JPH0551236B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to an encoding method and apparatus for highly efficient compression encoding of moving image signals.

(Prior Art) Conventionally, IEEE Transactions on Communications (IEEE Transactions on Communications) has been used as a video signal encoding method that combines orthogonal transform encoding and interframe predictive encoding.
Communications) Magazine, Vol.COM−33, pp.1291
-1302 (December 1985) (Reference 1) is known. The configuration of the encoding method described in this document 1 is shown in FIG. That is, on the encoding side, the orthogonal transform coefficients are inter-frame encoded, and on the decoding side, the orthogonal transform coefficients are inter-frame decoded and inverse orthogonally transformed to obtain a decoded image. When encoding the inter-frame difference value of orthogonal transform, all orthogonal transform coefficients are quantized with the same quantization characteristic, and the amount of generated information is controlled by changing the step size of the quantization characteristic and the size of the dead zone. I'm going. In addition, the 2nd International Technical Symposium on Optical and Electro-Optical Applied Science and...
Engineering (2nd International
Technical Symposium on Optical and
Electro-Optical Applied Science and
Enginering）SPIE Conf.B594Image Coding
(December 1985) (Reference 2) describes a coding method for orthogonal transformation coding of an interframe difference signal. The configuration of this encoding system is shown in FIGS. 7a and 7b. This method of encoding orthogonal transform coefficients can be directly applied to encoding inter-frame difference values of orthogonal transform coefficients in Document 1. In the encoding method of orthogonal transform coefficients in Reference 2, the orthogonal transform coefficients are first divided into several sets, the variance is measured, and the orthogonal transform coefficients with small variance are discarded, thereby reducing the deterioration of the image quality of the decoded image signal. While suppressing as much as possible,
It is devised so that encoding is performed at a constant bit rate.

(Problems to be Solved by the Invention) The amount of information can be compressed by cutting down the orthogonal transform coefficients to be encoded to reduce the number.
At this time, in the conventional method, for coefficients encoded with zero interframe prediction error due to truncation, on the decoding side, the coefficients of the previous frame are decoded as they are as orthogonal transform coefficients of the current frame. For example, in the block diagram of FIG. 6, a case as shown in FIG. 8b may be considered.
Figure 8 is an example of using one-dimensional orthogonal transformation, where the two waveforms in the squares represent the waveforms corresponding to two of the orthogonal transformation coefficients, and the amplitude of the waveform is the same as the coefficient. represents size. Note that the DC component is not shown. In FIG. 8b, a locally decoded signal such as waveform 508 is obtained during encoding of the previous frame, and orthogonal transform coefficients corresponding to waveform 506 and waveform 507, which are components of waveform 508, are sent to the decoding side, Assume that waveform 517 is obtained as a decoded signal. When using a method of simply calculating the difference from the signal of the previous frame as interframe encoding, on the encoding side, the waveform 506 is used as the predicted signal of the current frame output from the predictor 91 in FIG.
The orthogonal transform coefficients corresponding to the waveform 507 are used, and on the decoding side, the waveform 515 and waveform 51, which are the components of the waveform 517, are used as the output of the predictor 92 in FIG.
The orthogonal transform coefficient corresponding to 6 is used as the interframe prediction signal. In the current frame, waveform 5
When encoding an input signal like 01, waveform 5
An inter-frame difference is taken between orthogonal transform coefficients corresponding to waveform 502 and waveform 503 and waveform 506 and waveform 507, which are components of 01. By this inter-frame difference operation, waveform 5
04 and waveform 505 are obtained. In the coefficient truncation circuit 5, the waveform 504
The orthogonal transform coefficients corresponding to are saved and the waveform 505
When the orthogonal transform coefficients corresponding to the waveforms 511 and 512 are truncated, the coefficients corresponding to the waveforms 511 and 512 are encoded. In the coefficient truncation circuit 5, the mark ◯ indicates that the coefficient is saved, and the mark x indicates that the coefficient is truncated. On the decoding side, the orthogonal transform coefficients of the current frame corresponding to waveforms 513 and 514 are outputted from the predictor 92 in FIG. 6, and the orthogonal transform coefficients of the current frame are decoded. be converted into The orthogonal transform coefficients corresponding to the waveform 514 of the waveform 513 are inversely orthogonally transformed to obtain the waveform 52.
3 decoded signals are obtained. The orthogonal transform coefficients used to predict the next frame signal are the orthogonal transform coefficients corresponding to waveform 509 and waveform 510 on the encoding side, and the orthogonal transform coefficients corresponding to waveform 518 and waveform 519 on the decoding side.
are orthogonal transform coefficients corresponding to , and are input to predictors 91 and 92 in FIG. 6, respectively. Decoded waveform 5
23 is compared with the input waveform 501 and the previous frame decoded waveform 517. Due to the effect of coefficient truncation, the position of the falling slope of the decoded waveform 523 is shifted to the right compared to the input waveform 501, and it is different from the previous frame. It has a shape similar to waveform 517. In this manner, in the conventional method, the signal waveform of the previous frame corresponding to the truncated coefficient remains as it is in the decoded signal of the current frame, resulting in significant image quality deterioration. Particularly for areas with movement, if you cut off the high-frequency coefficients of the frame difference,
The high-frequency component of the object before it moves, that is, the limbus, remains in the background.

The purpose of the present invention is to eliminate these drawbacks of the conventional method, prevent unnecessary effects from the previous frame signal from remaining on the decoded image signal even if orthogonal transform coefficients are truncated during encoding, and improve image quality. An object of the present invention is to provide a video signal encoding device with little deterioration.

(Means for Solving the Problems) The encoding method of the present invention performs orthogonal transformation on an input image signal, and transform coefficients of the current frame predicted from the orthogonal transform coefficients and the orthogonal transform coefficients of the previous frame, that is, the predicted orthogonal and the difference signal is used to locally decode the orthogonal transform coefficient of the current frame to predict the orthogonal transform coefficient of the next frame by calculating the difference between the difference signal and the predicted transform coefficient, and calculate the sum of this difference signal and the predicted transform coefficient. In a video signal encoding method that outputs a coded image signal as a coded image signal, a part of the coefficients is truncated from a difference signal between an orthogonal transform coefficient obtained by orthogonally transforming an input image signal and a predicted orthogonal transform coefficient, and the predicted orthogonal transform coefficient is The present invention is characterized in that coefficients related to the truncated coefficients in the difference signal are truncated with respect to the orthogonal transform coefficients.

The configuration of the first encoding device that realizes the present invention is as follows:
means for orthogonally transforming an input image signal; means for taking a difference between the orthogonal transform coefficients obtained by the orthogonal transform and an output signal of an interframe predictor; and truncating some coefficients of the output signal of the means for taking the difference. a first truncation means; a second truncation means for truncating coefficients related to the coefficients truncated in the first truncation means from the output signal of the interframe predictor; and an output signal of the first truncation means; means for calculating the sum of the output signal of the second truncation means; an inter-frame predictor for predicting the signal of the next frame from the output signal of the summation means; The present invention is characterized by comprising means for outputting an encoded image signal, and means for generating mode information instructing which coefficients are to be discarded by the first discarding means.

Further, the configuration of a second encoding device realizing the present invention includes means for orthogonally transforming an input image signal, and means for calculating the difference between the orthogonal transform coefficients obtained by the orthogonal transform and the output signal of the interframe predictor. , first truncation means for truncating some coefficients of the output signal of the difference taking means, and means for calculating the sum of the output signal of the first truncation means and the output signal of the interframe predictor. and a second truncating means for truncating coefficients related to the coefficients truncated by the first truncating means from the output signal of the summing means.
an interframe predictor that predicts and outputs a signal of the next frame from the output signal of the second truncation means; and means that outputs the output signal of the first truncation means as an encoded image signal. , and means for generating mode information for instructing which coefficient is to be truncated in the first truncation means.

Further, the configuration of a third encoding device realizing the present invention includes means for orthogonally transforming an input image signal, a first truncation means for truncating some coefficients of the orthogonal transform coefficients obtained by the orthogonal transform, Said first
means for taking the difference between the output signal of the truncation means and the output signal of the interframe predictor; means for taking the sum of the output signal of the difference taking means and the output signal of the interframe predictor; a second truncation means for truncating a coefficient related to the coefficient truncated by the first truncation means from the output signal of the truncation means; and a second truncation means for predicting and outputting a signal of the next frame from the output signal of the second truncation means. a means for outputting an output signal of the means for taking a difference as an encoded image signal;
and means for generating mode information for instructing which coefficient to truncate in the truncation means.

Further, the configuration of a fourth encoding device that realizes the present invention includes means for performing orthogonal transformation on an input image signal, and means for calculating the difference between the orthogonal exchange coefficient obtained by the orthogonal transformation and the output signal of the interframe predictor. a first truncation means for cutting off some coefficients from the output signal of the difference taking means; a means for calculating the sum of the output signal of the difference taking means and the output signal of the interframe predictor; a second truncation means for truncating coefficients related to the coefficients truncated by the first truncation means from the output signal of the summation means; and a second truncation means for predicting the next frame signal from the output signal of the second truncation means. means for outputting the output signal of the first truncation means as an encoded image signal; and means for generating mode information for instructing which coefficient to truncate in the first truncation means. It is characterized by consisting of.

Further, the configuration of a fifth encoding device that realizes the present invention includes means for orthogonally transforming an input image signal, a first truncation means for truncating some coefficients of the orthogonal coefficients obtained by the orthogonal transformation, and the means for taking the difference between the output signal of the first truncation means and the output signal of the interframe predictor; and truncating coefficients related to the coefficients truncated by the first truncation means from the output signal of the difference taking means. Second
, means for summing the output signal of the difference-taking means and the output signal of the inter-frame predictor, and inter-frame prediction for predicting the signal of the next frame from the output signal of the sum-taking means. a means for outputting the output signal of the second truncation means as an encoded image signal, and a means for generating mode information for instructing which coefficient to truncate in the first truncation means. shall be.

(Operation) In FIG. 8b, the output signal 52 of the decoding device
3 has a shape similar to the decoded signal 517 of the previous frame because the orthogonal transform coefficient corresponding to waveform 516, which is the orthogonal transform coefficient of the previous frame, was not modified in the interframe decoding of the current frame. It was the cause. Such image quality deterioration can be alleviated by discarding the exchange coefficients discarded on the encoding side when performing interframe decoding in the decoding device as well. In this case, the example shown in FIG. 8b becomes as shown in FIG. 8a. FIG. 8a shows the encoding device according to claim 2 of the present invention and the fifth
This is an encoding/decoding device that combines the decoding devices shown in the figure. In FIG. 5, the input terminal 30
The adder 11 calculates the sum with the output signal of the interframe predictor 15 for the encoded image signal input from 0, and the coefficient truncation circuit 14 calculates the sum of the output signal of the interframe predictor 15
Part of the coefficient is truncated based on the mode information input from 0. The output signal of the coefficient truncation circuit 14 is subjected to inverse orthogonal transformation in the inverse orthogonal transformation circuit 13, and then outputted from the terminal 400 as a decoded image signal. The interframe predictor 15 predicts the next frame signal from the output signal of the coefficient truncation circuit 14 and outputs the predicted signal. The difference between Figures 8a and 8b is that in Figure 8a, the orthogonal transform coefficients corresponding to waveforms 513 and 514, which are interframe decoded signals of the current frame, are decoded; The transform coefficient corresponding to the truncated waveform 505 is truncated by the coefficient truncation circuit 14 to obtain orthogonal transform coefficients corresponding to the waveforms 520 and 521, and then inverse orthogonal transform is performed to obtain the output signal waveform 522. On the encoding side, when locally decoding the orthogonal transform coefficients of the current frame, the orthogonal transform coefficients corresponding to the truncated waveform 505 are truncated by the coefficient truncation circuit 6, and the waveforms 530 and 531 are
The orthogonal transform coefficients corresponding to are obtained. Waveform 522
Compared to the waveform 523, the position where the waveform falls and crosses level zero is closer to the input waveform 501, and the difference in level from the input waveform 501 is smaller. This tendency becomes more pronounced when the difference between the value of the coefficient to be truncated in the current frame and the value of the corresponding coefficient in the previous frame is large. As described above, in the present invention, deterioration of a decoded image signal can be suppressed compared to the conventional method in which coefficients are not truncated on the decoding side during encoding. Here, the difference between the first and second configurations of the encoding device is that in the first configuration, when locally decoding the orthogonal transform coefficients of the current frame, the orthogonal transform coefficients of the previous frame are discarded. However, in the second configuration, the coefficients are truncated after the orthogonal transform coefficients of the current frame are locally decoded, or the orthogonal transform coefficients of the current frame obtained by interframe prediction are truncated. . In the first and second configurations, the same encoded image signal can be obtained for the same input image signal. The difference between the second, third, and fourth configurations of the encoding device is that in the second configuration, the truncation of the orthogonal transform coefficients, which was performed on the interframe prediction error signal of the orthogonal transform coefficients, is performed in the third configuration. In the above configuration, this is performed on the signal before taking the interframe prediction error, and in the fourth configuration, it is performed after the interframe prediction loop. 2nd, 3rd, 4th
With this configuration, the same encoded image signal can be obtained for the same input image signal. The fifth part of the encoding device
Compared to the third configuration, this configuration eliminates the truncation of orthogonal transform coefficients that was performed for the locally decoded signal or the interframe predicted signal in the third configuration, and eliminates the interframe loop. I'm going after. Considering the point at which a certain orthogonal transform coefficient is truncated, in the case of the third configuration, that orthogonal transform coefficient is also truncated as a coded image signal, and the orthogonal transform of the current frame that is the basis of interframe prediction The coefficients are also truncated, but in the case of the fifth configuration, the orthogonal transform coefficients are truncated as a coded image signal, but the orthogonal transform coefficients of the current frame, which is the basis of interframe prediction, are truncated. It has not been done. However, if the orthogonal transform coefficient is truncated in the encoding of the next frame, the orthogonal transform coefficient will be canceled in the interframe local decoding and will enter the interframe predictor, and the interframe prediction loop Since the coefficients are truncated by the coefficient truncation means at the point where they exit, the coefficients are not sent to the decoding side, and it is possible to obtain the same encoded image signal as in the encoding apparatus of the third configuration.

(Embodiment) FIGS. 1a to 1e are block diagrams showing the configuration of an encoding device according to the present invention.

In FIG. 1A, an image signal inputted from an input terminal 100 is orthogonally transformed by an orthogonal transform circuit 1, and the difference between the image signal and the output signal of an interframe predictor 2 is calculated by a subtractor circuit 3. For this difference signal, the coefficient truncation circuit 5 truncates a part of the coefficient based on the mode information generated by the truncation coefficient determination circuit 81. The output signal of the coefficient truncation circuit 5 is outputted from an output terminal 200 as an encoded image signal. The truncation coefficient determination circuit 81 generates mode information instructing which coefficient to truncate.
Mode information is output from output terminal 210. The coefficient truncation circuit 6 truncates coefficients related to the coefficients truncated by the coefficient truncation circuit 5 from the output signal of the interframe predictor 2. Adder 4 adds the output signals of coefficient truncation circuits 5 and 6. The interframe predictor 2 predicts the next frame signal from the output signal of the adder 4 and outputs the predicted signal.

In FIG. 1b, the image signal inputted from the input terminal 100 is orthogonally transformed by the orthogonal transform circuit 1, and the difference from the output signal of the circuit 7 is calculated by the subtracter 3. For this difference signal, the coefficient truncation circuit 5 truncates a part of the coefficient based on the mode information generated by the truncation coefficient determination circuit 81. The output signal of the coefficient truncation circuit 5 is outputted from an output terminal 200 as an encoded image signal. The truncation coefficient determination circuit 81 generates mode information instructing which coefficient to truncate. Mode information is output from output terminal 210. In adder 4,
The output signal of the coefficient truncation circuit 5 and the output signal of the circuit 7 are summed. The circuit 7 cuts off the coefficients related to the coefficients cut off by the coefficient cutting circuit 5 for the output signal of the adder 4 and predicts and outputs the signal of the next frame from this signal, or outputs the signal of the next frame. A signal for the next frame is predicted from the signal, and a coefficient truncating circuit 5 truncates coefficients related to the truncated coefficients from this signal, and then outputs the signal.

In FIG. 1c, the image signal input from the input terminal 100 is orthogonally transformed in the orthogonal transform circuit 1, and then in the coefficient truncation circuit 5, a part of the coefficients is determined based on the mode information generated by the truncation coefficient determination circuit 81. is truncated. The difference between the output signal of the coefficient truncation circuit 5 and the output signal of the circuit 7 is obtained in the subtracter 3. The output signal of the subtracter 3 is outputted from an output terminal 200 as an encoded image signal. In the truncation coefficient determination circuit 81,
Generate mode information that indicates which coefficients to truncate. Mode information is output from output terminal 210. In the adder 4, the output signal of the subtracter 3 and the output signal of the circuit 7 are summed. The circuit 7 cuts off the coefficient related to the coefficient cut off by the coefficient cutting circuit 5 from the output signal of the adder 4 and predicts and outputs the signal of the next frame from this signal, or outputs the signal of the next frame. The signal of the next frame is predicted from the signal, and the coefficients related to the truncated coefficients are truncated by the coefficient truncation circuit 5 and then outputted.

In FIG. 1d, the image signal input from the input terminal 100 is orthogonally transformed by the orthogonal transform circuit 1.
The subtracter 3 calculates the difference between the signal and the output signal of the circuit 7. For this difference signal, the coefficient truncation circuit 5 truncates a part of the coefficient based on the mode information generated by the truncation coefficient determination circuit 81. The output signal of the coefficient truncation circuit 5 is outputted from an output terminal 200 as an encoded image signal. The truncation coefficient determination circuit 81 generates mode information instructing which coefficient to truncate. The mode information is
It is output from the output terminal 210. In adder 4,
The output signal of the subtracter 3 and the output signal of the circuit 7 are summed. The circuit 7 cuts off coefficients related to the coefficients cut off by the coefficient cutting circuit 5 from the output signal of the adder 4, predicts and outputs the next frame signal from this signal, or outputs the predicted signal from the adder 4.
The signal of the next frame is predicted from the output signal of , and the coefficients related to the truncated coefficients are truncated from this signal by the coefficient truncation circuit 5 and then output.

In FIG. 1e, the image signal input from the input terminal 100 is orthogonally transformed in the orthogonal transform circuit 1, and then the coefficients are truncated in the coefficient truncation circuit 5 based on the mode information generated by the truncation coefficient determination circuit 81. part is truncated. The output signal of the coefficient truncation circuit 5 is sent to the subtracter 3 by
The difference with the output signal of the interframe predictor 2 is taken. The output signal of the subtracter 3 is sent to a coefficient truncation circuit 6.
In this step, the coefficients related to the truncated coefficients are truncated by the coefficient truncation circuit 5. The output signal of the coefficient truncation circuit 6 is outputted from an output terminal 200 as an encoded image signal. The truncation coefficient determination circuit 81 generates mode information instructing which coefficient to truncate. The mode information is output from the output terminal 210. The adder 4 sums the output signal of the subtracter 3 and the output signal of the interframe predictor 2. The interframe predictor 2 predicts and outputs the signal of the next frame from the output signal of the adder 4.

In each of the encoding devices FIGS. 1a-e, two coefficient truncation circuits are included. The coefficients to be truncated in these two coefficient truncation circuits may be the same or different.

The difference between a and b in FIG. 1 is that in a, when the orthogonal transform coefficients of the current frame are locally decoded by the adder 4, the orthogonal transform coefficients of the previous frame are discarded by the coefficient truncation circuit 6, but in b. The orthogonal transform coefficients of the current frame are locally decoded by an adder 4 and then discarded by a circuit 7. In a and b, the same encoded image signal can be obtained for the same input image signal. The difference between Figure 1 b, c, and d is that in b, the orthogonal transform coefficients are truncated on the interframe prediction error signal of the orthogonal transform coefficients output from subtractor 3, and in c, the subtractor 3 is performed on the signal before taking the interframe prediction error, and step d is performed after the interframe prediction loop after branching the signal to the adder 4. b, c, and d can obtain the same encoded image signal for the same input image signal. In FIG. 1e, when compared with FIG. 1c, the truncation of orthogonal transform coefficients that was performed by the circuit 7 in c is no longer performed, but is performed by the coefficient truncation circuit 6 after the interframe prediction loop. Alternatively, considering the point at which the orthogonal transform coefficient is truncated, in the case of c, the orthogonal transform coefficient is also truncated as a coded image signal, and in the orthogonal transform coefficient of the current frame that is the basis of interframe prediction. In the case of e, the orthogonal transform coefficients of the encoded image signal are truncated, but the orthogonal transform coefficients of the current frame, which is the basis of interframe prediction, are not truncated. but,
If the orthogonal transform coefficient is also truncated in the encoding of the next frame, the orthogonal transform coefficient will be canceled in the interframe local decoding and enter the interframe prediction circuit 2, and the interframe prediction After exiting the loop, the coefficients are truncated by the coefficient truncation circuit 6, so they are not sent to the decoding side, and the same encoded image signal as in the encoding device of FIG. 1c can be obtained.

The mode information outputted from the encoding device through the terminal 210 and inputted into the decoding device through the terminal 310 includes information that directly instructs which coefficients are to be truncated, as well as information in which several patterns of coefficients to be truncated are prepared in advance. The pattern number can be used as mode information. The mode information is
It does not have to be dedicated information that instructs coefficient truncation, and may be a signal that can be derived from other encoded information.

FIG. 2 is a block diagram showing an embodiment of an encoding/decoding apparatus using the encoding apparatus of FIG. 1a and the decoding apparatus of FIG. 5 of the present invention. In FIG. 2, a quantizer and a multiplexer are added to the embodiment of the encoding device of FIG. Figure 2a is a block diagram of the encoding device, where 100 is an input terminal, 1 is an orthogonal transformer, 3 is a subtracter, 5 and 6 are coefficient truncation circuits, 53 is a quantizer, and 71 is an inverse quantizer. , 4 is an adder, 2 is an inter-frame predictor, 81 is a truncation coefficient determination circuit, and 60 is a coded image signal output from the quantizer 53, information on the quantization characteristics in the quantizer 53, and a truncation coefficient determination circuit 81. This is a multiplexer for outputting the mode information output from the output terminal 290 to the output terminal 290. FIG. 2b is a block diagram of the decoding device, where 390 is an input terminal, 61 is a demultiplexer that separates the signal from the input terminal 390 into a coded image signal, quantization information, and mode information, and 72 is an inverse quantizer. 11 is an adder; 12 is a truncated part of the output signal of the adder 11 based on the mode information, and predicts and outputs the signal of the next frame from that signal; or 13 is an inverse orthogonal transformer; 13 is an inverse orthogonal transformer; 13 is an inverse orthogonal transformer;
The output signal of the inverse orthogonal transformer 13 is outputted from an output terminal 400 as a decoded image signal. Quantizer 53 and inverse quantizer 7 in the encoding device of FIG. 2a
1 and multiplexer 60 can also be incorporated in the other encoding devices of FIG. 1, FIGS. 1b-e.

An example of the interframe prediction function included in the interframe predictor 2 or the circuit 7 can be realized by simply delaying the input signal until the next frame and outputting the delayed signal. At this time, the interframe predictor 15 on the decoding side or the interframe prediction function included in the circuit 12 simply delays the input signal until the next frame and outputs it. FIG. 3a shows an embodiment in which a motion compensation technique is introduced as the interframe predictor 2 in the encoding apparatus of FIG. 1a. The output signal of the adder 4 is subjected to inverse orthogonal transformation by an inverse orthogonal transformer 31. In the circuit 33,
The motion detection circuit 3 is connected to the output signal of the inverse orthogonal transformer 31.
The image is corrected by the amount of movement of the image detected in step 4, and output as a signal for the next frame. The output signal of the circuit 33 is orthogonally transformed by the orthogonal transformer 32 . The output signal of this orthogonal transformer 32 becomes the output image signal of the interframe predictor 2. The motion detection circuit 34 calculates the amount of image movement from the input image signal and outputs it to the circuit 32 and the multiplexer 60 as motion information. The multiplexer 60 multiplexes the motion information with other information and outputs the result. Figure 3b
shows an embodiment in which a motion compensation technique is introduced as the interframe predictor 15 in the decoding device shown in FIG. The output signal of the coefficient truncation circuit 14 is subjected to inverse orthogonal transformation by an inverse orthogonal transformer 41 . In the circuit 43, based on the motion information output from the demultiplexer 61, the inverse orthogonal transform circuit 4
1 output signal is corrected and output as a predicted signal for the next frame. The output signal of the circuit 43 is orthogonally transformed by the orthogonal transformer 42 . This orthogonal transformer 42
The output signal becomes the output signal of the interframe predictor 15. Motion compensation can also be applied to the circuits 7 and 12 that perform interframe prediction and coefficient truncation. 4a and 4b show an embodiment in which motion compensation is applied to the circuit 7. FIG. In Figure 4 a and b, terminal 5
The output signal of the adder 4 is input from 11, the mode information output from the truncation coefficient determination circuit 81 is input from the terminal 513, the input image signal is input from the terminal 514, and the signal processed by the circuit 7 is input from the terminal 512. The image signal is outputted to the terminal 515.
The motion information is output. In FIGS. 4a and 4b, a circuit 30 includes an inverse orthogonal transformer 31, a circuit 33 that combines an orthogonal transformer 32, a frame memory, and a variable delay circuit, and a motion detection circuit 34.
The operation is the same as that of circuit 2 in FIG. 3a.
In FIG. 4a, the signal inputted from the input terminal 511 is processed by the coefficient truncation circuit 35, and then processed by the circuit 30 and outputted to the output terminal 512. In FIG. 4b, the signal inputted from the input terminal 511 is The signal is processed by the circuit 30, then processed by the coefficient truncation circuit 36, and outputted to the output terminal 512.

(Effects of the Invention) In the conventional method, the signal waveform due to the transform coefficient of the previous frame may remain in the decoded signal of the current frame, resulting in a large deterioration of image quality. According to the present invention, such a phenomenon can be suppressed, and image quality deterioration can be reduced.

[Brief explanation of the drawing]

1A to 1E are block diagrams of the encoding device of the present invention, FIGS. 2A and 2B and 3A and 3B are block diagrams showing embodiments of the encoding device of the present invention, and FIGS. b
6 is a block diagram illustrating an embodiment of a circuit for performing coefficient truncation and motion-compensated interframe prediction according to the present invention; FIG. 5 is a block diagram illustrating an example of a decoding device; and FIG.
7A and 7B are explanatory diagrams of the conventional system, and FIGS. 8A and 8B are diagrams illustrating the operation of the present invention. In the figure, 1 is an orthogonal transform circuit, 2 is an interframe predictor, 3 is a subtracter, 4 is an adder, 5 and 6 are coefficient truncation circuits, 7 is a circuit that performs coefficient truncation and interframe prediction, and 81 is a truncation coefficient. This is a judgment circuit.

Claims (1)

[Scope of Claims] 1. Perform orthogonal transformation on an input image signal, calculate the difference between the orthogonal transform coefficients and the transform coefficients of the current frame predicted from the orthogonal transform coefficients of the previous frame, that is, the predicted orthogonal transform coefficients, and calculate the difference signal. and the predicted transform coefficient, locally decodes the orthogonal transform coefficient of the current frame, and uses the sum to predict the orthogonal transform coefficient of the next frame, and outputs the difference signal as a coded image signal. In the encoding method, a part of the coefficients is truncated from a difference signal between an orthogonal transform coefficient obtained by orthogonally transforming an input image signal and a predicted orthogonal transform coefficient, and the predicted orthogonal transform coefficient is truncated in the difference signal. A method for encoding a moving image signal, the method comprising: truncating coefficients related to the calculated coefficients. 2 means for orthogonally transforming an input image signal, means for taking the difference between the orthogonal transform coefficient obtained by the orthogonal transform coefficient and the output signal of the interframe predictor, and a part of the coefficient of the output signal of the means for taking the difference. a first truncation means for truncating from the output signal of the interframe predictor a coefficient related to the coefficient truncated by the first truncation means; and an output of the first truncation means. means for summing the signal and the output signal of the second truncation means; an interframe predictor for predicting a signal of the next frame from the output signal of the summing means; and an output of the first truncation means. A moving image signal and encoding device comprising: means for outputting a signal as an encoded image signal; and means for generating mode information for instructing which coefficients are to be truncated in the first truncation means. 3 means for orthogonally transforming an input image signal; means for taking the difference between the orthogonal transform coefficients obtained by the orthogonal transform and the output signal of the interframe predictor; and a part of the output signal of the means for taking the difference. a first truncation means for truncating a coefficient; a means for summing an output signal of the first truncation means and an output signal of the interframe predictor; and a first truncation from the output signal of the summation means. a second truncation means for truncating coefficients related to the coefficients truncated in the means; an interframe predictor for predicting and outputting a signal of the next frame from an output signal of the second truncation means; A code for a moving image signal, comprising: means for outputting the output signal of the truncation means as an encoded image signal; and means for generating mode information for instructing which coefficient to truncate in the first truncation means. conversion device. 4 means for orthogonally transforming an input image signal; a first truncation means for truncating some coefficients of the orthogonal transform coefficients obtained by the orthogonal transformation; an output signal of the first truncation means and an output of an interframe predictor; a means for taking a difference between the signals, a means for taking a sum of an output signal of the means for taking a difference and an output signal of the interframe predictor, and a means for truncating from the output signal of the means for taking a sum. a second truncation means for truncating a coefficient related to the truncated coefficient; an interframe predictor for predicting and outputting a signal of the next frame from the output signal of the second truncation means; and a difference calculating means. An encoding device for a moving image signal, comprising means for outputting an output signal as an encoded image signal, and means for generating mode information for instructing which coefficients are to be truncated in the first truncation means. 5 means for orthogonally transforming an input image signal; means for taking the difference between the orthogonal transform coefficients obtained by the orthogonal transform and the output signal of the interframe predictor; and extracting some coefficients from the output signal of the means for taking the difference. a first truncation means for truncating; a means for summing the output signal of the difference taking means and an output signal of the interframe predictor; a second truncation means that truncates a coefficient related to the coefficient that has been truncated; an interframe predictor that predicts and outputs a signal of the next frame from the output signal of the second truncation means; means for outputting the output signal as an encoded image signal;
and means for generating mode information for instructing which coefficients are to be truncated in the truncation means. 6 means for orthogonally transforming an input image signal; a first truncation means for truncating some of the orthogonal coefficients obtained by the orthogonal transformation; an output signal of the first truncation means and an output signal of an interframe predictor; a second truncation means for truncating a coefficient related to the coefficient truncated in the first truncation means from the output signal of the difference taking means; and an output signal of the difference taking means. and an output signal of the interframe predictor; an interframe predictor that predicts a signal of the next frame from the output signal of the summation means; and an output signal of the second truncation means. An encoding apparatus for a moving image signal, comprising means for outputting the encoded image signal as an encoded image signal, and means for generating mode information for instructing which coefficients are to be truncated by the first truncating means. 7 means for orthogonally transforming an input image signal; means for taking the difference between the orthogonal transform coefficient obtained by the orthogonal transformation and the output signal of the second truncation means; and a part of the output signal of the means for taking the difference. a first truncation means for truncating a coefficient of , a means for summing an output signal of the first truncation means and an output signal of the second truncation means, and a method for determining the next frame from the output signal of the summation means. an interframe predictor for predicting the progression of the interframe predictor; a second truncation means for truncating and outputting coefficients related to the coefficients truncated by the first truncation means from the output signal of the interframe predictor; A means for outputting the output signal of the truncation means as an encoded image signal, and a means for generating mode information for instructing which coefficient to truncate in the first truncation means. Encoding device. 8 means for orthogonally transforming an input image signal; a first truncation means for truncating some of the orthogonal transformation coefficients obtained by the orthogonal transformation; and an output signal of the first truncation means and a second truncation means. means for calculating the difference between the output signal and the output signal; means for calculating the sum of the output signal of the means for calculating the difference and the output signal of the second truncating means; and a signal of the next frame from the output signal of the means for calculating the sum. an interframe predictor for predicting, a second truncation means for truncating and outputting a coefficient related to the coefficient truncated by the first truncation means from the output signal of the interframe predictor, and a means for taking the difference. means for outputting the output signal of the encoded image signal as an encoded image signal;
and means for generating mode information for instructing which coefficients are to be truncated in the truncation means. 9 means for orthogonally transforming an input image signal; means for taking the difference between the orthogonal transform coefficients obtained by the orthogonal transform and the output signal of the second truncation means; a first truncation means for truncating a coefficient; a means for summing the output signal of the difference taking means and an output signal of the second truncation means; and a signal of the next frame from the output signal of the sum summing means. an interframe predictor for predicting, a second truncation means for truncating and outputting a coefficient related to the coefficient truncated by the first truncation means from the output signal of the interframe predictor, and the first truncation. Encoding of a moving image signal, characterized by comprising means for outputting an output signal of the means as an encoded image signal, and means for generating mode information instructing which coefficients to be truncated in the first truncation means. Device.