairing the glyphs 37 that occupy logically corresponding 6 positions in the two sync code subpatterns with each other and by then linearly interpolating between the x, y coordinates of these logically paired glyphs.
This process provides strong explicit synchronization axially of the sync code subpatterns 33a - 33d because it is sensitive to nonline~r distortion along thataxis. Weaker synchronization is provided orthogonally to the subpatterns 33a 10 - 33d because orthogonal positioning is determined by interpolation from the spacing of the neighboring sync code subpatterns 33a - 33d..
Fortunately, the implicit synchronization that is provided by the self-clocking property of the data glyphs 34 often is sufficient for maint~ining the 15 synchronization of the decode process while the data glyphs 34 that are disposed between the parallel sync code subpatterns 33a-33d are being read.
For example, the data glyphs 34 that are logically aligned between sync code glyphs 37 that occupy logically corresponding positions in two neighboring sync code subpatterns, suitably are identified on a "best efforts basis," (1) by20 explicitly resynchronizing the decode process at one of the logically paired sync code glyphs 37, and (2) by then advancing the decode process, data glyph-by-data glyph, toward the second of the paired sync code glyphs 37. This process isrepeated for one after another of the logical pairings of the sync code glyphs 37.
It, of course, will be understood that this is a "best efforts" process because 26 local damage can frustrate the attempt to identify some of the data glyphs 34with the precision that is required to reliably determine their logical ordering.
~ 2~3~
However, synchronization losses are easily detected. In that connection, it willbe noted that the number of data glyphs 34 that are read while the decode process is searching for data glyphs 34 that are logically aligned between any two logically paired sync code glyphs 37 is a reliable intlir<qtion of whether the anticipated number of data glyphs 34 was found or not. Moreover, the success 5 or failure of the decode process in tracking from one to the other of the logically paired sync code glyphs 37 is a strong indication of whether synchronization was preserved during the reading of the intervening data glyphs 34. Thus, if synchronization errors occur, suitable remedial and/or corrective action can be taken. This action may be transparent to the user, such as when there is 10 sufficient error correction capacity to correct the errors or may involve more drastic measures, such as notifying the user that the decode results are invalidand/or incomplete.
As shown in Fig. 3, the computational uncertainty of relying on data glyphs 34 15 as implicit spatial synchronization references can be avoided by employing a sync code pattern 43 that is composed of a plurality of nonparallel linear sync code subpatterns 43a - 43b that jointly span the composite code pattern 31( (e.g., along its x-axis and its y-axis). Two such subpatterns 43a and 43b are sufficient for reducing the synchronization errors that may tend to occur; for 20 example, as a result of one or two dimensional distortion of the image of thecomposite code pattern 31. Moreover, if the sync codes that are employed provide unique addresses over some definite span of each of the sync code subpatterns, such as is provided by the some of the digital codes that are discussed hereinbelow, a single pair of nonparallel sync lines can further 2~ reduce synchronization errors caused by dropouts of individual data glyphs 34, groups of neighboring data glyphs, and/or full lines of data glyphs.
213~4~4 .
Linearly extrapolating from the x and y coordinates of the glyphs 37 of the sync code subpatterns 43a and 43b in directions parallel to the y-axis and x-axis, respectively, of the composite code pattern 31 yields lines that intersect at prospective center locations for the data glyphs 34, thereby providing explicit 5 spatial synchronization references for the data glyphs 34. In practice, the scanned in image of the composite code pattern 31 may be skewed somewhat, but any slight skew error may be elimin.qted if the angular orientations of the sync code subpatterns 43a and 43b relative to the x and y axes of the composite code pattern 31 is known. Specifically, that information can be used to compute 10 a global or local skew correction factor that brings the sync code strings 43a and 43b into the proper angular orientation with the x and y axes of the composite code pattern 31. While any two nonparallel sync code strings that jointly span the composite code pattern 31 along its major axes will suffice, two perpendicular sync code strings 43a and 43b that span the composite code 15 pattern 31 along its x-axis and its y-axis, respectively, are shown to provide a straightforward example.
Still further, as shown in Figs 4 - 12, two non-colinear sets of parallel linearsync code subpatterns can be used to form a lattice-like sync code pattern 53 20 that is composed of a two dimensional tiled array of sync code cells 54a - 54i that have rectangular or parallelogram-like configurations. The x, y coordinates on which the glyphs 37 of such a sync code pattern are centered may be identified, and the glyphs 37 may be decoded, systematically throughout the composite code pattern 31 by following the sync code lattice.
25 Since there are four paths to most vertices of this lattice, almost any givenglyph 37 of the sync code pattern 53 can be reached by following a path of ~ 2134~9~
logically ordered sync code glyphs ~this notion of imposing a predetermined logical order on the values of neighboring glyphs in the sync code pattern 53 isdiscussed in more detail hereinbelow), even when damage frustrates any attempt to determine the logical ordering of the glyphs 37 along one or more of the alternative paths. For example, see Fig.8 where the overwrites 81,82, and 6 83 partially obliterate portions of the lattice-like sync code pattern 53. The closed line geometry of the overwrite 81 may prevent the sync glyphs within its interior from being reached, but it is possible to navigate around the open line overwrites 82 and 83 on sync code glyphs.
As will be appreciated, the use a two dimensional sync code lattice, such as thesync code pattern 53, permits positive verification of the number of data glyphs34 that are read in a linear array of glyphs, such as a row or column of data glyphs 34 that has both ends terminated by sync code glyphs 37. This, in turn, means that such a sync code pattern permits substantial verification that branching to an adjacent row of glyphs has not occurred during the reading of the data glyphs 34 that are fully confined by the sync code lattice.
Advantageously, the dimensions of the lattice cells 54a - 54i (as measured by sync code glyph count) are uniform because this permits the periodicity of the cyclical synchronization codes that are discussed below to be matched to the periodicity of the sync code pattern 53. While Fig 4 shows that each cell 54a -54i of the sync code pattern 53 contains a 6 x 6 array of data glyphs 34, it will be understood that any M x N lattice may be employed in practice. For example, each of the cells 54a -54i of the lattice pattern 53 suitably frames a logical data block composed of a 16 x 16 array of data glyphs 34, thereby reducing the overhead cost of including the sync code pattern 53 to provide the ~ 213~4 spatial references that are required to explicitly synchronize the decoding of the data glyphs 34.
DIGITAL CODES FOR SYNC GLYPH CODES.
5 As is known, digital codes of various descriptions have been extensively utilized to synchronize communication ~ign~l~. This usually is a one dimensional synchronization problem. The present invention, on the other hand, deals with a two dimensional synchronization problem. As will be seen, however, it has been found that very effective two dimensional 10 synchronization code patterns can be implemented by using combinations of one dimensional digital codes. Moreover, it has been further found that some of these two dimensional sync code patterns have attractive attributes for more detailed applications, including absolute and relative logical addressing of thestructure and/or information content of self-clocking glyph codes, orientational15 disambiguation of the information content of such codes, and enh~nced tolerance to damage and/or distortion.
There clearly is a fundamental advantage to associating digital sync codes with glyph patterns that are spatially well ordered and relatively easily 20 identified. The above-described linear glyph code patterns are good examples of patterns that satisfy these general criteria .
Indeed, as will be recalled, even parallel line sync code patterns (see Fig. 2) can provide synchronization information with respect to two dimensions: i. e., 25 along and transverse to the lines. Furthermore, it will be evident that the positive confirmation that a sequence of glyphs on a line encode a predefined 213~4 digital sync code makes the line a registration mark with respect to transverse position; e.g. the edge of a glyph subregion. Simple repetitive digital codes such as 11111111.. or 10101010... can be used for this purpose. Theyclearly are simple to detect, but they provide limited addressing information along the line and minirn~l absolute addressing information. Moreover, the high symmetry of these repetitive codes provides little aid in resolving orientation ambiguity (this subject is discussed in more detail hereinbelow).
It also will be understood that the intersections of sync code patterns that arecomposed of nonparallel linear strings of sync code (Figs. 3 - ~ can serve as 10 references for synchronization, even if the digital codes themselves lack unique addressing, such as in the simple code sequences mentioned hereinabove. For example two sets of intersecting linear subpatterns of glyphs that encode sync codes, as in Fig. 4, can provide some two dimensional absolute registration capability. Speçific~lly, the confirmation that each of these linear 15 glyph strings encode sync code establishes linear boundaries for a glyph region or cell that is framed by sync code glyphs, and that framed region is known (because the spatial formatting rule is known) to have a predefined pattern of interior glyph positions.
20 Cyclic digital code sequences that conform to so-called m~im~l length sequences (MLS) can be generated by appropriately configured digital shift registers or by reference to a bit-level description of the desired MLS (see, Shift Register Sequences by Solomon W. Golomb, Holden-Day, Inc. 1967). It now has been found that these MLS digital sequences have properties that make them 25 very useful for more sophisticated proce~qing of two dimensional glyph codes.
213~4~
More particularly, it is known that an n-stage shift register that produces a MLS will generate a binary sequence of length L=2n-1, in which every subsequence of length n is unique. Thus any subsequence of at least length 1 =n, llefine.s position along the sequence and, therefore, a local 5 synchronization reference and a local address with a periodic ambiguity of L.
Continuation of such a shift register sequence is cyclically periodic with period L. For example, if n = 3, then L = 7 and a corresponding MLS shift register bit sequence is S= 00111010011101.... Thus, it will be evident that n may be easily chosen, if desired, to be large enough so that there is no repeat of the bit 10 sequence S within a bit string of any given finite length. More generally, itwill be observed that when a spatially ordered array of glyphs encode the values of one after another of the bits of a non-repeating shift register subsequence in logical order, the bit values that are encoded by the glyphs provide absolute one dimensional addressing transversely of the array of 15 glyphs. As will be appreciated, this absolute addressing property follows from the mapping of the logically ordered, non-repeating shift register subsequence on to the spatially ordered pattern of glyphs, so it is not limited to linear patterns of sync code glyphs.
20 As a general rule, MLS sequences are not symmetric for n>2. Although an address d~fining MLS subsequence of length n may be symmetric in the forward and reverse directions (e.g., 101), extending the sequence to no more than length 2n will break the symmetry (e. g., 101001). As will be seen, this property of MLS codes can aid in glyph pattern symmetry r~ nbiguation.
213~494 .
As previously pointed out, an n-stage shift register m~xim~l length sequence has a cyclic period of bit length L=2n-1,which is exactly one less than a "round" binary number. Thus, for n= 1, 2,3, 4, 5,6,7, ...; L= 1,3,7,15, 51,63, 127,.... These bit lengths do not naturally match the round binary units in 5 which digital data is typically assembled; e.g., nibbles(4), bytes(8), blocks(512x8). This is me~nin~ful because it implies that MLS codes have the drawback of not precisely sp~nning data blocks that are arrayed in convenient binary units.
10 Fortunately, however, it has been shown that for an n- stage shift register with a m~qxim~ql length sequence of 2n- 1 that every cycle period between 1 and 2n can be obtained from it by one of 2n- 1 simple st~nfl~qrd modifications to a nonlinear shift register [Golomb, op. ct. p. 192 - 194]. In many cases, the resulting sequence is a subsequence of a m~ximal length sequence. For 15 example:
~xim~l n=4, L=15, S=111000100110101111000....
Truncated L' =8, S= 11100010.111000....
M~xim~l n=5, L=31, S=0000100101100111110001101110101.
TruncatedL'=17, S=00001001011001111.00001... (Note, each 5 bit subsequence is unique) In general, these cyclic codes, including the cyclic truncated sequences, provide 25 unique n bit subsequence addressing within each cycle (i. e., within each local region of the printed glyph code).
21~9~
An n-stage shift register with a m~im~l length sequence of L=2n- 1 generally contains a subsequence of (n - 1) O's adjacent to (n) 1's. Thus, such a sequence can be padded by inserting one or more (i. e., "r") additional O's adjacent to the (n - 1) O's. Adding one 0, yields a sequence of period length 5 (L' =2n), a round binary value. Thus, padding of this type can be used in a glyph code to align each cycle of sync code with a block of data code that is written in a convenient binary format. Inserting two or more O's enables alignment with (1) a data block having an ~'even" binary format, together with (2) one or more regions of additional glyphs which may be used for separate 10 purposes, such as transverse lines of sync code glyphs. As will be seen, thiszero padding has the useful property of preserving the uniqueness of all n-bit subsequences, except for the string of O's that is immediately followed by and unambiguously identified by the (n) 1's, as unambiguous addresses along the code.
Examples:
n=2, r=0, L=3, S=011011 (Cautions: symmetric) n=2,r=l,L'=4,S=00110011 (Cautions: bit inversion= reverse of sequence; symmetric) n=2, r=2, L' =5, S =0001100011.
n=3, r=0, L' =7, S =00111010011101 n=3,r=1,L'=8,S=0001110100011101 [Caution: bit inversion =sequence reverse]
n=3,r=2, L'=9, S=000011101000011101.
2~3~
.
n=4, r=0, L' = 16, S=000111101011001000111101011001 n=4, r= 1, L' = 16, S=00001111010110010000111101011001 n=4, r=2, L' = 17, S=0000011110101100100000111101011001 5 It also has been recognized that a periodic sync code that provides unique addressing within a subsequence of predetermined length (such as an MLS
code) can be employed to provide directional steering to the glyph location for an intersecting transverse pattern of sync glyphs. Thus, when a periodic sync code of the foregoing type is employed, it is useful to match the spacing of 10 transverse lines of sync glyphs to the period of the periodic sync code so that there is consistent transverse referencing between the periodic sync code and the transverse sync glyphs.
The maintenance of consistent transverse references between intersecting 15 strings of sync code is, of course, extensible to some of the more interesting glyph code patterns, including the lattice-like glyph code patterns shown in Figs. 4 - 8 . More particularly, it will be evident that two intersecting sets of parallel glyph code subpatterns that encode periodic sync codes may be spaced at periodic intervals fitting both sync codes. Moreover, it can be shown that a 20 spacing that matches the periodic intervals of both codes yields an interlocking periodic grid of sync with the same bit patterns in the neighborhood of each intersection. Indeed, this is illustrated in Fig. 5 for the code 0011101 in a square lattice. And, it also applies to Figs. 6 - 8 where the sync code pattern is a square lattice of glyph strings that encode the 17 bit cycle code with two zero 25 bit ps3~1rling discussed above. Figs. 6 and 7 show the sync lattice and the data glyphs printed separately, while Fig. 8 illustrates the combined or composite 213~94 glyph code which shows that the sync code visually blends uniformly with the variable data encoded glyphs.
On a somewhat more general level, if intersecting strings glyphs encode linear address information and share common glyphs, the values encoded in the 6 shared glyphs preferably are consistent with both of the encoded sequences.
On the other hand where the sync codes do not otherwise provide appreciable unique addressing, as in simple codes(ll1 or 1010), it may be advantageous to arrange the intersection glyphs to be discordant with one or both of the sync sequences to facilitate the identification of the intersection location; e.g., 1111101111111, or 101010001010.
As shown in Fig. 9, the translational ambiguity of a repeating periodic s.ync sequence can be overcome by writing additional predefined data in "reserved"
glyphs 58a-58i which have a known order and which reside at predefined 15 positions with respect to the encoded sync code glyphs. For robustness, the glyphs that are reserved for encoding this additional information preferably are spatially adjacent to the sync code glyphs 37. The additional information that might be encoded in these reserved glyphs includes physical and/or logical data block identification. For example, the cyclic sync codes that are 20 contemplated by this invention can be supplemented by encoding appropriate data block specific information in these reserved glyphs 58a - 68i to enable thesync code to be used for synchronizing the of data glyphs, on a data block basisand the reserved glyphs 58a - 58i to be employed for identifying the data blocksphysically with respect to the overall glyph code pattern. To this end, in Fig. 9 25 there are nine data blocks 54a - 54i which are surrounded by a cyclic MLS code [n=4, L' = 16 (padded with a single zero)] and labeled with a byte of data in --2~--213~4~
.
the upper left corner giving block coordinates in the form of two hex code numbers; e.g. 00100011 corresponds to 23, which is a suitable identifier for the data block in the second column and third row of the tiled format.
Physical addressing of the data blocks 54a - 54i provides a powerful in-le~in~
5 mec~ni~m for random access to the data recorded in any of the data blocks 54a - 54i. Furthermore, if desired, this inde2~ing may be e~tenfle~
hierarchically to include, for example, a page index value, a volume index value and similar logical correlators in the data that is written into the reserved glyphs 58a - 58i. Standard random access techniques can then be 10 employed for advancing the glyph reading process from one data block to another data block within the same glyph code pattern or a different glyph code pattern. As will be appreciated, the ability to index reliably and efficiently from one glyph code pattern to another is dependent on having a substantially stable and/or easily reconstructable relationship between those 15 patterns, so that the indexing can be performed using straight forward mathematical functions (i. e., addition and subtraction). Image distortions and other factors may perturb the physical relationship among well-linked glyph code patterns slightly, so it sometimes may be best to perform random access seeks using a convergence strategy; whereby each seek starts with a coarse 20 indexing to at least roughly approach the target location, and then continueswith one or more progressively finer seeks until the target is located precisely.
As will be appreciated, the indexing data in the reserved glyphs 58a-58i can be used to control the coarse seeks, while the finer seeks can be controlled using the precision provided by the above-described addressing codes.
213~g4 .
Considering the issues involved in f~i~Ambiguating the sync code glyphs 37 from the data glyphs 34 in some additional detail, the general rule is that the ability to distinguish a predefined ordered subset of k glyphs that encode sync code from glyphs that encode variable data with a r2qn-1om occurrence of bits is5 governed by the probability of random match of all bits, 1/2k, which decreasesexponentially (geometrically) with k. Even for nonrandom data, matching to a pseudorandom synchronization digital code, such as the m~im~l length sequences, will be ~imil~rly improbable. Thus, the first key test for disambiguating sync glyphs from data glyphs focuses on determininE~ whether 10 a given spatially ordered sequence of glyphs encodes values of the known sync code in valid logical order.
As will be appreciated, the linear sync glyph patterns that are proposed hereinabove facilitate this logically ordered sorting of glyphs, but there still is 15 an uncertainty of translation by an integer number of glyphs, together with apossible gross rotational ambiguity as more fully described hereinbelow.
Therefore, when a cyclic sync code having a period of L bits is employed, the testing to identify the glyphs that are encoded with this sync code suitably begins with a search for glyphs that encode a logically ordered subsequence of 20 k bits that matches on the known sync code, where k is long enough to uniquely define synchronized address within a single period of the sync code.
Then the probability of this subsequence randomly matching a translation phase of the sync code is L / 2k of course, if k + L is equal to or greater than m, where m is the total number of sync glyphs in a full line of sync code, then the25 number of valid sequences is reduced from L to (m - k), yielding a random match probability, of (m - k)/ 2k.
2134~94 Once a tentatively valid sync code subsequence is found, a follow-on test typically is performed to confirm that the particular subsequence continues to match on the known sync code as it is being extended by the decode values for one after another of the glyphs that logically extend the subsequence in 5 accordance with the spatial formatting rules. In other words, the apparent sync code pattern is further delineated and verified by confirming that the decode values it yields for subsequences of increasing k continue to match on the known sync code. ~enerally, adequate sync distinguishment is achieved when the probability of mistakenly identifying data as sync is deemed to be 10 acceptable. For most applications, this condition can be satisfied while using a moderately sized k for sync verification because of the fast growth of the exponential probability dependence on k.
Verification of known cyclic sync patterns is further facilitated by the 15 bidirectional (i. e., forward and backward), bit-by-bit verifiability of the logical extensions of such patterns. Furthermore, the use of two dimensional lattices of sync glyphs that encode interlocking cyclic sync codes, as described hereinabove, facilitates robust distinguishment of the sync pattern by providing connectivity among linear strings of sync glyph. As shown in Figs.
20 11 and 12, the additional sync verification tools that are provided by the bidirectional and interlocking logical attributes of this class of sync code areespecially valuable for reestablishing explicit spatial synchronization for the decoding of glyph code when the sync code pattern is disrupted (i. e., logicallybroken erroneously) by local damage and/or degradation of the glyph code.
2134~
As will be understood, absolute verification of synchronization may also be achieved on finite area bounded glyph code patterns, such as shown in Fig. 3, by pre~efining and detecting a geometrically unique subpattern of sync code glyphs for which there is no congruent pattern of glyph centers. In the 5 absence of glyph "dropouts", such a sync code pattern can be verified in the image domain, without having to perform any pattern matching at the bit level (i. e., in the digital domain). For e~ mple, if the glyph code is written on a rectangular array of glyph centers, a sync code pattern that is written into a predetermined column and row of glyphs to span the array completely (i. e., 10 including the border glyphs, as in Figure 3) is geometrically identifiable assync code. Such a geometrically identifiable sync code pattern also provides a degree of absolute 2D addressing, referenced by counting glyphs off this sync pattern, rather than by reference to a boundary feature.
15 It has been found that sync code encoded glyph patterns also can be used to identify and correct ambiguities in glyph codes due to symmetries that exist in the pattern of glyph centers with respect to operations such as image rotations (e.g., 90~, 180~, 270~) and/or image reflections. When these orientational ambiguities transform individual glyphs into permissible members of the 20 glyph code, disambiguation is not trivial.
For example, slash-like glyphs that are tilted to the left or right at 45~ ( i. e., +45~ and -45~, respectively) to encode bit values of "0" or "1," respectively, might be written in accordance with a spatial formatting rule that maps their 25 centers into a square lattice array. As previously pointed out such a glyph pattern has a four-fold rotational ambiguity. Specifically, a 180~ rotation of 2134~9~
. ~
the glyph code pattern m~int~in~ the identities of the individual glyphs because each of the glyphs is transformed into the same, orient~tionally matched member of the glyph set. In other words, the values encoded in the individual glyphs are preserved. However, this 180~ rotation inverts the apparent ordering of linearly aligned glyphs. Fortunately, there is a 5 straightforward solution to this problem because the glyphs still are linearlyaligned, and the direction in which they are aligned is llnch~n~ed. In particular, if the glyph code pattern includes a linear pattern of sync glyphs, as at 61 in Fig. 10, that is encoded in accordance with an asymmetrical sync code sequence (e.g., a m~im~l length shift register sequence),testing for the sync 10 code sequence and its reverse resolves the ambiguity. On the other hand, 90~or 270~ rotations transform each of the two permitted glyphs into the other, thereby interchanging their apparent encoded values. Furthermore, the apparent directions of orthogonal lines (e.g., rows and columns, or 45~
diagonals) also are interchanged. And to further complicate the problem, the 15 apparent ordering of a linear pattern of sync code encoded glyphs may be normal (in the event of a 90~ rotation) or reversed (in the event of a 270~
rotation) with respect to the sync code sequence. Thus, to resolve this ambiguity through the use of a n-stage shift register m~im~l length sequence sync code, a test sequence of at least 2n is employed to avoid inadvertent match20 to a subsequence that is reversed and/or bit-inverted. Hence, it will be understood that four tests for the sync sequence, deciding bit inversion and sequence reversal, will resolve the ambiguity for four rotation states. [0~, 90~, 180~, 270~].
25 Mirror symmetry, which might occur in connection with the use of transparencies, is an additional ambiguity that might be encountered, so ~ 2~3~
additional encoding precautions may be taken to assure ~ qmhiguation For example, to resolve the combination of possible ambiguities, row and column sync codes suitably are selected to be distinct or to at least provide a different phase shift along the code as referenced from sync line intersections.
5 An alternate approach to ~ mbiguation would be to use an asymmetric spatial patterns for the sync code. This requires r~ mhiguation search in image space of glyphs rather than digital space of codes.
For some applications, sync code encoded glyph patterns may be used in 10 combination with coarser synchronization information, such as image reference marks or features, to support a dual phase, coarse/fine synchronization process) In decoding glyph arrays cont~ining glyphs encoding a predefined sync code 15 and variable data glyphs, the predefined sync code glyphs generally are decoded first. For example, in the case of a lattice-like sync code pattern, thedecoding can be initiated at any desired location. To locate nearby sync code encode glyphs possible sync subsequences of glyphs are tested in a local region that is large enough to definitely include a section of sync pattern. If there is 20 concern for orientation ambiguity, these possible sync subsequences are evaluated for each of the possible orientations until a satisfactory result is found. When an initial section of the known sync pattern is found, the sync pattern then is extended glyph-by-glyph. If the glyph code sequence is blocked by damage, alternate sync pattern paths may be followed. When required 2~ jumps to adjacent regions may be logically connected via digital sync code addressing properties. After the sync glyphs have been located as completely ~ 2 1 ~
as possible, variable data may read with logical order referenced and confirmed by local sync glyph code. Optionally the lattice of sync glyph locations may be used to interpolate the expected image positions of vaAable data glyphs.
6 Alternatively, the sync code may be used only after the data code has been transformed into a two dimensional array of decode values. In this event, the sync code is used in a feed forward mode to reorder and correct the decode values as may be required to compensate for decode errors, such as additions, dropouts and shearing of data.
Claims (7)
1. A process for spatially referencing a self-clocking glyph code that is composed of logically ordered data glyphs that are selected from a predetermined symbol set for encoding information in said data glyphs; said process comprising writing said logically ordered data glyphs in a two dimensional field on a recording medium in accordance with a predetermined spatial formatting rule;
writing a synchronization code in said field, said synchronization code being composed of logically ordered glyphs that are distinguishable from said data glyphs, said synchronization code being written in said field in accordancewith said spatial formatting rule plus at least one additional rule that establishes a predetermined geometric relationship among the glyphs of said synchronization code and said data glyphs, whereby the logical ordering of at least some of the data glyphs in said field is inferable from the glyphs of saidsynchronization code based upon said geometric relationship.
writing a synchronization code in said field, said synchronization code being composed of logically ordered glyphs that are distinguishable from said data glyphs, said synchronization code being written in said field in accordancewith said spatial formatting rule plus at least one additional rule that establishes a predetermined geometric relationship among the glyphs of said synchronization code and said data glyphs, whereby the logical ordering of at least some of the data glyphs in said field is inferable from the glyphs of saidsynchronization code based upon said geometric relationship.
2. An optically readable record for storing encoded information; said record comprising a recording medium;
a self-clocking data code for encoding said information; said data code being composed of data glyphs that are written on said recording medium on centers that are spatially distributed in nominal accordance with a two dimensional spatial formatting rule; said data glyphs being defined by respective symbols that are selected from a finite set of optically discriminable symbols to encode said information; and at least one self-clocking synchronization code for establishing a persistent spatial reference for said data code; said synchronization code beingdiscriminable from said data code and being composed of additional glyphs that are written on said recording medium in nominal accordance with said spatial formatting rule plus at least one additional rule that establishes a predetermined geometric relationship among the glyphs of the synchronization code and said data glyphs, whereby the logical ordering of at least some of the data glyphs is inferable from their geometric relationship to the glyphs of the synchronization code.
a self-clocking data code for encoding said information; said data code being composed of data glyphs that are written on said recording medium on centers that are spatially distributed in nominal accordance with a two dimensional spatial formatting rule; said data glyphs being defined by respective symbols that are selected from a finite set of optically discriminable symbols to encode said information; and at least one self-clocking synchronization code for establishing a persistent spatial reference for said data code; said synchronization code beingdiscriminable from said data code and being composed of additional glyphs that are written on said recording medium in nominal accordance with said spatial formatting rule plus at least one additional rule that establishes a predetermined geometric relationship among the glyphs of the synchronization code and said data glyphs, whereby the logical ordering of at least some of the data glyphs is inferable from their geometric relationship to the glyphs of the synchronization code.
3. An optically readable record for storing encoded information; said record comprising a recording medium;
a self-clocking glyph code for storing said information on said recording medium; said glyph code being composed of a data code and at least one a synchronization pattern for providing a persistent spatial reference for said data code said data code being composed of a first logically ordered set of discrete data glyphs that are written on said recording medium on spatially offset centers in nominal accordance with a two dimensional spatial formatting rule;
said data glyphs being defined by respective symbols that are selected from a finite set of optically discriminable symbols to encode said information; and said synchronization pattern being discriminable from said data code and being composed of a second logically ordered set of glyphs that are written on said recording medium in nominal accordance with said spatial formatting rule plus at least one additional rule that establishes a predetermined geometric relationship among the glyphs of the synchronization pattern and said data glyphs, whereby the logical ordering of at least some of the data glyphs is inferable from their geometric relationship to the glyphs of the synchronization pattern.
a self-clocking glyph code for storing said information on said recording medium; said glyph code being composed of a data code and at least one a synchronization pattern for providing a persistent spatial reference for said data code said data code being composed of a first logically ordered set of discrete data glyphs that are written on said recording medium on spatially offset centers in nominal accordance with a two dimensional spatial formatting rule;
said data glyphs being defined by respective symbols that are selected from a finite set of optically discriminable symbols to encode said information; and said synchronization pattern being discriminable from said data code and being composed of a second logically ordered set of glyphs that are written on said recording medium in nominal accordance with said spatial formatting rule plus at least one additional rule that establishes a predetermined geometric relationship among the glyphs of the synchronization pattern and said data glyphs, whereby the logical ordering of at least some of the data glyphs is inferable from their geometric relationship to the glyphs of the synchronization pattern.
4. An optically readable record for storing encoded information; said record comprising a recording medium;
a self-clocking data code for encoding said information; said data code being composed of data glyphs that are written on said recording medium on centers that are spatially distributed in nominal accordance with a two dimensional spatial formatting rule; said data glyphs being defined by respective symbols that are selected from a finite set of optically discriminable symbols to encode said information; and a synchronization code pattern composed of at least two intersecting nonparallel synchronization code subpatterns for establishing a two dimensional, persistent spatial reference for said data code, each of said synchronization patterns being discriminable from said data code and being composed of glyphs that are written on said recording medium in accordance with said spatial formatting rule plus at least one additional rule that establishes a predetermined geometric relationship between the glyphs of the said synchronization patterns and said data glyphs, whereby the logical ordering of at least some of the data glyphs is inferable from their geometric relationship to the glyphs of the synchronization patterns.
a self-clocking data code for encoding said information; said data code being composed of data glyphs that are written on said recording medium on centers that are spatially distributed in nominal accordance with a two dimensional spatial formatting rule; said data glyphs being defined by respective symbols that are selected from a finite set of optically discriminable symbols to encode said information; and a synchronization code pattern composed of at least two intersecting nonparallel synchronization code subpatterns for establishing a two dimensional, persistent spatial reference for said data code, each of said synchronization patterns being discriminable from said data code and being composed of glyphs that are written on said recording medium in accordance with said spatial formatting rule plus at least one additional rule that establishes a predetermined geometric relationship between the glyphs of the said synchronization patterns and said data glyphs, whereby the logical ordering of at least some of the data glyphs is inferable from their geometric relationship to the glyphs of the synchronization patterns.
5. The record of Claim 4 wherein the glyphs of said data code are written into a bounded, two dimensional code field; and said synchronization patterns are linear patterns that are written into said code field in orthogonal alignment with each other, said subpatterns are linear and are written into said code field in orthogonal alignment with each other.
said synchronization patterns being dimensionally coextensive with said code field axially of said synchronization patterns, whereby scale and skew are computable from said synchronization patterns.
said synchronization patterns being dimensionally coextensive with said code field axially of said synchronization patterns, whereby scale and skew are computable from said synchronization patterns.
6. The record of any of Claims 2, 3, 4 or 5 wherein said glyphs are composed of elongated slash-like symbols that oriented at different angles to encode different digital values.
7. The record of Claim 6 wherein said glyphs are oriented at +45° and -45° with respect to a major axis of said recording medium for encoding different binary values.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US08/171,592 US5449895A (en) | 1993-12-22 | 1993-12-22 | Explicit synchronization for self-clocking glyph codes |
US171592 | 1993-12-22 |
Publications (2)
Publication Number | Publication Date |
---|---|
CA2134494A1 CA2134494A1 (en) | 1995-06-23 |
CA2134494C true CA2134494C (en) | 1999-03-16 |
Family
ID=22624362
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CA002134494A Expired - Fee Related CA2134494C (en) | 1993-12-22 | 1994-10-27 | Explicit synchronization for self-clocking glyph codes |
Country Status (5)
Country | Link |
---|---|
US (1) | US5449895A (en) |
EP (1) | EP0660261B1 (en) |
JP (1) | JP3642813B2 (en) |
CA (1) | CA2134494C (en) |
DE (1) | DE69430517T2 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9190105B1 (en) | 2014-05-22 | 2015-11-17 | Guzik Technical Enterprises | Radially-coherent reading and writing system and method for a magnetic disk |
Families Citing this family (140)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5748763A (en) | 1993-11-18 | 1998-05-05 | Digimarc Corporation | Image steganography system featuring perceptually adaptive and globally scalable signal embedding |
US6122403A (en) | 1995-07-27 | 2000-09-19 | Digimarc Corporation | Computer system linked by using information in data objects |
US6681029B1 (en) * | 1993-11-18 | 2004-01-20 | Digimarc Corporation | Decoding steganographic messages embedded in media signals |
US6944298B1 (en) | 1993-11-18 | 2005-09-13 | Digimare Corporation | Steganographic encoding and decoding of auxiliary codes in media signals |
US7171016B1 (en) | 1993-11-18 | 2007-01-30 | Digimarc Corporation | Method for monitoring internet dissemination of image, video and/or audio files |
US6757406B2 (en) | 1993-11-18 | 2004-06-29 | Digimarc Corporation | Steganographic image processing |
US6614914B1 (en) | 1995-05-08 | 2003-09-02 | Digimarc Corporation | Watermark embedder and reader |
US6516079B1 (en) | 2000-02-14 | 2003-02-04 | Digimarc Corporation | Digital watermark screening and detecting strategies |
US6611607B1 (en) | 1993-11-18 | 2003-08-26 | Digimarc Corporation | Integrating digital watermarks in multimedia content |
US5768426A (en) | 1993-11-18 | 1998-06-16 | Digimarc Corporation | Graphics processing system employing embedded code signals |
US6983051B1 (en) | 1993-11-18 | 2006-01-03 | Digimarc Corporation | Methods for audio watermarking and decoding |
US6449377B1 (en) | 1995-05-08 | 2002-09-10 | Digimarc Corporation | Methods and systems for watermark processing of line art images |
JPH07220035A (en) * | 1993-12-22 | 1995-08-18 | Xerox Corp | Optically readable record |
GB2288476A (en) * | 1994-04-05 | 1995-10-18 | Ibm | Authentication of printed documents. |
US5939703A (en) * | 1995-01-03 | 1999-08-17 | Xerox Corporation | Distributed dimensional labeling for dimensional characterization of embedded data blocks |
US5892900A (en) | 1996-08-30 | 1999-04-06 | Intertrust Technologies Corp. | Systems and methods for secure transaction management and electronic rights protection |
US7133846B1 (en) | 1995-02-13 | 2006-11-07 | Intertrust Technologies Corp. | Digital certificate support system, methods and techniques for secure electronic commerce transaction and rights management |
ATE441897T1 (en) | 1995-02-13 | 2009-09-15 | Intertrust Tech Corp | SYSTEMS AND METHODS FOR MANAGING SECURED TRANSACTIONS AND PROTECTING ELECTRONIC RIGHTS |
US5943422A (en) | 1996-08-12 | 1999-08-24 | Intertrust Technologies Corp. | Steganographic techniques for securely delivering electronic digital rights management control information over insecure communication channels |
US6157721A (en) | 1996-08-12 | 2000-12-05 | Intertrust Technologies Corp. | Systems and methods using cryptography to protect secure computing environments |
US6948070B1 (en) | 1995-02-13 | 2005-09-20 | Intertrust Technologies Corporation | Systems and methods for secure transaction management and electronic rights protection |
US6658568B1 (en) | 1995-02-13 | 2003-12-02 | Intertrust Technologies Corporation | Trusted infrastructure support system, methods and techniques for secure electronic commerce transaction and rights management |
US7555139B2 (en) * | 1995-05-08 | 2009-06-30 | Digimarc Corporation | Secure documents with hidden signals, and related methods and systems |
US5761925A (en) * | 1995-12-21 | 1998-06-09 | Ebara Corporation | Absorption heat pump and desiccant assisted air conditioner |
US6000621A (en) * | 1995-12-21 | 1999-12-14 | Xerox Corporation | Tilings of mono-code and dual-code embedded data pattern strips for robust asynchronous capture |
EP0780790A3 (en) * | 1995-12-21 | 2003-06-25 | Xerox Corporation | Tiled embedded data blocks |
USRE39985E1 (en) * | 1995-12-25 | 2008-01-01 | Olympus Optical Co., Ltd. | Information reproducing system, information recording medium, and information recording apparatus |
US20010011253A1 (en) | 1998-08-04 | 2001-08-02 | Christopher D. Coley | Automated system for management of licensed software |
US6381341B1 (en) | 1996-05-16 | 2002-04-30 | Digimarc Corporation | Watermark encoding method exploiting biases inherent in original signal |
US5862255A (en) * | 1996-06-18 | 1999-01-19 | Xerox Corporation | Broad bandwidth image domain communication channel with symbol interference suppression |
US5761686A (en) * | 1996-06-27 | 1998-06-02 | Xerox Corporation | Embedding encoded information in an iconic version of a text image |
US5765176A (en) * | 1996-09-06 | 1998-06-09 | Xerox Corporation | Performing document image management tasks using an iconic image having embedded encoded information |
US5864127A (en) * | 1996-10-10 | 1999-01-26 | Xerox Corporation | Analog glyph detector and detector arrays |
US5901224A (en) * | 1996-10-21 | 1999-05-04 | Xerox Corporation | Quasi-reprographics with variable embedded data with applications to copyright management, and distribution control |
US6327395B1 (en) | 1996-12-20 | 2001-12-04 | Xerox Parc | Glyph address carpet methods and apparatus for providing location information in a multidimensional address space |
US5937110A (en) | 1996-12-20 | 1999-08-10 | Xerox Corporation | Parallel propagating embedded binary sequences for characterizing objects in N-dimensional address space |
US5920861A (en) | 1997-02-25 | 1999-07-06 | Intertrust Technologies Corp. | Techniques for defining using and manipulating rights management data structures |
US6112181A (en) | 1997-11-06 | 2000-08-29 | Intertrust Technologies Corporation | Systems and methods for matching, selecting, narrowcasting, and/or classifying based on rights management and/or other information |
US6330976B1 (en) | 1998-04-01 | 2001-12-18 | Xerox Corporation | Marking medium area with encoded identifier for producing action through network |
WO1999050736A1 (en) * | 1998-04-01 | 1999-10-07 | Xerox Corporation | Paper indexing of recordings |
US6325420B1 (en) | 1998-08-17 | 2001-12-04 | Inspectron Corporation | Method for embedding non-intrusive encoded data in printed matter and system for reading same |
US6470096B2 (en) | 1998-11-13 | 2002-10-22 | Xerox Corporation | Method for locating user interface tags in a document processing system |
US7168036B2 (en) | 1998-11-13 | 2007-01-23 | Xerox Corporation | User interface identification and service tags for a document processing system |
US7394573B1 (en) * | 1999-04-14 | 2008-07-01 | Xerox Corporation | System for authenticating hardcopy documents |
US6641051B1 (en) * | 1999-09-24 | 2003-11-04 | Xerox Corporation | System for embedded digital data that allows embedding of data around known obstructions |
SE517445C2 (en) | 1999-10-01 | 2002-06-04 | Anoto Ab | Position determination on a surface provided with a position coding pattern |
US6296192B1 (en) | 1999-12-16 | 2001-10-02 | Xerox Corporation | Machine-readable record with a two-dimensional lattice of synchronization code interleaved with data code |
US6714677B1 (en) * | 1999-12-17 | 2004-03-30 | Xerox Corporation | Use of correlation histograms for improved glyph decoding |
US6854821B2 (en) | 2000-04-05 | 2005-02-15 | Anoto Ab | Systems and methods for printing by using a position-coding pattern |
US7493250B2 (en) * | 2000-12-18 | 2009-02-17 | Xerox Corporation | System and method for distributing multilingual documents |
US6726104B2 (en) * | 2000-12-18 | 2004-04-27 | Symbol Technologies, Inc. | Scaling techniques for printing bar code symbols |
US6966495B2 (en) * | 2001-06-26 | 2005-11-22 | Anoto Ab | Devices method and computer program for position determination |
US6732927B2 (en) * | 2001-06-26 | 2004-05-11 | Anoto Ab | Method and device for data decoding |
TWI235926B (en) | 2002-01-11 | 2005-07-11 | Sonix Technology Co Ltd | A method for producing indicators and processing system, coordinate positioning system and electronic book system utilizing the indicators |
US7359884B2 (en) * | 2002-03-14 | 2008-04-15 | Contentguard Holdings, Inc. | Method and apparatus for processing usage rights expressions |
US20030229593A1 (en) * | 2002-03-14 | 2003-12-11 | Michael Raley | Rights expression profile system and method |
US7107009B2 (en) * | 2002-06-26 | 2006-09-12 | Nokia Corporation | Method, system and computer program product for personalizing the functionality of a personal communication device |
US20040010446A1 (en) * | 2002-07-08 | 2004-01-15 | Marko Vanska | Mobile customer relationship management |
US7274909B2 (en) * | 2002-10-31 | 2007-09-25 | Nokia Corporation | Method and system for selecting data items for service requests |
US7072672B1 (en) * | 2002-11-01 | 2006-07-04 | Nokia Corporation | Disposable mini-applications |
US20040093274A1 (en) * | 2002-11-08 | 2004-05-13 | Marko Vanska | Method and apparatus for making daily shopping easier |
US8719053B2 (en) | 2003-07-17 | 2014-05-06 | Ventana Medical Systems, Inc. | Laboratory instrumentation information management and control network |
US7860727B2 (en) * | 2003-07-17 | 2010-12-28 | Ventana Medical Systems, Inc. | Laboratory instrumentation information management and control network |
US7407107B2 (en) * | 2003-12-08 | 2008-08-05 | Nokia Corporation | Apparatus, system, method and computer program product for creating shortcuts to functions in a personal communication device |
EP1749273A4 (en) * | 2004-05-18 | 2011-12-28 | Silverbrook Res Pty Ltd | Authentication of an object using a signature encoded in a number of data portions |
US7304585B2 (en) * | 2004-07-02 | 2007-12-04 | Nokia Corporation | Initiation of actions with compressed action language representations |
JP4556705B2 (en) * | 2005-02-28 | 2010-10-06 | 富士ゼロックス株式会社 | Two-dimensional coordinate identification apparatus, image forming apparatus, and two-dimensional coordinate identification method |
US8181261B2 (en) * | 2005-05-13 | 2012-05-15 | Xerox Corporation | System and method for controlling reproduction of documents containing sensitive information |
JP4532349B2 (en) | 2005-06-07 | 2010-08-25 | 株式会社リコー | Image processing method, image processing apparatus, image processing program, and recording medium |
US7481374B2 (en) * | 2005-06-08 | 2009-01-27 | Xerox Corporation | System and method for placement and retrieval of embedded information within a document |
JP4810918B2 (en) * | 2005-08-01 | 2011-11-09 | 富士ゼロックス株式会社 | Code pattern image generating apparatus and method, code pattern image reading apparatus and method, and code pattern image medium |
US8050502B2 (en) * | 2006-06-21 | 2011-11-01 | Namco Bandai Games Inc. | Two-Dimensional code generation method, two-dimensional code, two-dimensional code recognition method, and image recognition device |
JP5557419B2 (en) * | 2007-10-17 | 2014-07-23 | スパンション エルエルシー | Semiconductor device |
US7991153B1 (en) | 2008-08-26 | 2011-08-02 | Nanoglyph, LLC | Glyph encryption system and related methods |
JP5412778B2 (en) * | 2008-09-18 | 2014-02-12 | 富士ゼロックス株式会社 | Business support system |
AU2008255212A1 (en) * | 2008-12-10 | 2010-06-24 | Canon Kabushiki Kaisha | Page mark-up using printed dot barcodes |
AU2008255227A1 (en) * | 2008-12-10 | 2010-06-24 | Canon Kabushiki Kaisha | Document security method |
JP5321352B2 (en) * | 2009-08-27 | 2013-10-23 | 富士ゼロックス株式会社 | Two-dimensional code generation apparatus, image forming system, two-dimensional code reading apparatus, computer program, and medium |
US8453922B2 (en) * | 2010-02-09 | 2013-06-04 | Xerox Corporation | Method for one-step document categorization and separation using stamped machine recognizable patterns |
US9288089B2 (en) | 2010-04-30 | 2016-03-15 | Ecole Polytechnique Federale De Lausanne (Epfl) | Orthogonal differential vector signaling |
US9246713B2 (en) | 2010-05-20 | 2016-01-26 | Kandou Labs, S.A. | Vector signaling with reduced receiver complexity |
US8593305B1 (en) | 2011-07-05 | 2013-11-26 | Kandou Labs, S.A. | Efficient processing and detection of balanced codes |
US9124557B2 (en) | 2010-05-20 | 2015-09-01 | Kandou Labs, S.A. | Methods and systems for chip-to-chip communication with reduced simultaneous switching noise |
US9985634B2 (en) | 2010-05-20 | 2018-05-29 | Kandou Labs, S.A. | Data-driven voltage regulator |
US9251873B1 (en) | 2010-05-20 | 2016-02-02 | Kandou Labs, S.A. | Methods and systems for pin-efficient memory controller interface using vector signaling codes for chip-to-chip communications |
US9596109B2 (en) | 2010-05-20 | 2017-03-14 | Kandou Labs, S.A. | Methods and systems for high bandwidth communications interface |
US9288082B1 (en) | 2010-05-20 | 2016-03-15 | Kandou Labs, S.A. | Circuits for efficient detection of vector signaling codes for chip-to-chip communication using sums of differences |
US9479369B1 (en) | 2010-05-20 | 2016-10-25 | Kandou Labs, S.A. | Vector signaling codes with high pin-efficiency for chip-to-chip communication and storage |
US9564994B2 (en) | 2010-05-20 | 2017-02-07 | Kandou Labs, S.A. | Fault tolerant chip-to-chip communication with advanced voltage |
US9450744B2 (en) | 2010-05-20 | 2016-09-20 | Kandou Lab, S.A. | Control loop management and vector signaling code communications links |
US9077386B1 (en) | 2010-05-20 | 2015-07-07 | Kandou Labs, S.A. | Methods and systems for selection of unions of vector signaling codes for power and pin efficient chip-to-chip communication |
WO2011151469A1 (en) | 2010-06-04 | 2011-12-08 | Ecole Polytechnique Federale De Lausanne | Error control coding for orthogonal differential vector signaling |
US8430315B2 (en) | 2010-06-29 | 2013-04-30 | SignaKey LLC | Data encoding and decoding |
US9275720B2 (en) | 2010-12-30 | 2016-03-01 | Kandou Labs, S.A. | Differential vector storage for dynamic random access memory |
US9268683B1 (en) | 2012-05-14 | 2016-02-23 | Kandou Labs, S.A. | Storage method and apparatus for random access memory using codeword storage |
CN105122758B (en) | 2013-02-11 | 2018-07-10 | 康杜实验室公司 | High bandwidth interchip communication interface method and system |
CN105379170B (en) | 2013-04-16 | 2019-06-21 | 康杜实验室公司 | High-bandwidth communication interface method and system |
CN105393512B (en) | 2013-06-25 | 2019-06-28 | 康杜实验室公司 | Vector signaling with low receiver complexity |
WO2015077608A1 (en) | 2013-11-22 | 2015-05-28 | Kandou Labs SA | Multiwire linear equalizer for vector signaling code receiver |
US9806761B1 (en) | 2014-01-31 | 2017-10-31 | Kandou Labs, S.A. | Methods and systems for reduction of nearest-neighbor crosstalk |
WO2015117102A1 (en) | 2014-02-02 | 2015-08-06 | Kandou Labs SA | Method and apparatus for low power chip-to-chip communications with constrained isi ratio |
WO2015131203A1 (en) | 2014-02-28 | 2015-09-03 | Kandou Lab, S.A. | Clock-embedded vector signaling codes |
US9509437B2 (en) | 2014-05-13 | 2016-11-29 | Kandou Labs, S.A. | Vector signaling code with improved noise margin |
US11240076B2 (en) | 2014-05-13 | 2022-02-01 | Kandou Labs, S.A. | Vector signaling code with improved noise margin |
US9148087B1 (en) | 2014-05-16 | 2015-09-29 | Kandou Labs, S.A. | Symmetric is linear equalization circuit with increased gain |
US9852806B2 (en) | 2014-06-20 | 2017-12-26 | Kandou Labs, S.A. | System for generating a test pattern to detect and isolate stuck faults for an interface using transition coding |
US9112550B1 (en) | 2014-06-25 | 2015-08-18 | Kandou Labs, SA | Multilevel driver for high speed chip-to-chip communications |
WO2016007863A2 (en) | 2014-07-10 | 2016-01-14 | Kandou Labs, S.A. | Vector signaling codes with increased signal to noise characteristics |
US9432082B2 (en) | 2014-07-17 | 2016-08-30 | Kandou Labs, S.A. | Bus reversable orthogonal differential vector signaling codes |
KR102243423B1 (en) | 2014-07-21 | 2021-04-22 | 칸도우 랩스 에스에이 | Multidrop data transfer |
US9461862B2 (en) | 2014-08-01 | 2016-10-04 | Kandou Labs, S.A. | Orthogonal differential vector signaling codes with embedded clock |
US9674014B2 (en) | 2014-10-22 | 2017-06-06 | Kandou Labs, S.A. | Method and apparatus for high speed chip-to-chip communications |
KR102517583B1 (en) | 2015-06-26 | 2023-04-03 | 칸도우 랩스 에스에이 | High speed communications system |
CN109359657B (en) | 2015-10-19 | 2022-11-25 | 松翰科技股份有限公司 | Index structure |
US9557760B1 (en) | 2015-10-28 | 2017-01-31 | Kandou Labs, S.A. | Enhanced phase interpolation circuit |
US9577815B1 (en) | 2015-10-29 | 2017-02-21 | Kandou Labs, S.A. | Clock data alignment system for vector signaling code communications link |
US10055372B2 (en) | 2015-11-25 | 2018-08-21 | Kandou Labs, S.A. | Orthogonal differential vector signaling codes with embedded clock |
US10003315B2 (en) | 2016-01-25 | 2018-06-19 | Kandou Labs S.A. | Voltage sampler driver with enhanced high-frequency gain |
CN116012992A (en) | 2016-03-08 | 2023-04-25 | 达斯特一致有限责任公司 | Generating unique codes from orientation information |
US10003454B2 (en) | 2016-04-22 | 2018-06-19 | Kandou Labs, S.A. | Sampler with low input kickback |
CN115085727A (en) | 2016-04-22 | 2022-09-20 | 康杜实验室公司 | High performance phase locked loop |
US10153591B2 (en) | 2016-04-28 | 2018-12-11 | Kandou Labs, S.A. | Skew-resistant multi-wire channel |
US10056903B2 (en) | 2016-04-28 | 2018-08-21 | Kandou Labs, S.A. | Low power multilevel driver |
US10333741B2 (en) | 2016-04-28 | 2019-06-25 | Kandou Labs, S.A. | Vector signaling codes for densely-routed wire groups |
US9906358B1 (en) | 2016-08-31 | 2018-02-27 | Kandou Labs, S.A. | Lock detector for phase lock loop |
US10411922B2 (en) | 2016-09-16 | 2019-09-10 | Kandou Labs, S.A. | Data-driven phase detector element for phase locked loops |
US10200188B2 (en) | 2016-10-21 | 2019-02-05 | Kandou Labs, S.A. | Quadrature and duty cycle error correction in matrix phase lock loop |
US10200218B2 (en) | 2016-10-24 | 2019-02-05 | Kandou Labs, S.A. | Multi-stage sampler with increased gain |
US10372665B2 (en) | 2016-10-24 | 2019-08-06 | Kandou Labs, S.A. | Multiphase data receiver with distributed DFE |
EP4216444A1 (en) | 2017-04-14 | 2023-07-26 | Kandou Labs, S.A. | Pipelined forward error correction for vector signaling code channel |
CN110945830B (en) | 2017-05-22 | 2022-09-09 | 康杜实验室公司 | Multimode data driven clock recovery circuit |
US10116468B1 (en) | 2017-06-28 | 2018-10-30 | Kandou Labs, S.A. | Low power chip-to-chip bidirectional communications |
US10686583B2 (en) | 2017-07-04 | 2020-06-16 | Kandou Labs, S.A. | Method for measuring and correcting multi-wire skew |
US10693587B2 (en) | 2017-07-10 | 2020-06-23 | Kandou Labs, S.A. | Multi-wire permuted forward error correction |
US10203226B1 (en) | 2017-08-11 | 2019-02-12 | Kandou Labs, S.A. | Phase interpolation circuit |
US10467177B2 (en) | 2017-12-08 | 2019-11-05 | Kandou Labs, S.A. | High speed memory interface |
US10326623B1 (en) | 2017-12-08 | 2019-06-18 | Kandou Labs, S.A. | Methods and systems for providing multi-stage distributed decision feedback equalization |
KR102347396B1 (en) | 2017-12-28 | 2022-01-04 | 칸도우 랩스 에스에이 | Synchronously Switched Multi-Input Demodulating Comparator |
US10554380B2 (en) | 2018-01-26 | 2020-02-04 | Kandou Labs, S.A. | Dynamically weighted exclusive or gate having weighted output segments for phase detection and phase interpolation |
US11831472B1 (en) | 2022-08-30 | 2023-11-28 | Kandou Labs SA | Pre-scaler for orthogonal differential vector signalling |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5168147A (en) * | 1990-07-31 | 1992-12-01 | Xerox Corporation | Binary image processing for decoding self-clocking glyph shape codes |
CA2044404C (en) * | 1990-07-31 | 1998-06-23 | Dan S. Bloomberg | Self-clocking glyph shape codes |
US5128525A (en) * | 1990-07-31 | 1992-07-07 | Xerox Corporation | Convolution filtering for decoding self-clocking glyph shape codes |
US5091966A (en) * | 1990-07-31 | 1992-02-25 | Xerox Corporation | Adaptive scaling for decoding spatially periodic self-clocking glyph shape codes |
US5245165A (en) * | 1991-12-27 | 1993-09-14 | Xerox Corporation | Self-clocking glyph code for encoding dual bit digital values robustly |
US5221833A (en) * | 1991-12-27 | 1993-06-22 | Xerox Corporation | Methods and means for reducing bit error rates in reading self-clocking glyph codes |
-
1993
- 1993-12-22 US US08/171,592 patent/US5449895A/en not_active Expired - Lifetime
-
1994
- 1994-10-27 CA CA002134494A patent/CA2134494C/en not_active Expired - Fee Related
- 1994-12-06 EP EP94309085A patent/EP0660261B1/en not_active Expired - Lifetime
- 1994-12-06 DE DE69430517T patent/DE69430517T2/en not_active Expired - Lifetime
- 1994-12-16 JP JP31374494A patent/JP3642813B2/en not_active Expired - Lifetime
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9190105B1 (en) | 2014-05-22 | 2015-11-17 | Guzik Technical Enterprises | Radially-coherent reading and writing system and method for a magnetic disk |
Also Published As
Publication number | Publication date |
---|---|
EP0660261A2 (en) | 1995-06-28 |
DE69430517D1 (en) | 2002-06-06 |
EP0660261B1 (en) | 2002-05-02 |
JP3642813B2 (en) | 2005-04-27 |
US5449895A (en) | 1995-09-12 |
DE69430517T2 (en) | 2002-10-31 |
JPH07225803A (en) | 1995-08-22 |
EP0660261A3 (en) | 1998-09-16 |
CA2134494A1 (en) | 1995-06-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CA2134494C (en) | Explicit synchronization for self-clocking glyph codes | |
US5449896A (en) | Random access techniques for use with self-clocking glyph codes | |
US5521372A (en) | Framing codes for robust synchronization and addressing of self-clocking glyph codes | |
EP0660260B1 (en) | Global addressability for self-clocking glyph codes | |
US6182901B1 (en) | Orientational disambiguation for self-clocking glyph codes | |
US6000621A (en) | Tilings of mono-code and dual-code embedded data pattern strips for robust asynchronous capture | |
US5717197A (en) | Tiled embedded data block-types for generic embedded data block systems | |
US5694102A (en) | Vector reconstruction of asynchronously captured tiled embedded data blocks | |
US5937110A (en) | Parallel propagating embedded binary sequences for characterizing objects in N-dimensional address space | |
CN100517378C (en) | Method for encoding and decoding two-dimensional code | |
US5611575A (en) | Distributed state flags or other unordered information for embedded data blocks | |
CA2502483C (en) | Orientation-indicating cyclic position codes | |
US5572010A (en) | Distributed type labeling for embedded data blocks | |
US5790715A (en) | Method and apparatus for recording/reproducing mesh pattern data | |
US5481103A (en) | Packet bar code with data sequence encoded in address/data packets | |
JPH0778817B2 (en) | A method for decoding a bitmap image space representation of a self-clocking symbol shape code | |
EP0849704B1 (en) | Multi-dimensional address spaces | |
US6341730B1 (en) | Method of encoding embedded data blocks containing occlusions | |
US20060289661A1 (en) | Bi-directional readable two-dimensional bar code system and its reading method | |
EP0780790A2 (en) | Tiled embedded data blocks | |
US5870510A (en) | Parallel propagating embedded binary sequences for parameterizing two dimensional image domain code patterns in N-dimensional address space | |
AU2003266852B2 (en) | Orientation-indicating cyclic position codes | |
JPH09214733A (en) | Picture capture mechanism | |
JPH08235333A (en) | Buried data block |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
EEER | Examination request | ||
MKLA | Lapsed |