JPS6120189B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a predictive encoding/decoding transmission apparatus for image signals such as television signals. Image signals such as television signals have strong correlations in the horizontal and vertical directions of the screen and in the time axis direction, so instead of directly transmitting the image signal, a predicted signal is generated from pixel signals in the vicinity of the pixel signal to be transmitted. By transmitting the difference signal between the signal to be created and transmitted and the predicted signal, that is, the prediction error signal, the amount of transmitted information can be saved. This kind of encoding method is called predictive encoding, and includes intra-frame predictive encoding in which the predicted signal is obtained from pixel signals in the same frame, inter-field predictive encoding in which the predicted signal is obtained from the pixel signal in the previous field, and predictive coding in which the predicted signal is obtained from the pixel signal in the previous frame. Interframe predictive coding obtained from a signal, composite predictive coding that combines these, and the like are known. Intraframe coding has a high predictive accuracy for flat or blurred images, and is therefore efficient. Interframe coding is efficient for images with little motion, and interfield coding and composite coding have properties intermediate between these. Conventional predictive coding devices generally use one of these prediction methods; however, when the image signal to be transmitted has little or no movement, such as a broadcast television signal, In cases where there are flat images and complex images, if the prediction method is fixed, there is a drawback that efficient encoding cannot always be performed. On the other hand, a prediction switching method has also been proposed in which two types of prediction signals, intra-frame prediction and inter-frame prediction, are created, and which one is more efficient is checked in advance for each block consisting of multiple pixels, and the more efficient one is selected. However, when the speed of screen change is high, the efficiency of both of the above two prediction methods decreases, and the prediction switching method is no longer able to fully demonstrate its effects.

The purpose of the present invention is to compare the prediction efficiency of a plurality of prediction signals in units of blocks consisting of a plurality of pixels,
A predictive coding device that selects at least two predictive signals showing good efficiency for each block from a plurality of predictive signals, and further selects one highly efficient predictive signal for each pixel from among the predictive signals to perform predictive coding. and,
An object of the present invention is to provide a decoding device for this purpose.

According to the present invention, the first selection means obtains a plurality of predicted signals from a locally decoded signal and selects at least two predicted signals from the plurality of predicted signals according to a control signal; means for comparing the prediction efficiencies of the prediction signals and generating the control signal for each block according to the comparison result; and a second selection means for selecting one of the prediction signals output from the first selection means. , means for controlling the second selection means pixel by pixel, means for obtaining a difference (prediction error signal) between the output prediction signal of the second selection means and the image signal, and an output of the second selection means. a predictive encoding device comprising: means for obtaining the local signal from a prediction signal and the prediction error signal; and means for converting the code of at least the control signal and the prediction error signal; and the code-converted control signal and prediction. means for inversely converting the code of the error signal; a third means for similarly obtaining a plurality of prediction signals in the predictive encoding device from the decoded signal and selecting at least two predicted signals therefrom according to the control signal that has been inversely code-converted; a selection means, a fourth selection means for selecting one of the prediction signals output from the third selection means, a means for controlling the fourth selection means pixel by pixel, and an output of the fourth selection means. A predictive encoding/decoding transmission device comprising: a predictive signal; and means for obtaining the decoded signal from the predicted error signal subjected to code inverse transformation. can get.

If the predictive encoding/decoding transmission device of the present invention is used, information specifying a plurality of predictive signals selected by the first selection means is transmitted in units of blocks.
If the second selection means selects each pixel based on the magnitude of the transmitted prediction error amplitude, the decoding device can correctly decode the selection without transmitting the information instructing the selection.
The amount of information per pixel is greatly reduced.

Also, the contents of the screen (the movement of the subject, the size of the pattern,
When the screen changes (black and white, color, etc.), the first selection means selects a prediction signal that efficiently encodes that screen, so there is already a considerable improvement in encoding efficiency at this stage. For example, the prediction signal that minimizes the prediction error signal is selected pixel by pixel from among the prediction signals of Since the image signal can be reproduced, the required transmission speed can be lowered when passing through a transmission path, and the required capacity can be reduced when recording on a digital signal recording medium. Therefore, the effect of putting the present invention into practical use is extremely large.

Next, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing a simplified configuration of the present invention. The input image signal input to terminal 1 is encoded by encoding device 500 and provided to A, which is a transmission line or a digital recording medium. A decoding device 60 from the output of A
0, the original image signal is reproduced and output to terminal 1'.

Each of the encoding device and decoding device will be explained in detail below.

FIG. 2 is a block diagram showing an example of the configuration of a predictive encoding device according to the present invention. The image signal applied to the terminal 1 is converted into a digital signal 1011 by an analog/digital converter (A/D converter) 10. The image signal thus digitally converted is supplied to a delay circuit 11 that delays the time T required for the prediction efficiency comparison circuit 15 to perform the prediction efficiency comparison. The output signal of the delay circuit 11 is sent to the switch circuit 16.
The difference is calculated by the subtracter 12 by the predicted signal 1612 which is the output of the subtracter 12. This difference, that is, the prediction error signal, is supplied to the quantization circuit 13. Quantization circuit 13
In this method, the effect of compressing the amount of information is further enhanced by limiting the number of levels that the prediction error signal can take. This prediction error signal 1314 is summed with the prediction signal 1612 in the adder 17, and the local decoded signal 1718 is
and is supplied to the predictor group 18. In the predictor group 18, the predicted signal specified by the control signal 1514 is supplied to the switch circuit 16, which is a second selection means, as two selected predicted signals 1801 and 1802, respectively. In the switch circuit 16, selection is made between these two predicted signals, and the control is performed according to the control signal 2016. How to create the control signal 2016 will be described below. The prediction error signal 1314 is compared with a predetermined value in the comparison circuit 19, and the comparison result of whether or not it is larger than this threshold value (for example, if it is large, the code is 1, if it is not, the code is 0, etc.) is 1.
It is supplied to a delay element 20 which provides pixel delay.
The output of this delay element 20 is the switch circuit 1.
6 control signal 2016. For example, if the control signal 2016 indicates 1, the switch circuit selects a signal other than the predicted signal at the previous sample time, and if it indicates 0, it continues to select the predicted signal at the previous sample time.

Note that in order to ensure consistency between encoding and decoding devices, it is necessary to forcibly select the same predicted signal at least once at the start of operation. For example, the interframe prediction signal has a high accuracy rate in the synchronization signal portion within each horizontal scanning line. Therefore, if the start of a horizontal scanning line is forcibly subjected to inter-frame prediction, efficiency is improved, and at the same time consistency between encoding and decoding devices can be achieved.

Note that in the above description, the number of output prediction signals of the predictor group 18 is set to two for simplicity, but in principle, any number of output prediction signals may be provided.

Next, a specific configuration example of the predictor group 18 will be explained.

FIG. 3 shows an example of the configuration of the predictor group 18.

The predictor group 18 is roughly divided into an intra-frame predictor and an inter-frame predictor. For example, reference number 1
The predictor denoted by 80 is an intra-frame predictor, and a specific example is the previous value prediction, that is, 1
It is composed of pixel delay elements.

As for the interframe predictor, in order to efficiently encode a variety of screens, we will provide a predictor configuration that can perform motion prediction. An example of motion prediction is shown below. Interframe prediction using a one-frame delay element is accurate for a stationary screen, but becomes less accurate when the screen moves from side to side, for example. For example, if the screen of the current frame moves one pixel to the right with respect to the screen of the previous frame, if a signal delayed by (1 frame + 1) pixel is used as the predicted signal, the prediction will be accurate. A predictor capable of such motion prediction is 181 (1 frame-T-
3) Consists of a pixel delay element, a T pixel delay element 183, and a shift register 184. i.e. 1
If the number of pixels in a frame is F, T pixel delay elements 1
The predicted signal X F _-3 delayed by (F-3) pixels is obtained at the output of 83. If this signal is further delayed pixel by pixel by the shift register 184, each predicted signal from X _F-2 to X _F+3 is obtained as an output. The above example shows changes between frames (up to 3 pixels)
In general, in order to deal with cases exceeding 1
81 may be a (F-T-n) pixel delay element, and the delay element of the shift register 184 may be 2n. The predicted signal group X _F-3 to X _F+3 obtained in this way
is given to the switch circuit 185 which is the first selection means, and the switch circuit 185 receives the control signal 151.
4, an appropriate motion prediction signal is selected. (F
-T-3) The output of the pixel delay element 181 is 1 at the same time
A total of seven signals 1815 delayed from (F-T-3) to (F-T+3) pixels are obtained. Used for comparison.

Next, the predictive efficiency comparison circuit 15 shown in FIG. 2 will be explained.

FIG. 4 shows a specific example of the predictive efficiency comparison circuit 15.

Digitized image signal 1011 and prediction group 1
The seven types of prediction signals 1815 from interframe prediction signals X _FT-3 to X _F-T+3 in 8 are subtracted in the subtracter 150, and the difference is calculated from X _FT-3 to X _F-T+3. The signal is applied to the threshold circuit 151 by each of the corresponding channels Ch.1 to Ch.7. Threshold circuit 15
1, it is determined whether the difference between these is greater than a certain threshold value, and if the difference is greater than a certain threshold, then a value of 1 is generated, for example, and if not, a value of 0 is generated, and this value is provided to the counter 152 through each channel. The counter 152 counts (adds) this numerical value for each channel, and the counting result is given to the comparator 153. Comparator 15
3, it is compared and determined which channel number shows the smallest count result among the count results from Ch.1 to Ch.7 in block units, and the channel number is outputted as a switching signal 1514. Based on this switching signal 1514, a switch circuit 185 that selects the output prediction signal of the predictor group 18 is controlled. Both the prediction error signal and the control signal of the predictor group 18 are transmitted to the receiving side, but upon transmission, the code converter 14 converts them into codes for efficient transmission. As an example, an unequal length code such as a Huffman code obtained from the distribution of each signal can be applied. The code converter 14 further arranges the code-converted control signal and prediction error signal in a predetermined order,
After performing speed matching with the transmission line, the code is output to the output terminal 7. Furthermore, when recording on a digital signal recording medium such as a disk or drum instead of a transmission path, the code is outputted to the output terminal 7 after matching the writing speed of the recording medium.

Next, the decoding device will be explained.

FIG. 5 is a diagram illustrating an example of the configuration of a decoding device.

During transmission or from a digital signal recording medium, the code is applied to the code inverter 14' through the input terminal 7', and by an operation opposite to that of the code converter 14 in the encoder, a control signal 151 in the encoder is generated.
1418′ corresponding to 4 is the prediction error signal 1314
is separated into a prediction error signal 1440' corresponding to the prediction error signal 1440'. The prediction error signal 1440' is the prediction signal 164
0' and the decoded signal 40 which is added in the adder 40'.
It becomes 30'. The decoded signal 4030' is converted into an original analog image signal by a digital/analog (D/A) converter 30' and outputted from a terminal 1'. At the same time, it is also given to the predictor group 18', and the predicted signal 18
01', 1802'.

As shown in FIG. 6, the predictor group 18' has the same function as the predictive encoder group 18 in the encoding device except that the shift register 182 and prediction signal 1815 are removed. Therefore, one of the seven prediction signals X _F-3 to X _F+3 is selected by the switch circuit 185', which is the third selection means, in accordance with the switching signal 1418', and as a result, 2
As in the encoding device, the two prediction signals 1801' and 1802' are applied to the switch circuit 16', which is the fourth selection means. Next, how to create the control signal 2016' will be described. The separated prediction error signal 1440' is compared with the same value as the encoding device in the comparator circuit 1', and is converted into a code 1 or 0 by the same logic as the encoding device. The control signal 2016' is obtained by delaying this code by one pixel in the one-pixel delay element 20'. The image signal encoded in this manner is converted into the original image signal by a decoding device. In addition,
For vertical and diagonal movements within the screen, both the encoding and decoding devices can use multiple delay elements for one horizontal scanning line to further increase the predicted signal, making it even more efficient. It is.

FIG. 7 is a diagram illustrating a case where the control signals 2016, 2016' for the second selection means are generated using another method.

Since the changes are the same for both the encoding device and the decoding device, the encoding device will be explained as an example. X ₁ -X 0 is a subtracter 31, and X ₂ -X ₀ is a subtracter, which takes the difference by a locally decoded signal 1718 (abbreviated as X ₀ ) from predicted signals 1801 and ₁₈₀₂ (abbreviated as X ₁ and X ₂ , respectively). 32, respectively. X ₁ −X ₀ and X ₂ −
The absolute value of X ₀ is taken by absolute value circuits 33 and 34, respectively, and compared by a comparator 35. ｜x ₁ −x ₀ ｜
If _{> |} The switch circuit 16 outputs X 2 when the control signal 2016 has a code of 0, and outputs X ₁ when the control signal 2016 has a code of ₁ .
shall be able to select.

That is, in this way, the control signals 2016, 2
016', the circuit shown in FIG. 7 may be inserted in place of the comparator 19 of FIG. 2 and the comparator 19' of FIG. 5. Note that the local decoded signal 1718 is converted into a decoded signal 1440' in the decoding device.
shall be read as Such control signal 2
According to the method of making 016 and 2016', the control signal is made more finely than in the case of the previously described embodiment, so that the prediction efficiency is further improved.

FIG. 8 is a diagram illustrating an example of simplifying the encoding/decoding apparatus of the present invention.

In the previous embodiments, the prediction efficiencies of a plurality of prediction signals were compared in advance for each block, but in this embodiment, the compared results are used in the next block. The circuit configuration can be greatly simplified. That is, in the encoding device, the shift registers 182 and 184 in the predictor group 18 shown in FIG. 3 are made common (F
-T-3) Pixel delay element 181 and T pixel delay element 183 are integrated (F-3) Pixel delay element 18
Set it to 6. An example of this modification is shown in FIG. At the same time as above, the T pixel delay circuit 11 in FIG. 2 is removed.

No changes are required in the decoding device.

In this way, by using the comparison result of the prediction efficiency in block units for the selection of the prediction signal in the first selection means for the image signal in the block at the next time, the encoding apparatus can be simplified.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a configuration of a predictive coding/decoding transmission device of the present invention, FIG. 2 is a block diagram showing a configuration example of a predictive coding device of the present invention, and FIG. 3 is a block diagram of a predictive coding device of the present invention. FIG. 4 is a block diagram showing an example of the configuration of the predictor group 18, FIG. 4 is a block diagram showing an example of the configuration of the prediction efficiency comparison circuit 15 of the predictive encoding device, and FIG. 5 is a block diagram showing an example of the configuration of the decoding device in the present invention. Figure 6 shows predictor group 1 of the decoding device.
8' is a block diagram showing an example of the configuration of the second selection means.
The figure is a block diagram showing another example of the configuration of the predictor group 18 of the predictive encoding device. In the figure, A is a transmission line or a recording medium for digital signals, 10 is an analog/digital (A/
D) Converter, 11...T pixel delay element, 12...
Subtractor, 1... Input terminal, 13... Quantizer, 1
4... Code converter, 15... Prediction efficiency comparison circuit,
16...Switch circuit, 17...Adder, 18...
...Predictor group, 19...Comparison circuit, 20...1 pixel delay element, 7...Output terminal, 180...Intraframe predictor (1 pixel delay element), 181... (F-
T-3) Pixel delay element, 182, 184...Shift register, 183...T pixel delay element, 185
...Switch circuit, 150...Subtractor, 151...
...Threshold circuit, 152...Counter, 153...Comparator, 7'...Input terminal, 14'...Sign inverse converter,
40'... Adder, 16'... Switch circuit, 1
8'...Predictor group, 19'...Comparison circuit, 20'...
...1 pixel delay element, 30'...Digital/analog (D/A) converter, 1'...Output terminal, 18
0'...Intraframe predictor (1 pixel delay element), 1
86'...(F-3) Pixel delay element, 184'...
Shift register, 185'...Switch circuit, 3
1, 32... subtractor, 33, 34... absolute value circuit, 35... comparator, 186... (F-3) pixel delay element.

Claims (1)

[Claims] 1. When transmitting an image signal, a first selection means obtains a plurality of predicted signals from a locally decoded signal and selects at least two predicted signals from the plurality of predicted signals according to a control signal; means for comparing the prediction efficiencies of the predicted signals and generating the control signal for each block according to the comparison result; and a second selection means for selecting one of the predicted signals output from the first selection means. and means for controlling the second selection means pixel by pixel, means for obtaining a difference (prediction error signal) between the output prediction signal of the second selection means and the image signal, and the second selection means. a predictive encoding device comprising: means for obtaining the local signal from the output prediction signal and the image signal; and means for transcoding at least the control signal and the prediction error signal; and the transcoding of the control signal and prediction. a third means for inversely converting the code of the error signal, obtaining a plurality of prediction signals from the decoded signal in the same manner as in the predictive encoding device, and selecting at least two prediction signals from the above according to the control signal that has been inversely code-converted; a fourth selection means for selecting one of the prediction signals output from the third selection means; a means for controlling the fourth selection means pixel by pixel; A predictive decoding device comprising means for obtaining the decoded signal from the output prediction signal and the code-inversely transformed prediction error signal;
A predictive encoding device/decoding/transmission device comprising: