JP3474862B2 - Moving picture decoding apparatus and moving picture decoding method - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a moving picture decoding apparatus and a moving picture decoding method for decoding a compression coded signal to reproduce an original moving picture signal, and more particularly to an error in a transmission line / storage medium. The present invention relates to a moving picture decoding apparatus and a moving picture decoding method that are strong in use.

[0002]

2. Description of the Related Art Video signals are transmitted / received in TV telephones, TV conference systems, personal digital assistants, digital video disc systems, digital TV broadcasting systems, etc.
In a storage system, a moving image signal is compression-encoded into the smallest possible amount of information, the obtained compression-encoded signal is transmitted / stored in a transmission line / storage medium, and the transmitted / stored code sequence is stored. The original moving image signal is reproduced by decoding.

As compression encoding techniques for moving image signals applied to such a system, methods such as motion compensation, discrete cosine transform (DCT), subband encoding, pyramid encoding, etc., or a combination of these methods, etc. Various methods have been developed. In addition, ISO / MPEG1, MPEG2, and ITU- are used as international standard methods for compressing and encoding moving images.
T.H. 261, H.H. 262 is specified. All of these coding methods are based on a combination of motion-compensated adaptive prediction and discrete cosine transform. Reference 1: Hiroshi Yasuda, "International Standard for Multimedia Coding," Maruzen, (1991). June) etc. for details.

By the way, when a code string obtained by coding a moving image signal in this way is transmitted / stored through a medium such as a wireless transmission line where an error is likely to occur, an image reproduced on the decoding side. Signals can be degraded by errors in transmission / storage. As a countermeasure against such an error,
For example, under the condition that a code string can be transmitted via a plurality of transmission paths with different error probabilities, the code string is divided into several layers in order to reduce the image quality deterioration due to an error, and the code string in the upper layer has a higher error probability. A hierarchical coding method is known in which transmission is performed on a transmission path having a low transmission rate. To divide the hierarchy,
A method has been proposed in which mode information, motion compensation information, and low frequency components of an image signal are used as upper layers, and high frequency components of image signals are used as upper layers.

As an example of the conventional hierarchical moving picture coding apparatus, a moving picture coding apparatus using motion compensation adaptive prediction and DCT will be described with reference to FIG. In FIG. 5, the input moving image signal 11 is first subjected to prediction by the prediction circuit 12. The prediction circuit 12 detects a motion vector between the input moving image signal 11 and the reference image signal already stored in the frame memory 13 and obtained by encoding / local decoding, and based on this motion vector Motion-compensated prediction is performed for each predetermined unit area (referred to as a prediction area) to create a motion-compensated prediction signal.

The prediction circuit 12 has a motion-compensated prediction mode (inter-frame prediction mode) and an intra-frame prediction mode in which the input moving picture signal 11 is coded as it is, and the optimum mode for coding is selected from these prediction modes. It selects and outputs the prediction signal 14 corresponding to each mode. That is, the prediction circuit 12 outputs a motion compensation prediction signal in the motion compensation prediction mode and "0" as the prediction signal 14 in the intra-frame prediction mode, respectively.

The subtractor 15 subtracts the prediction signal 14 from the input moving image signal 11 to generate a prediction residual signal 16
Is generated. The prediction residual signal 16 is subjected to a discrete cosine transform in a discrete cosine transform (DCT) circuit 17 in units of blocks having a certain size, and becomes DCT coefficient information 18. The DCT coefficient information 18 is quantized by the quantization circuit 19. In the intra-frame prediction mode, the prediction signal 14 is "0".
Therefore, the input moving image signal 11 is directly output from the subtractor 15 as the prediction residual signal 16.

Quantized DCT from quantization circuit 19
The coefficient information 20 is bifurcated into two parts. One is variable length coded by the first variable length coding circuit 21 and the other is dequantized by the dequantization circuit 22. The output of the inverse quantization circuit 22 is inverse discrete cosine transformed by an inverse discrete cosine transform (inverse DCT) circuit 23. That is, the inverse quantization circuit 2
2 and the inverse DCT circuit 23, quantizer circuits 19 and D
A process reverse to that of the CT circuit 17 is performed, and a signal approximate to the prediction residual signal 16 is obtained at the output of the inverse DCT circuit 23.
The output of the inverse DCT circuit 23 is added to the prediction signal 14 from the prediction circuit 12 in the addition circuit 24 to generate the local decoded signal 25. The locally decoded signal 25 is stored in the frame memory 13 as a reference image signal.

Prediction mode / motion vector information 26 is output from the prediction circuit 12 as information relating to prediction, and variable length coding circuit 27 performs variable length coding. The code strings output from the variable-length coding circuits 21 and 27 are multiplexed by the multiplexer 30, divided into an upper layer code string 31 and a lower layer code string 32, and output to a transmission line / storage medium (not shown). It That is, the upper layer code string 31 is output to the transmission line / storage medium where the error probability due to transmission / accumulation is relatively low, and the lower layer code string 32 is the transmission line / storage medium where the error probability due to transmission / storage is relatively high. It is output to the storage medium.

Here, the division of the upper layer code string 31 and the lower layer code string 32 by the multiplexer 30 is a mode indicating the prediction mode in the prediction circuit 12 from the variable length coding circuit 22 as shown in FIG. 6, for example. The information, the motion vector information (MV), and the low-frequency DCT coefficient information of the variable-length coded DCT coefficient information from the variable-length coding circuit 21 are applied to the upper layer code string 31, and the variable length coding circuit 21 The remaining high-frequency DCT coefficient information of the long-coded DCT coefficient information is applied to the lower layer code string 32.

Such a conventional hierarchical moving picture coding apparatus has the following problems. First, since there is only one motion vector information in each prediction region, which causes the image quality to be greatly degraded when an error occurs, if an error occurs in the motion vector information, the motion information cannot be decoded at all for that prediction region. That is, the image quality is greatly deteriorated. In order to reduce such image quality deterioration, FIG.
As shown in (a), all the motion vector information (MV) may be applied to the higher layer code sequence 31. However, in general, there is a limit to the ratio of the code amount of the code string of each layer to the total code amount of the code strings of all layers, and thus when all the motion vector information is included in the upper layer code string 31. May exceed this limit. If motion vector information is applied to the lower layer code string 32 in order to prevent this, there arises a problem that the error resistance is significantly reduced.

Further, code strings 31, 32 to be transmitted / stored
Since each code word is composed of a variable-length code created by the variable-length coding circuits 21 and 27, an error may cause out-of-sync of the variable-length code during decoding.
However, in the conventional moving image coding apparatus, as shown in FIG. 6, important predictions such as mode information and motion vector information that cause a large deterioration in a decoded image when an error occurs in the code strings 31 and 32 of each layer are provided. Related information and DCT of prediction residual signal that does not cause much deterioration even if an error occurs
Coefficient information and the like are mixed and multiplexed. For this reason, if out-of-sync occurs during decoding of a codeword carrying unimportant information, it may spread errors to the codeword carrying important information, resulting in a large deterioration in the decoded image. Give rise to. In such a case, since the synchronization cannot be recovered until the synchronization code appears, all the information of the decoded image up to that point becomes erroneous, and there is a problem that a large deterioration occurs in a wide range in the screen. is there.

[0013]

As described above, in the conventional moving picture coding apparatus, information related to prediction, such as motion vector information, which greatly deteriorates the quality of a decoded image when an error occurs, is used in each prediction. Since only one is coded for each area, error resistance is low.

In order to improve error resilience, all prediction information must be transmitted / stored by a transmission line / storage medium having a low error probability, but the code amount of the code string of each layer is the code of all layers. Since the ratio that can be taken in the total code amount of the sequence is limited, as a result, multiple transmission lines with different error probabilities /
A problem arises in that the characteristic of hierarchical encoding, which reduces the deterioration of image quality due to errors, cannot be utilized by transmitting / accumulating a code string using a storage medium.

Further, in the conventional moving image coding apparatus, relatively important information such as information related to prediction and other relatively unimportant information are mixed and included in the code string. There is a problem that an error caused by unimportant information spreads to important information and causes a large deterioration in image quality.

Therefore, an object of the present invention is to provide a moving picture decoding apparatus and a moving picture decoding method having high error resistance.

[0017]

In order to solve the above problems, the present invention provides an input moving image signal and a motion compensation prediction mode and intra-frame for each of a plurality of blocks into which the input image forming the input moving image signal is divided. Quantized DCT of a prediction residual signal which is an error signal with respect to a prediction signal by performing prediction in a prediction mode selected from a plurality of prediction modes including a prediction mode
Means for inputting a code string including coefficient information, comprising (1)
When the intra-frame prediction mode is selected, the DC component information of the DCT coefficient and the AC component information of the DCT coefficient relating to each block are collected for each of a plurality of blocks, and the DC of the DCT coefficient is
A means for inputting a code string in which component information is multiplexed in the vicinity of information indicating a prediction mode related to each block rather than the AC component information, (2) DCT related to each block when the intra-frame prediction mode is selected Means for collecting a plurality of blocks of DC component information and AC component information of the coefficient, and inputting a code string obtained by multiplexing the DC component information of the DCT coefficient adjacent to information indicating a prediction mode related to each block, (3) When the intra-frame prediction mode is selected, the DC component information and the AC component information of the DCT coefficient relating to each block are collected for a plurality of blocks, and the D
Means for inputting a code string in which the DC component information of the CT coefficient is arranged closer to the information indicating the prediction mode relating to each block than the AC component information, (4) each block when the intra-frame prediction mode is selected Of DCT coefficient related to
A means for inputting a code string in which the C component information and the AC component information are grouped into a plurality of blocks and the DC component information of the DCT coefficient is arranged adjacent to the information indicating the prediction mode related to each block, (5) intra-frame prediction When the mode is selected, the DC component information of the DCT coefficients relating to a plurality of blocks is multiplexed in the vicinity of the information indicating the prediction mode relating to each block as a code string having a relatively high degree of importance, and the DCT coefficients relating to a plurality of blocks are Means for inputting a code string in which AC component information is multiplexed as a code string having a relatively low degree of importance, (6) DC component information of DCT coefficients relating to a plurality of blocks when the intra-frame prediction mode is selected As a code string having a relatively high degree of importance, is multiplexed adjacent to the information indicating the prediction mode related to each block, and DC related to a plurality of blocks Means for inputting a code string in which the AC component information of the coefficient is separated and multiplexed as a code string having a relatively low degree of importance, and (7) indicates a prediction mode related to a plurality of blocks when the intra-frame prediction mode is selected. Information and DC
A means for inputting a code string in which the DC component information of the T coefficient is arranged as a code string of relatively high importance, and the AC component information of the DCT coefficient relating to a plurality of blocks is arranged as a code string of relatively low importance, 8) When the intra-frame prediction mode is selected, the information indicating the prediction mode related to a plurality of blocks and the DC component information of the DCT coefficient are generated as a code string of relatively high importance, and the DCT coefficient related to a plurality of blocks is generated. It is provided with any one of means for inputting a code string generated as a code string with relatively low importance of AC component information, and means for decoding the input code string to reproduce the original moving image signal. And

Further, according to the present invention, an input moving image signal and a plurality of prediction modes including a motion compensation prediction mode and an intra-frame prediction mode are selected for each of a plurality of blocks into which the input image forming the input moving image signal is divided. A code string including quantized DCT coefficient information of a prediction residual signal which is an error signal from the prediction signal by performing prediction in the prediction mode, wherein (1)
When the intra-frame prediction mode is selected, the DC component information of the DCT coefficient and the AC component information of the DCT coefficient relating to each block are collected for each of a plurality of blocks, and the DC of the DCT coefficient is
A code string in which the component information is multiplexed in the vicinity of the information indicating the prediction mode related to each block rather than the AC component information,
(2) When the intra-frame prediction mode is selected, the DC component information and the AC component information of the DCT coefficient related to each block are grouped for a plurality of blocks, and the DC component information of the DCT coefficient is set to the prediction mode related to each block. DC associated with each block when (3) intra-frame prediction mode and a code string multiplexed adjacent to the information shown are selected
A code string in which the DC component information and the AC component information of the T coefficient are grouped for a plurality of blocks, respectively, and the DC component information of the DCT coefficient is arranged closer to the information indicating the prediction mode related to each block than the AC component information, (4 ) DC related to each block when the intra-frame prediction mode is selected
A code string in which the DC component information and the AC component information of the T coefficient are collected for a plurality of blocks respectively, and the DC component information of the DCT coefficient is arranged adjacent to the information indicating the prediction mode related to each block, (5) intra-frame prediction mode When is selected, the DC component information of the DCT coefficients relating to a plurality of blocks is multiplexed in the vicinity of the information indicating the prediction mode relating to each block as a code string having a relatively high degree of importance, and the AC of the DCT coefficients relating to a plurality of blocks is A code string in which the component information is multiplexed behind it as a code string of relatively low importance, (6)
When the intra-frame prediction mode is selected, the DC component information of the DCT coefficients relating to a plurality of blocks is multiplexed as a code string having a relatively high degree of importance adjacent to the information indicating the prediction mode relating to each block, and is divided into a plurality of blocks. DCT involved
A code string in which the AC component information of the coefficient is separated and multiplexed as a code string of relatively low importance, (7) information indicating a prediction mode relating to a plurality of blocks and a DCT coefficient when the intra-frame prediction mode is selected (8) Intra-frame prediction mode, in which the DC component information of the above is arranged as a code string of relatively high importance, and the AC component information of the DCT coefficients related to a plurality of blocks is arranged as a code string of relatively low importance. When is selected, the information indicating the prediction mode related to a plurality of blocks and the DC component information of the DCT coefficient are generated as a code string having a relatively high degree of importance, and the DC related to a plurality of blocks is selected.
It is characterized in that one of the code strings generated as the AC coefficient information of the T coefficient as a code string having a relatively low importance is input, and the code string is decoded to reproduce the original moving image signal.

[0019]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram of an embodiment of a moving picture coding apparatus according to the present invention, and shows an example in which the present invention is applied to a moving picture coding apparatus in which motion compensation adaptive prediction and DCT are combined.

In FIG. 1, the input moving image signal 101 is first used for prediction in the prediction circuit 102. That is, the predicting circuit 102 detects a motion vector between the input moving image signal 101 and the reference image signal already stored in the frame memory 103 by the encoding / local decoding, and based on this motion vector, A motion compensated prediction signal is created. The prediction circuit 102 has a motion-compensated prediction mode (inter-frame prediction mode) and an intra-frame prediction mode in which the input moving image signal 101 is coded as it is, and the optimum prediction mode for coding is selected to select each mode. And outputs the prediction signal 104 corresponding to. That is, the prediction circuit 102 outputs a motion compensation prediction signal in the motion compensation prediction mode and “0” as the prediction signal 104 in the intra-frame prediction mode, respectively.

In the subtractor 105, the input moving image signal 101
The prediction residual signal 106 is generated by subtracting the prediction signal 104 from the prediction residual signal 106. The prediction residual signal 106 is subjected to discrete cosine transform in a block unit of a certain size in a discrete cosine transform (DCT) circuit 107, and becomes DCT coefficient information 108. The DCT coefficient information 108 is quantized by the quantization circuit 109. In the intra-frame prediction mode, since the prediction signal 104 is “0”, the subtractor 105
From the input moving image signal 101 as the prediction residual signal 106.
Is output as is.

Quantized DC from quantization circuit 109
The T coefficient information 110 is bifurcated into two parts. One is variable length coded by the first variable length coding circuit 111, and the other is dequantized by the dequantization circuit 112. The output of the inverse quantization circuit 112 is an inverse discrete cosine transform (inverse DCT) circuit 1
Inverse discrete cosine transform is performed by 13. That is, the inverse quantization circuit 112 and the inverse DCT circuit 113 perform the reverse processing of the quantization circuit 109 and the DCT circuit 107,
A signal approximate to the prediction residual signal 106 is obtained at the output of the inverse DCT circuit 113. The output of the inverse DCT circuit 113 is the prediction signal 10 from the prediction circuit 102 in the addition circuit 114.
4 is added to generate a locally decoded signal 115. The locally decoded signal 115 is stored in the frame memory 103 as a reference image signal.

As will be described later, the prediction circuit 102 outputs large area prediction mode / motion vector information 116 and small area prediction mode / motion vector information 117 as information related to prediction, and the variable length coding circuits 118 and 11 are output.
9 is variable length coded. The code strings output from the variable-length coding circuits 111, 118, and 119 are multiplexed by the multiplexer 120, and the upper layer code string 121 and the lower layer code string 122 are also multiplexed.
And is output to a transmission line / storage medium (not shown).

Here, under the condition that the code sequence can be transmitted / stored via a plurality of transmission lines / storage media having different error probabilities, the upper layer code sequence 121 has a transmission line with a lower error probability. / Transmitting / storing via the storage medium and transmitting / storing via the transmission path / storing medium having a higher error probability with respect to the lower layer code string 122, the upper layer code string 121 is as error-free as possible. To prevent the occurrence of. Further, when performing error correction coding on the coded sequences 121 and 122, more powerful error correction is performed so that the higher-layer coded sequence 121 has a lower error rate than the lower-layer coded sequence 122. Encode.

Next, the configuration and operation of the prediction circuit 102 will be described in detail with reference to FIG. In the prediction circuit 102, the input moving image signal 101 is sequentially divided into more regions in stages from the upper stage to the lower stage, and motion compensation prediction is performed on the input moving image signal for each region divided in each stage. Generate a prediction signal. In the example of FIG. 2, the area division and prediction in the prediction circuit 102 are performed in two stages. That is, in the first stage, the prediction circuit 102 largely divides the input moving image signal 101 as shown by a solid line area (referred to as a large area) in FIG. 2, and performs motion compensation prediction on these large areas with coarse pixel accuracy. Next, in the second stage, the large area is further divided into smaller areas as shown by the broken line areas (referred to as small areas) in the drawing, and motion compensation prediction is performed on these small areas with fine pixel accuracy.

Then, the variable length coding circuits 118 and 119.
Thus, not only the information about the prediction about the large area output from the prediction circuit 102 but also the information about the prediction about the large area is encoded about the small area. By doing so, even if the information about the prediction about the small area is lost due to an error, if the information about the prediction about the large area is decoded without error in the decoding device, the prediction can be performed with a rough accuracy. Therefore, it is possible to prevent the deterioration of the quality of the decoded image.

The information related to the prediction output from the prediction circuit 102 is composed of information indicating the prediction mode and information indicating the motion vector. The information indicating the prediction mode of the large area and the information indicating the motion vector (indicated by a solid arrow in FIG. 2), that is, the large area prediction mode / motion vector information 116 is variable-length coded by the variable-length coding circuit 118. . At this time, with respect to the motion vector information, the difference between the motion vector information and the motion vector information of the adjacent already-encoded large area may be subjected to the variable length coding, or the fixed length coding may be performed without taking the difference. Alternatively, the motion vector information may be fixed-length coded for each area at a certain interval, and the motion vector information of other areas may be variable-length coded.

On the other hand, information indicating a prediction mode of a small area and information indicating a motion vector (indicated by a dashed arrow in FIG. 2), that is, small area prediction mode / motion vector information 117,
The variable length coding circuit 119 performs variable length coding. In this case, variable length coding may be performed by taking the difference between the motion vector information and the large area motion vector information for each small area, or block coding, vector quantization, etc. may be collectively used for each large area. Encoding may be performed. When the difference value is variable-length coded, if a large area motion vector can be represented by a reversible function (for example, an average value) based on the small area motion vector, one of the small area motion vectors will be large. Since it can be obtained by calculation based on the regional motion vector and the other small region motion vector, no particular coding is required.

FIG. 3 shows a configuration example of the upper layer code sequence 121 and the lower layer code sequence 122. Figure 3
In the upper layer code sequence 121 shown in (a), a uniquely decodable synchronization code is inserted at the beginning for each encoded frame or for each certain area unit. PSC indicates a synchronization code in frame units. A picture header indicating the coding information of the frame is attached after the synchronization code PSC.
The picture header includes a frame number indicating the temporal position of the frame, information indicating the prediction mode of the frame (mode information), and information indicating the lengths of the code strings of the upper layer and the lower layer of the frame (code amount). Consists of.

Further, as shown in FIG. 3A, if information indicating the size of the large area and the small area and the pixel accuracy of the motion compensation (large area MC accuracy, small area MC accuracy) is added to the picture header, the motion is detected. It is possible to control the code amount of the motion vector information by changing the compensation accuracy for each frame. Thereby, even when the ratio of the code amount of the upper layer code sequence 121 and the lower layer code sequence 122 is defined by the constraint of the transmission path / storage medium, etc., the code amount distribution corresponding to it can be performed. Also, since the total code amount of the motion vector information of each frame can be controlled, it is possible to select the optimum motion compensation accuracy in view of the relationship between the motion compensation accuracy and the code amount of the motion vector information, and the encoding It is possible to improve efficiency.

A characteristic point is that, after the picture header of the higher layer code string 121 shown in FIG. 3A, the coded information of each area is arranged in order from the most important information. Here, the information of high importance is information that causes a large deterioration in a decoded image when an error occurs. That is, after the picture header of the higher layer code sequence 121, information indicating the prediction mode having the highest importance (mode information), that is, the large area prediction mode / motion vector information 116 and the small information output from the prediction circuit 102 is first. The code string of the prediction mode information of the code strings obtained by coding the region prediction mode information 117 by the variable length coding circuit 118 is inserted.

Next, the DC component (intra DC) of the DCT coefficient information of the region in which the intra-frame prediction mode is selected, that is, the prediction residual signal 106 is changed through the DCT circuit 107 and the quantization circuit 109 to change the DCT coefficient information. DC of the code string obtained by encoding in the long encoding circuit 111
Enter the code string of the component. Furthermore, in the area where the motion compensation prediction mode is selected, large area motion vector information (large area MV) indicating rough motion information, that is, the prediction circuit 102.
Large area prediction mode / motion vector information 1 output from
The code string of the motion vector information in the code string obtained by coding 16 with the variable length coding circuit 118 is inserted.

On the other hand, the lower layer code string 1 shown in FIG.
22 includes motion vector information of a small area (small area MV),
That is, the small-region prediction mode / motion vector information 117 output from the prediction circuit 102 is coded by the variable-length coding circuit 119, and the code string of the motion vector information is inserted in the code string, and the DCT is added after that. The high frequency component of the coefficient information, that is, the prediction residual signal 106 is transferred to the DCT circuit 107.
Then, the DCT coefficient information obtained through the quantizing circuit 109 is coded by the variable length coding circuit 111, and the code string of the high frequency component is inserted.

As described above, the motion compensation prediction is performed hierarchically, and the mode information indicating the prediction mode and the large area motion vector information indicating the rough prediction information are used as the upper layer code sequence 121.
To the lower layer code sequence 12
2 are assigned respectively. Therefore, even if the small area motion vector information included in the lower layer code string 122 is lost due to an error in the transmission path / storage medium, the moving picture decoding apparatus may use the large area motion vector information included in the upper layer code string 121. Since motion-compensated prediction can be performed with high accuracy using, it is possible to reduce the probability of significant image quality deterioration in the decoded image.

Further, in this embodiment, the code sequences are arranged in the order of important information in the upper and lower hierarchical code sequences 121 and 122 as well, so that the error generated in the unimportant information spreads to the important information. Therefore, it is possible to prevent a significant deterioration in image quality.

Next, an embodiment of the moving picture decoding apparatus according to the present invention will be described. FIG. 4 is a block diagram showing the configuration of a moving picture decoding apparatus corresponding to the moving picture coding apparatus of FIG.

In FIG. 4, the upper layer code sequence 121 and the lower layer code sequence 122 output from the moving image encoding apparatus of FIG. 1 are input via the transmission line / storage medium, and the lower layer code sequence 201 and the lower layer sequence. The hierarchical code string 202 is separated by the demultiplexer 203 into a variable length code 204 of quantized DCT coefficient information, a variable length code 205 of large area prediction mode / motion vector information, and a variable length code 206 of small area prediction mode / motion vector information. Then, it is input to the variable length decoding circuits 207, 208 and 209, respectively.

The variable length decoding circuit 207 uses the variable length code 2
The quantized DCT coefficient information 210 is output by variable-length decoding 04. This quantized DCT coefficient information 210
Is inversely quantized by the inverse quantization circuit 213, and further inversely discrete cosine transformed by the inverse DCT circuit 214, thereby generating the prediction residual signal 215. The prediction residual signal 215 is added to the prediction signal 219 from the prediction circuit 217 in the addition circuit 216 to generate the reproduced image signal 221. The reproduced image signal 221 is output to the outside of the moving image decoding apparatus and is also stored in the frame memory 218 as a reference image signal.

On the other hand, the variable length decoding circuits 208 and 209 perform variable length decoding of the large area prediction mode / motion vector information 211 and the small area prediction mode / motion vector information 212 from the variable length codes 205 and 206. These pieces of information 211 and 212 are input to the prediction circuit 217. The prediction circuit 217 uses the reference image signal and the large area prediction mode / motion vector information 2 stored in the frame memory 218.
11 and small region prediction mode / motion vector information 212
Then, the moving image signal is predicted from, and a prediction signal 219 is generated.

The error decision circuit 220 includes a demultiplexer 203 and variable length decoding circuits 207, 208 and 209.
The presence or absence of an error in the upper layer code sequence 201 and the lower layer code sequence 202 is determined based on the above state, and the determination result is given to the prediction circuit 217. If no error is detected by the error determination circuit 220 in either the upper layer or lower layer code sequence 201 or 202, the prediction circuit 217 uses the reference image signal stored in the frame memory 218 as the basis of the prediction image signal shown in FIG.
The prediction signal 219 that is the same as the prediction signal 104 in FIG.

The above processing is processing for reproducing an image signal corresponding to the moving picture coding apparatus of FIG. 1, and is processing performed by the inverse quantization circuit 213, the inverse DCT circuit 214, the addition circuit 216, and the frame memory 218. Are respectively the inverse quantization circuit 112, the inverse DCT circuit 113, and the addition circuit 11 in FIG.
4. The process is basically the same as that of the frame memory 103. In addition, the variable length decoding circuits 207, 208, 209,
The multiplexer 203 includes the variable length coding circuits 111, 118, 119 and the multiplexer 1 in FIG. 1, respectively.
The reverse process of the process of 20 is performed.

On the other hand, when the error determination circuit 220 detects an error in at least one of the upper and lower layer code sequences 201 and 202, for example, the following information of higher importance than the information in which the error is detected is detected. A reproduced image is created by using this.

That is, (1) when an error is detected in the DCT coefficient information of the block for which motion compensation prediction is included in the lower layer code string 122, the prediction residual signal of that block is set to 0 and the block is correctly decoded. The reproduced image signal 221 is obtained by using the motion compensation prediction signal obtained by using the converted mode information, large area motion vector information and small area motion vector information as the prediction signal 219.

(2) When an error occurs in the small area motion vector information, a motion compensated prediction signal obtained by using the large area motion vector information for the small area is used as the prediction signal 219 to reproduce the image signal. 221 is obtained.

Further, (3) when an error occurs in the large area motion vector information, the motion compensation prediction of the large area is performed from the surrounding area or the motion vector information of the already decoded frame for the large area. If possible, the estimated value is used, and if an error also occurs in the surrounding motion vector information, the image signal of the already decoded frame is directly used as the reproduced image signal 221.

(4) When an error occurs in the AC component of the DCT coefficient information of the block in which the intra-frame prediction mode is selected, the DC component of the DCT coefficient information and the correctly decoded image signal of the surrounding block The image signal of the block is predicted and used as the reproduced image signal 221, or the image signal of the block is predicted using the predicted signal of the correctly decoded image signal of the surrounding blocks from the image signal of the already decoded frame. To be reproduced image signal 221.

By the way, when a variable length code is used for encoding various kinds of information, an out-of-sync occurs due to an error, and the error may spread to subsequent codes until the synchronization is recovered by detecting the sync code or the like. In such a case, the subsequent code is not used for decoding. For example, the lower layer code sequence 12
When an error occurs in the small area motion vector information in 2, the error may spread to the motion vector information after the small area and the subsequent DCT coefficient information. In such a case, the error spreads. The existing information is not used for decryption. Even if such a loss of synchronization occurs, the codewords are arranged in the code string in the order of high importance, so that the error generated in the information of low importance can spread to the information of high importance. Therefore, it is possible to prevent the reproduced image from being significantly deteriorated.

The error determination unit 220 uses the code sequences 201 and 202.
The following method can be given as a specific method for detecting that an error has occurred.

The first is a method using an error detection code such as a parity code or a CRC code. In this case, the multiplexer 120 of the moving picture coding apparatus shown in FIG. 1 performs error detection coding on the variable length code, and the demultiplexer 203 of the moving picture decoding apparatus shown in FIG. 4 performs error detection processing. The detection result is given to the error determination unit 220.

Second, the variable length decoding circuits 207 and 20.
8 and 209, when a codeword that does not exist in the codeword table representing the correspondence between the input codeword and the output value is detected, it is determined as an error. In particular, when a variable length code is used, the error may spread not only in the part where the error was detected but also in the code sequence before and after it. Perform processing.

Third, a signal such as motion vector information reproduced based on a decoded code word, prediction signal, DCT coefficient information, prediction residual signal, reproduced image signal, etc., which cannot occur in encoding a moving image. This is a method of performing error determination by determining whether or not The use of this method is characteristic of the present invention and will be described in more detail.

For example, if the motion vector indicated by the motion vector information exceeds the predetermined search range or is outside the screen, it is determined to be an error.

It is also possible to detect an error by determining the DCT coefficient information dequantized by the dequantization circuit 213. The possible range of the pixel value of the input image signal 101 is 0 to D−1, and the DCT block size is N ×
If N, the DCT coefficient takes a value within the following range.

<Intra-frame prediction mode> DC component: 0 to NxD AC component:-(N / 2xD) to (N / 2xD) <Interframe prediction mode>-(N / 2xD) ˜ (N / 2 × D) Then, if the value of the decoded DCT coefficient takes a value outside this range, it is determined as an error. In this case, all or some of the D blocks in which an error is detected are
The CT coefficient may be set to 0 or the decoded value may be estimated from the decoded values of surrounding blocks.

Further, it is possible to judge an error by the pixel value of the reproduced image signal 221. Assuming that the range of pixel values of the input image signal 101 is 0 to D-1, the DCT block size is N × N, and the quantization width is Q (in the case of linear quantization), the pixel values of the reproduced image signal 221 are calculated. The obtained range is -NxQ to D + NxQ. Therefore, when the pixel value of the decoded reproduced image signal 221 exceeds this range, it is determined as an error. In that case, for example, the prediction residual signal 215 is set to “0” in the inter-frame prediction mode (motion compensation prediction mode), and a part of the DCT coefficient input to the inverse DCT circuit 214 is set to “0” in the intra-frame prediction mode. Inverse DCT is performed to obtain the reproduced image signal 221, or the reproduced image signal 2 is obtained.
The estimation may be performed from the pixel values of blocks around 21.

As described above, according to the present invention, whether or not the reproduced information or signal is the information or signal which cannot be generated in the transmission line / storage medium is used for the error judgment in the error judgment unit 220 in the moving picture decoding apparatus. By adding such a judgment, it becomes possible to make a more accurate error judgment. For this reason, it is possible to suppress the quality deterioration of the reproduced image which is caused by directly using the information or the signal in which the error has occurred without performing the error processing for reproducing the moving image signal.

The present invention can be implemented with various modifications. For example, in the above embodiment, an example in which the code string output from the moving image coding apparatus is divided into two layers has been shown, but the coding may be performed by dividing into three layers or more. For example, the frame synchronization code (PSC), picture header, and mode information are set to the highest first layer, and DC in the intra-frame prediction mode is used.
The DC component (intra DC) of the T coefficient information and the large area motion vector information are in the second layer, and the small area motion vector information is in the third layer.
DC other than the DCT coefficient information assigned to the hierarchy and the first hierarchy
The T coefficient information may be assigned to each of the fourth layers. Further, the DCT coefficient may be further divided into several layers, such as a low frequency component and a high frequency component, to be encoded.

In the coding of the large area motion vector information, the motion of variable length coding the difference between the motion vector information for which fixed length coding is performed as described above and the motion vector information for which this fixed length coding is performed. When the vector information is divided into two types and encoded, the fixed-length encoded motion vector information in which an error does not spread to the subsequent code sequence due to the loss of synchronization is collected in a frame or a certain area unit first. It is effective to put it and put the motion vector information coded in variable length after it. In this way, even if an error occurs in the variable-length coding part and the synchronization is lost, the error does not propagate to the variable-length coded motion vector information, so the fixed-length coding is performed at the time of decoding. Since it is possible to estimate a motion vector in which an error has occurred based on the motion vector information and create a prediction signal with rough accuracy, it is possible to reduce deterioration in image quality of a reproduced image due to the error.

The method for determining whether or not the information reproduced in the moving picture decoding apparatus of FIG. 4 is information that cannot occur in moving picture coding is not limited to the hierarchically coded code string, It can also be applied to a moving picture decoding apparatus that decodes an original image signal from a code string obtained by a general moving picture coding apparatus.

[0060]

As described above, according to the present invention, a moving picture decoding apparatus and a moving picture having a high error resistance, that is, a deterioration in the image quality of a decoded image due to an error occurring at the time of transmitting / accumulating a coded sequence is small A decoding method can be provided.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of a moving picture coding apparatus according to the present invention.

FIG. 2 is a diagram showing motion compensation areas and corresponding motion vectors in the moving picture coding apparatus of FIG.

FIG. 3 is a diagram showing an example of upper layer and lower layer code strings output from the moving image encoding apparatus of FIG.

4 is a block diagram showing an embodiment of a moving picture decoding apparatus corresponding to the moving picture coding apparatus of FIG.

FIG. 5 is a block diagram showing an example of a conventional moving image encoding device.

FIG. 6 is a diagram showing an example of a code string output from the moving picture coding apparatus in FIG.

[Explanation of symbols]

101 ... Input image signal 102 ... Prediction circuit 103 ... Frame memory 104 ... Prediction signal 105 ... Subtractor 106 ... Prediction residual signal 107 ... Discrete cosine transform circuit 108 ... DC
T coefficient information 109 ... Quantization circuit 110 ... Quantized DCT coefficient information 111 ... Variable length coding circuit 112 ... Inverse quantization circuit 113 ... Inverse discrete cosine transform circuit 114 ... Adder 115 ... Local decoded signal 116 ... Large area prediction mode / Motion vector information 117 ... Small region prediction mode / Motion vector information 118,119 ... Variable length coding circuit 120 ... Multiplexer 121 ... Upper layer code string 122 ... Lower layer code string 201 ... Upper layer code string 202 ... Lower layer code string 203 ... Demultiplexer 204-20
6 ... Variable-length codes 207 to 209 ... Variable-length coding circuit 210 ... Quantized DCT coefficient 211 ... Large area prediction mode / motion vector information 212 ... Small area prediction mode / motion vector information 213 ... Inverse quantization circuit 214 ... Inverse discrete Cosine transform circuit 215 ... Prediction residual signal 216 ... Adder 217 ... Prediction circuit 218 ... Frame memory 219 ... Prediction signal 220 ... Error determination circuit 221 ... Reproduced image signal

─────────────────────────────────────────────────── ─── Continuation of the front page (72) Takashi Ida, Komukai Toshiba Town, Komukai-shi, Kawasaki-shi, Kanagawa 1st, Toshiba Research & Development Center (72) Inventor Noboru Yamaguchi Komukai-shiba, Saiwai-ku, Kawasaki-shi, Kanagawa No. 1 Toshiba Research & Development Center Co., Ltd. (72) Inventor Ken Nakajo No. 1 Komukai Toshiba-cho, Kouki-ku, Kawasaki-shi, Kanagawa Toshiba Research & Development Center Co., Ltd. (58) Fields investigated (Int.Cl. ⁷ , DB name) ) H04N 7/24-7/68

Claims (16)

(57) [Claims] 1. A have rows predicted by the selected prediction mode from among a plurality of prediction modes including a plurality of motion compensation prediction modes for each block and intra-frame prediction mode obtained by dividing the input image constituting the input moving image signal, In the motion compensation prediction mode
Prediction signal generated when prediction is performed and the input moving image
A prediction residual signal that is an error signal with the image signal, or
Input video when prediction is made in intra-frame prediction mode
A code string including DCT coefficient information quantized for an image signal , when the intra-frame prediction mode is selected,
DC component information of DCT coefficient related to each block and AC
The component information is collected for each of a plurality of blocks, and the DCT
Means for inputting a code string in which the DC component information of the coefficient is multiplexed in the vicinity of the information indicating the prediction mode relating to each block rather than the AC component information, and decoding the input code string to obtain the original moving image signal. A moving picture decoding apparatus comprising: a reproducing unit. 2. A have rows predicted by the prediction mode selected from among a plurality of prediction modes including a plurality of motion compensation prediction modes for each block and intra-frame prediction mode obtained by dividing the input image constituting the input moving image signal, In the motion compensation prediction mode
Prediction signal generated when prediction is performed and the input moving image
A prediction residual signal that is an error signal with the image signal, or
Input video when prediction is made in intra-frame prediction mode
A code string including DCT coefficient information quantized for an image signal , when the intra-frame prediction mode is selected,
DC component information of DCT coefficient related to each block and AC
The component information is collected for each of a plurality of blocks, and the DCT
A means for inputting a code string in which the DC component information of the coefficient is adjacently multiplexed with information indicating a prediction mode related to each block, and a means for decoding the input code string to reproduce the original moving image signal are provided. A moving picture decoding device comprising: 3. There line prediction by the prediction mode selected from among a plurality of prediction modes including a plurality of motion compensation prediction modes for each block and intra-frame prediction mode obtained by dividing the input image constituting the input moving image signal, In the motion compensation prediction mode
Prediction signal generated when prediction is performed and the input moving image
A prediction residual signal that is an error signal with the image signal, or
Input video when prediction is made in intra-frame prediction mode
A code string including DCT coefficient information quantized for an image signal , when the intra-frame prediction mode is selected,
DC component information of DCT coefficient related to each block and AC
The component information is collected for each of a plurality of blocks, and the DCT
A means for inputting a code string in which the DC component information of the coefficient is arranged closer to the information indicating the prediction mode relating to each block than the AC component information, and the input video sequence is decoded to obtain the original moving image signal. A moving picture decoding apparatus comprising: a reproducing unit. 4. There line prediction by the prediction mode selected from among a plurality of prediction modes including a plurality of motion compensation prediction mode and the intra prediction mode for each block obtained by dividing the input image constituting the input moving image signal, In the motion compensation prediction mode
Prediction signal generated when prediction is performed and the input moving image
A prediction residual signal that is an error signal with the image signal, or
Input video when prediction is made in intra-frame prediction mode
A code string including DCT coefficient information quantized for an image signal , when the intra-frame prediction mode is selected,
DC component information of DCT coefficient related to each block and AC
The component information is collected for each of a plurality of blocks, and the DCT
A means for inputting a code string in which the DC component information of the coefficient is arranged adjacent to the information indicating the prediction mode related to each block, and a means for decoding the input code string to reproduce the original moving image signal are provided. A moving picture decoding device comprising: 5. There line prediction by the prediction mode selected from among a plurality of prediction modes including a plurality of motion compensation prediction modes for each block and intra-frame prediction mode obtained by dividing the input image constituting the input moving image signal, In the motion compensation prediction mode
Prediction signal generated when prediction is performed and the input moving image
A prediction residual signal that is an error signal with the image signal, or
Input video when prediction is made in intra-frame prediction mode
A code string including DCT coefficient information quantized for an image signal , when the intra-frame prediction mode is selected,
The DC component information of the DCT coefficients relating to a plurality of blocks is multiplexed in the vicinity of the information indicating the prediction mode relating to each block as a code string having a relatively high importance, and the AC component information of the DCT coefficients relating to a plurality of blocks is relatively important. And a means for inputting a multiplexed code sequence behind the low code sequence and a means for decoding the input code sequence and reproducing the original video signal. apparatus. 6. There line prediction by the prediction mode selected from among a plurality of prediction modes including a plurality of motion compensation prediction modes for each block and intra-frame prediction mode obtained by dividing the input image constituting the input moving image signal, In the motion compensation prediction mode
Prediction signal generated when prediction is performed and the input moving image
A prediction residual signal that is an error signal with the image signal, or
Input video when prediction is made in intra-frame prediction mode
A code string including DCT coefficient information quantized for an image signal , when the intra-frame prediction mode is selected,
The DC component information of the DCT coefficients relating to a plurality of blocks is adjacently multiplexed with the information indicating the prediction mode relating to each block as a code string having a relatively high degree of importance, and the AC component information of the DCT coefficients relating to a plurality of blocks is relatively important. Moving picture decoding, characterized in that it comprises means for inputting a code string that has been separated and multiplexed as a low-degree code string, and means for decoding the input code string and reproducing the original moving picture signal. Device. 7. There line prediction by the prediction mode selected from among a plurality of prediction modes including a plurality of motion compensation prediction mode and the intra prediction mode for each block obtained by dividing the input image constituting the input moving image signal, In the motion compensation prediction mode
Prediction signal generated when prediction is performed and the input moving image
A prediction residual signal that is an error signal with the image signal, or
Input video when prediction is made in intra-frame prediction mode
A code string including DCT coefficient information quantized for an image signal , when the intra-frame prediction mode is selected,
Information indicating a prediction mode related to a plurality of blocks and DC component information of DCT coefficients are arranged as a code string having a relatively high importance, and AC component information of DCT coefficients related to a plurality of blocks is a code string having a relatively low importance. A moving image decoding apparatus comprising: a means for inputting the arranged code string; and a means for decoding the input code string to reproduce the original moving image signal. 8. There line prediction by the prediction mode selected from among a plurality of prediction modes including a plurality of motion compensation prediction modes for each block and intra-frame prediction mode obtained by dividing the input image constituting the input moving image signal, In the motion compensation prediction mode
Prediction signal generated when prediction is performed and the input moving image
A prediction residual signal that is an error signal with the image signal, or
Input video when prediction is made in intra-frame prediction mode
A code string including DCT coefficient information quantized for an image signal , when the intra-frame prediction mode is selected,
Information indicating a prediction mode related to a plurality of blocks and DC component information of DCT coefficients are generated as a code string having a relatively high importance, and AC component information of DCT coefficients related to a plurality of blocks is a code string having a relatively low importance. A moving image decoding apparatus comprising: a unit for inputting the generated code string; and a unit for decoding the input code string to reproduce the original moving image signal. 9. There line prediction by the prediction mode selected from among a plurality of prediction modes including a plurality of motion compensation prediction mode and the intra prediction mode for each block obtained by dividing the input image constituting the input moving image signal, In the motion compensation prediction mode
Prediction signal generated when prediction is performed and the input moving image
A prediction residual signal that is an error signal with the image signal, or
Input video when prediction is made in intra-frame prediction mode
A code string including DCT coefficient information quantized for an image signal , when the intra-frame prediction mode is selected,
DC component information of DCT coefficient related to each block and AC
The component information is collected for each of a plurality of blocks, and the DCT
A code string in which the DC component information of the coefficient is multiplexed in the vicinity of the information indicating the prediction mode related to each block rather than the AC component information is input, and the code string is decoded to reproduce the original moving image signal. A characteristic video decoding method. 10. There line prediction by the prediction mode selected from among a plurality of prediction modes including a plurality of motion compensation prediction mode and the intra prediction mode for each block obtained by dividing the input image constituting the input moving image signal, The motion compensation prediction mode
The prediction signal generated when the prediction is performed in
Prediction residual signal which is an error signal with the image signal, or
When the prediction is performed in the intra-frame prediction mode, the input motion
A code string including DCT coefficient information quantized with respect to an image signal, and when the intra-frame prediction mode is selected, the DC component information and the AC component information of the DCT coefficient relating to each block are collected for a plurality of blocks, respectively, and D
A moving image characterized by inputting a code string in which DC component information of CT coefficients is adjacent to information indicating a prediction mode related to each block, and decoding the code string to reproduce an original moving image signal. Image decoding method. 11. There line prediction by the prediction mode selected from among a plurality of prediction modes including a plurality of motion compensation prediction modes for each block and intra-frame prediction mode obtained by dividing the input image constituting the input moving image signal, The motion compensation prediction mode
The prediction signal generated when the prediction is performed in
Prediction residual signal which is an error signal with the image signal, or
When the prediction is performed in the intra-frame prediction mode, the input motion
A code string including DCT coefficient information quantized with respect to an image signal, and when the intra-frame prediction mode is selected, the DC component information and the AC component information of the DCT coefficient relating to each block are collected for a plurality of blocks, respectively, and D
A code string in which the DC component information of the CT coefficient is arranged closer to the information indicating the prediction mode related to each block than the AC component information is input, and the code string is decoded to reproduce the original moving image signal. A characteristic video decoding method. 12. There line prediction by the prediction mode selected from among a plurality of prediction modes including a plurality of motion compensation prediction mode and the intra prediction mode for each block obtained by dividing the input image constituting the input moving image signal, The motion compensation prediction mode
The prediction signal generated when the prediction is performed in
Prediction residual signal which is an error signal with the image signal, or
When the prediction is performed in the intra-frame prediction mode, the input motion
A code string including DCT coefficient information quantized with respect to an image signal, and when the intra-frame prediction mode is selected, the DC component information and the AC component information of the DCT coefficient relating to each block are collected for a plurality of blocks, respectively, and D
A moving image characterized by inputting a code string in which DC component information of CT coefficients is arranged adjacent to information indicating a prediction mode relating to each block, and decoding the code string to reproduce an original moving image signal. Decryption method. 13. There line prediction by the prediction mode selected from among a plurality of prediction modes including a plurality of motion compensation prediction modes for each block and intra-frame prediction mode obtained by dividing the input image constituting the input moving image signal, The motion compensation prediction mode
The prediction signal generated when the prediction is performed in
Prediction residual signal which is an error signal with the image signal, or
When the prediction is performed in the intra-frame prediction mode, the input motion
A code string including DCT coefficient information quantized for an image signal, and when the intra-frame prediction mode is selected, DC component information of DCT coefficients relating to a plurality of blocks is used as a code string having a relatively high degree of importance for each block. Is input in the vicinity of the information indicating the prediction mode related to, and the AC component information of the DCT coefficients related to a plurality of blocks is input as a code string having a relatively low importance, and the code string is decoded. A method for decoding a moving image, which is characterized by reproducing the original moving image signal. 14. There line prediction by the prediction mode selected from among a plurality of prediction modes including a plurality of motion compensation prediction modes for each block and intra-frame prediction mode obtained by dividing the input image constituting the input moving image signal, The motion compensation prediction mode
The prediction signal generated when the prediction is performed in
Prediction residual signal which is an error signal with the image signal, or
When the prediction is performed in the intra-frame prediction mode, the input motion
A code string including DCT coefficient information quantized for an image signal, and when the intra-frame prediction mode is selected, DC component information of DCT coefficients relating to a plurality of blocks is used as a code string having a relatively high degree of importance for each block. Is input adjacent to the information indicating the prediction mode related to, the AC component information of the DCT coefficients related to a plurality of blocks is separated as a code string having a relatively low importance, and the multiplexed code string is input. A moving picture decoding method characterized by decoding and reproducing an original moving picture signal. 15. There line prediction by the prediction mode selected from among a plurality of prediction modes including a plurality of motion compensation prediction mode and the intra prediction mode for each block obtained by dividing the input image constituting the input moving image signal, The motion compensation prediction mode
The prediction signal generated when the prediction is performed in
Prediction residual signal which is an error signal with the image signal, or
When the prediction is performed in the intra-frame prediction mode, the input motion
A code string including DCT coefficient information quantized with respect to an image signal, and information indicating a prediction mode related to a plurality of blocks and DC when an intra-frame prediction mode is selected.
The DC component information of the T coefficient is arranged as a code string of relatively high importance, and the AC component information of the DCT coefficient related to a plurality of blocks is arranged as a code string of relatively low importance. A moving picture decoding method characterized by decoding a sequence to reproduce an original moving picture signal. 16. There line prediction by the prediction mode selected from among a plurality of prediction modes including a plurality of motion compensation prediction modes for each block and intra-frame prediction mode obtained by dividing the input image constituting the input moving image signal, The motion compensation prediction mode
The prediction signal generated when the prediction is performed in
Prediction residual signal which is an error signal with the image signal, or
When the prediction is performed in the intra-frame prediction mode, the input motion
A code string including DCT coefficient information quantized with respect to an image signal, and information indicating a prediction mode related to a plurality of blocks and DC when an intra-frame prediction mode is selected.
The DC component information of the T coefficient is generated as a code string having a relatively high importance, and the AC component information of the DCT coefficient related to a plurality of blocks is generated as a code string having a relatively low importance, and the code string is input. A moving picture decoding method characterized by decoding a sequence to reproduce an original moving picture signal.