WO2001017263A1

WO2001017263A1 - Time multiplex method for recording, transmitting and reproducing video images from different video sources

Info

Publication number: WO2001017263A1
Application number: PCT/EP2000/007815
Authority: WO
Inventors: Matthias G. DÖRING
Original assignee: Geutebrück Gmbh & Co. Kg
Priority date: 1999-08-31
Filing date: 2000-08-09
Publication date: 2001-03-08
Also published as: PL353391A1; DE19941322B4; DE19941322A1; AU6837400A

Abstract

The invention relates to a time multiplex method for recording, transmitting and reproducing, on a monitor, digitized video images from different video sources (1a, 1b,..., 1c) in a closed circuit television (CCTV) system. In order to achieve parity locking (5) in each multiplex channel, the line offset of recorded semi-images is compensated independently from the parity thereof in such a way that the lines are expanded to form full images by respectively inserting a line in between adjacent lines, whereby said full images are used for reproduction on the monitor. For the purposes of insertion, a previous line can be copied or an interpolation of two adjacent lines can be used. The inventive method can be used with CCTV systems working with the multiplexing system, i.e. industrial television, corporate television or TV monitoring.

Description

Time-division multiplexing method for recording, transmission and playback of video images from various video sources

The invention relates to a time-division multiplex method for recording, transmission and monitor playback of digitized video images from various video sources in a CCTV (Closed Circuit Television) system.

For the recording, transmission and display of video data in CCTV systems, it is known to use multiplexing methods in order to reduce the number and thus the high costs of compression and digitization channels. For this purpose, the devices for video signal digitization and compression are preceded by an electronic analog crossbar switch (time-division multiplex switch). This is switched according to the recording rates to be realized for the various video sources of the CCTV system. The digitized and compressed video data are then recorded in a video database (recorder).

The time frames during the permanent recording or an additional event-triggered recording for each video source, usually these are video cameras, lead to a complex, generally non-deterministic time behavior for the acquisition of the individual fields and their assignment to the different multiplex channels. The recording grid is generally not synchronized with

Time grid of the video sources and the field parity. It is subject to constant, unpredictable changes due to the parameterization of the system or through event-controlled recording.

An essential performance parameter of a multiplex recording of several video sources is the total frame rate. This code number is the proportion of fields that can be used for the recording. Due to the possibly large temporal spacing of the individual fields recorded for a multiplex channel from one another, it is not sensible to combine fields into full images. Instead, the representation on the video image monitor, for example on a PC monitor, takes place in a known manner as a scaled field by time interpolation or simple doubling. If fields of different parity are mixed in the display of a multiplex channel, this leads to a jump by two lines with the original image size, which has a visually extremely disruptive effect.

It is therefore known to try to record only fields of one and the same parity in the multiplex recording for each multiplex channel. In this case, fields with the parity assigned to the multiplex channel are waited for, or generally only fields with a certain parity are used. However, this creates a number of significant problems. This reduces the total frame rate and thus the performance of the overall system. In addition, the response times for event-controlled recording are increased. In addition, a great deal of planning is required to optimize the recording behavior through parity-synchronous recording grids. In addition, complex algorithms must be developed and used to optimize the recording control.

The invention has for its object to design a time-division multiplexing method for recording, transmission and monitor playback of digitized video images from various video sources in a CCTV (Closed Circuit Television) system in such a way that better overall image rates result without loss of information and at the same time the Recording is not limited to fields of parity.

According to the invention, which relates to a method of the type mentioned at the outset, this object is achieved in that, for interlacing correction in each multiplex channel, the line offset of recorded fields is compensated for independently of their parity in such a way that they are inserted by inserting one Line between adjacent lines to full frames ex- panded, which are then used in playback.

In the interlace correction method (parity correction; "parity locking") operating according to the invention, it is generally assumed that the different channels of a multiplexed CCTV system supply fields of both parities. Using an algorithm, the line offset of fields of different parity can be compensated so that the line jump no longer appears visually. The method according to the invention is initially based on fields of the corresponding number of lines, e.g. according to the CCIR 288 standard. These fields are expanded into frames by inserting one line between adjacent lines.

To insert the lines, there are advantageous options either to copy the previous line or to use an interpolation of two adjacent lines.

Depending on the parity (even field or odd field), the full image obtained in this way is displayed directly (even field) or previously processed in a further step (odd field) in such a way that no jump is noticed when the images are strung together. This further step can expediently consist in that special fill lines are inserted at the beginning and / or end of the full image formed from the field.

Advantageously, the export of individual images from the CCTV system as an interlaced full image is also carried out in accordance with the method according to the invention.

When using the method according to the invention, the recording is not limited to fields of one of the two parities and this results in the implementation considerably simpler recording algorithms than before. There is no loss of information in the video source data.

Compared to the conventional methods, there are higher total image rates, which correspond to the rates that the hardware provides. In the case of synchronous source signals, these are 50 frames per second (in the case of a digitizing / compression channel) or rates reduced by synchronization losses.

The use of the method according to the invention also enables simpler system planning than hitherto, since the factor of field parity no longer plays a role in the planning of the recording grid.

Statistically, when using the method according to the invention, there are improved response times in event-controlled recording, since the field immediately following an event can be used for the recording.

In the following the invention will be explained with reference to drawings. Show it:

1 shows a block diagram of a schematic of a CCTV (closed circuit) system working with digital multiplex recording with interlace compensation (parity correction, "parity locking") in the playback path according to the invention,

FIG. 2 shows in a line diagram a first variant of the method according to the invention for compensating for the line jump, and

Figure 3 also in a line diagram a second variant of the method according to the invention for compensating the interlacing. In the CCTV system shown in FIG. 1, a plurality of video sources 1 a, 1 b,..., In, for example video cameras, are provided, the video signals of which are fed to an analog electronic crossbar switch 2 (time multiplexer). The crossbar switch 2 is therefore provided in order to reduce the number and hence the effort of compression and digitization channels. The crossbar switch 2 is followed by a digitizer / compressor 3, the output of which is connected to a video database 4 for storing the digitized video data. The crossbar switch 2 is switched in accordance with the recording rates to be implemented for the various video sources Ia, Ib, ..., In of the CCTV system. The for each video source la, lb, ..., in different time frames during the uninterrupted recording and, if necessary, an additional event-controlled recording result in a complex, generally non-deterministic time behavior for the acquisition of the individual fields and their assignment to the individual multi - plex channels. The recording grid is generally not synchronized with the time grid of the video sources la, lb, ..., In and the field parity and, moreover, is subject to unpredictable changes.

The digital video data which can be extracted from the video database 4 are decompressed in a decompression circuit 6 before they are treated in a line interlacing compensator 5 ("parity locking") according to the method according to the invention. The video data processed in the interlace compensator 5 are finally fed to a PC monitor 7 for display there in the playback path. In the interlace compensator 5, based on the knowledge that the different channels of a system operated in time-division multiplexing supply fields of both parities, the line offset of fields of different parity is compensated with the aid of an algorithm in such a way that the interlacing no longer occurs visually. In the process that uses interlaced pensator 5 is carried out, corresponding fields of the corresponding number of lines are initially assumed (for example CCIR 288). In interlaced compensator 5, these fields are expanded into full images by inserting one line between adjacent lines. To insert the lines, either the previous line is copied or an interpolation of two adjacent lines is carried out. The full image thus obtained with the help of the interlace compensator 5 is then directly displayed on the PC monitor 7, depending on the parity of the field from which the full image was generated, or is previously processed in a further step in such a way that there is no discernible jump in the sequence of images.

Figures 2 and 3 illustrate the method according to the invention using two variants of the correction of line breaks. Figures 2 (variant 1) and 3 (variant 2) each consist of three juxtaposed representations of a line scheme, the line scheme shown on the left in each case a normal full image, the middle line scheme a padded straight field (even parity) and the one shown on the right Line scheme represent a padded odd field (odd parity). In Figures 2 and 3, the lines of even parity with solid lines, the lines of odd parity with dashed lines, inserted fill lines with dash-dotted lines and fill lines at the end or at the beginning of an image with dotted lines are shown. In both variants, a fill line is inserted between adjacent lines of even parity in the even field for interlace correction, and a final fill line is added below in the direction of line doubling. The direction of the line doubling is in the straight field from top to bottom in both variants. The variant 1 shown in FIG. 2 and the variant 2 shown in FIG. 3 differ only with regard to the line correction in the case of posed odd field. In the first variant (FIG. 2) the direction of the line doubling is from bottom to top and in the second variant (FIG. 3) from top to bottom as in the even field. In the first variant (FIG. 2), the full image obtained from the odd field is supplemented beforehand by two special fill lines at the bottom, that is to say in the beginning in the direction of line doubling, so that there is no jump in the sequence of the images. In the second variant (FIG. 3), the full image obtained from the odd field is supplemented beforehand by a special fill line at the top and a special fill line at the bottom, i.e. at the beginning and end in the direction of line doubling, so that the sequence in time of the No jump can be seen in the pictures either. The usual, dash-dotted lines

Filling lines are identified in FIGS. 2 and 3 with a prime index for the respective line number.

Claims

claims

1. time-division multiplex method for recording, transmission and monitor playback of digitized video images from various video sources in a CCTV (Closed Circuit Television) system,

characterized,

that for interlace correction in each multiplex channel, the line offset of recorded fields is compensated for regardless of their parity in such a way that they are expanded by inserting one line between adjacent lines into frames, which are then used during playback.

2. The method according to claim 1,

characterized,

that the previous line is always copied to insert the lines.

The method of claim 1 or 2,

characterized,

that an interpolation of two adjacent lines is carried out to insert the lines.

Method according to one of the preceding claims,

characterized,

that the full frame obtained, depending on the parity of the field from which a full frame is assumed, is reproduced directly (even field image) or is processed in a further step (odd field) in such a way that there is no noticeable jump in the sequence of images.

5. The method according to claim 4,

characterized,

that the further step consists in that filler lines are inserted at the beginning and / or end of the full image formed from the field.

6. The method according to any one of the preceding claims,

marked by

its application for the export of individual images from the CCTV system in such a way that interlaced frames are taken for this in accordance with this method.