DE2624038C3 - Method and circuit arrangement for the transmission of images via image channels with a low bandwidth - Google Patents

Description

The invention relates to a method and a circuit arrangement for transmitting images via Image channels with low bandwidth between a sending and a receiving point of a Video telephone system in which the moving images are recorded with a high level of detail, but cannot be transferred in this way.

Numerous circuit arrangements are already known which take this fact into account and therefore send pictures with higher detail but reduced motion resolution to the remote station. At a the receiving station is equipped with a memory for such slow image transmission is scanned until a new image has arrived from the sending point. Until then he'll see Participant a still image. This method also has the disadvantage that there is a large one for each recipient Memory must be provided which stores the brightness values of all pixels contained in an image has to record.

Other known circuit arrangements instead to * fi mekamera maintain the high resolution of the motion Aufnah- and for that to transfer a few pixels per image, so that the resolution of detail is reduced.

Further circuit arrangements, for example according to DE-OS 23 20 243, allow between the f> o to choose one and the other method and to switch the transmitter and receiver if necessary. For one such image transmission, however, continues to require the large pixel memory.

Finally, it is also known, for example from ^ΙιΓ > Bell Lab. Record, May / June 1969, pages 143-147, to increase the detail resolution by selecting an image section from the entire image, which is then transmitted. All other parts of the image are then not shown on the receiving station's screen.

The object of the invention is to provide an image transmission option to provide with the garoje of the Capture camera scanned image is visible on the screen, and still at the higher detail resolution is not entirely dispensed with.

This object is achieved according to the invention in that the receiving station specified image section with higher detail and reduced motion resolution, the rest of the image but is transmitted with lower detail and higher motion resolution.

The invention is based on the assumption that the human eye can detect or read accurately of a document only adjusts to a small sub-area of the overall picture It is therefore not required to constantly offer the overall picture with a higher resolution, but it is sufficient to have one in his Select a shape that can be freely selected and moved in the image field with a certain maximum area, within which the detail resolution is optimal and the motion resolution is reduced. In this way does the user have the. Possibility to always have an overview of the overall picture of a template and at the same time to capture the sub-areas of interest with sufficient reproduction sharpness. Since the motion resolution is only within a limited image section is reduced, z. B. the partner in a television telephone system using a pen or light pen operate at will in the other areas. This would be the case for a full, slowly transferred image extremely annoying because of the smearing and blurring effects that occur here.

Another advantage of the invention is that the image memory is not a whole frame, but only one Part, e.g. B. must store a tenth of the image information. The cost of such scaled-down storage can be significantly reduced, so that, for example, subscriber terminals in one Video telephone system can be equipped with such a facility.

An example of the invention is described with reference to the drawings. While in FIG. 1 the for the Invention essential details of a video telephone system are shown in Fig.2 and 3 of the Signal curve of two picture lines shown, which contain information transmitted according to two standards. Based on FIG. 4 a line reduction is described.

For the example of the invention it is assumed that the known point comb method is used for the transmission of the detailed image section of interest, referred to in the following window. This window F is on the screen B2 of the receiving station ßin Fi g. 1 shown.

Depending on the original text or image D, the shape of this window can be selected and shifted from a large number of predefined possibilities with an operating device. This can e.g. B. consist of a light pen with which a corner or the center of the desired field is indicated on the screen. Another method is to use the light pen to mark the image section by circling it. The window of interest can also be specified with a joystick or a trackball.

The local position information of the selected screen

sters and their changes are preferably done digitally in a format-dependent grid, because a analog shift could lead to blurring effects.

The video signal of the original D scanned with a hoot-resolving camera Ki is sent either via a switch S1 directly to an image channel ή, which connects the subscriber device of the sending station } A via lines and exchanges L to the receiving station B , or when the image window is scanned an A \ & asteineinrichtung E 1. This is followed by a low-pass pulse shaper P , which broadens the sampling pulses so that their spectrum lies within the bandwidth of the transmission channel

The image signals captured by the camera K 1 are thereby adapted b for the direct transmission to the bandwidth of the image channel that either a low pass limits the high frame rates, or that the focusing of the scanning beam of the recording camera K is reduced. 1

As soon as the participant B sees the image of the original D on his screen B 2, he selects a certain image window, for example by means of a light pen and a format preselection button, and thus controls an image section generator G synchronized with the screen B 2 . This inserts the position and format commands in coded form into the timing grid of the camera (not shown here) of participant B , e.g. B. at the time of the Bidlaustastlücken. However, it is also possible to switch off the recording camera of participant B when starting up the light pen or when pressing the format preselection button and to transmit the coded coordinates of the image window in another way via the image channels a leading to participant A.

A code evaluator C is connected to the image channel a at subscriber A, which determines the size and position of the window from the incoming signals and then switches the electronic switch S1 accordingly at the right moment. When switching over to the one shown in FIG. 1, instead of the normal camera signal, the signal from the low-pass pulse shaper P, processed in the manner described using the point comb method, arrives at the image channel b.

At the receiver B , a blanking device E 2 is connected to the image channel b and is synchronized by the sampling rate of the scanning device E 1. A converter R ensures that the pulses obtained by the scanning device El reach the memory Sp in their correct width via the switch S3. All stored image points are then read out using switch S2 in the position shown

15 for the duration of the selected image window. In the rest of the time, the directly transmitted image signal is reproduced on screen B 2;

The image repetition memory Sp can be designed, for example, in the form of an analog shift register. Each window pixel is read out there (n- 1) times and replaced by a new one on the / jth time according to the point refresh principle

The image signal of such a mixed line is shown as an example in FIG. This is followed by the short scanning pulses of a width T each obtained by the device £ 1. These pulses in the second image A 2 lie between those of the first image A 1. This fact is illustrated even more clearly by FIG. 2 circled two places enlarged. A larger space is left between the pulse Λ2 and Al. This is intended to indicate that with n-fold scanning there are still further pulses at this point if a greater detail resolution is desired.

With the arrangement described so far, an increased horizontal resolution is achieved within the image section. However, it appears sensible to distribute the information gain evenly between the vertical and horizontal within the window by reducing the movement resolution. With a / η-fold reduction, for example from 4 MHz to 1 MHz, the number of lines for the overall image is therefore also reduced. With m = 4, this can be done by blanking every second line and expanding it over time to double the original line length in a line converter. On the receiving side, the lines are correspondingly compressed by a factor of 2, so that there is a uniform reduction in resolution both horizontally and vertically. This line reduction does not apply within the image window because both the vertical and the horizontal maximum resolution should be retained there. The latter is guaranteed in any case by the scanning method described.

4 illustrates this process. In the upper curve the camera signal is shown on the sending side A and below the curve on the receiving side B. In the middle, the converted signal is shown during transmission. It can be seen from this that the image signal and the horizontal blanking gaps are twice as large during the transmission, while the window size Fun has remained changed.

25th

30th

35

40

45

50

1 sheet of drawings

Claims (2)

Patent claims: 1. Method for the transmission of images over image channels with low bandwidth between a sending and a receiving point of a video telephone system in which the moving pictures recorded with a high level of detail, but cannot be transmitted in such a way characterized in that the image excerpt specified by the receiving body with higher detail and reduced motion resolution, but the rest of the picture with lower detail and higher motion resolution is transmitted. 2 Circuit arrangement for Druchführung the method according to claim 1, characterized in that the receiving unit (B) an image segment generator ^ is assigned, which stores the coordinates of the desired image (F) and on an image channel (a) for sending "Steep (A ) transmits and at the same time actuates a switch (S2, S3) which feeds the incoming image signals from outside the image section (F) directly and those from inside via an image memory (Sp) of the picture tube (B 2) , and that the sending point fine evaluator (C) , which records the transmitted coordinates and controls a switch (Si) within this image section, which adjusts the method of the image points to be transmitted