JP6038243B2 - Image predictive decoding method and image predictive decoding device - Google Patents

Description

The present invention relates to an image predictive decoding method and apparatus , and more particularly to a method and apparatus for predictive decoding using a signal in a screen.

  In order to efficiently transmit and store still images and moving image data, compression coding technology is used. In the case of moving images, MPEG1-4 and H.264 are used. 261-H. H.264 is widely used.

  In these encoding methods, encoding / decoding processing is performed after an image to be encoded is divided into a plurality of blocks. MPEG4 and H.264 In H.264, in order to further increase the encoding efficiency, adjacent pre-reproduced pixel signals (reconstructed compressed image data) in the same screen as the target block are used for predictive encoding in the screen. After generating the prediction signal, the difference signal obtained by subtracting the prediction signal from the signal of the target block is encoded. For predictive coding between screens, refer to another already-reproduced image signal on a different screen from the target block, perform motion correction, generate a prediction signal, and generate the generated prediction signal from the signal of the target block The subtracted difference signal is encoded.

  Specifically, H.C. H.264 intra-screen predictive encoding employs a method of generating a prediction signal by extrapolating already reproduced pixel values adjacent to a block to be encoded in a predetermined direction. FIG. 2 is a schematic diagram for explaining an intra-screen prediction method used for H.264. In FIG. 13A, a block 1302 is a target block, and a pixel group composed of pixels A to M (1301) adjacent to the boundary of the target block is an adjacent region, and is an image signal that has already been reproduced in past processing. is there. In this case, an adjacent pixel (A to D) directly above the target block 1302 is duplicated downward to generate a prediction signal. In FIG. 13B, the already reproduced pixels (I to L) on the left of the target block 1304 are duplicated on the right to generate a prediction signal. A specific method for generating a prediction signal is described in Patent Document 1, for example. Thus, the difference between each of the nine prediction signals generated by the method shown in FIGS. 13A to 13I and the pixel signal of the target block is taken, and the one with the smallest difference value is set as the optimum prediction method. . These extrapolation methods can be summarized as shown in FIG. The arrows in FIG. 14 indicate the directions in which the already reproduced pixels are stretched, and the numbers given in each direction are the identification numbers of the respective prediction modes. Also, the identification number predicted by the average of the surrounding already reproduced pixels is 2, which is shown as DC in FIG. This identification number is also referred to as identification information relating to the intra prediction method, mode information, or simply a prediction mode.

  It is necessary to send the prediction mode of the block predicted in the screen to the transmission side. At this time, the intra prediction mode of the target block is encoded with reference to the intra prediction modes of the block adjacent to the target block and the block adjacent to the left. That is, the intra prediction modes of the upper adjacent block and the left adjacent block are compared, and the one with the smaller value is determined as the reference mode information (most probable mode). The in-screen prediction mode of the target block is encoded based on this reference mode information.

  Specifically, a symbol indicating whether the intra prediction mode of the target block is the same as the reference mode information is encoded. When the symbol is 1, the in-screen prediction mode of the target block is the same as the reference mode information. When the symbol is 0, information on the intra prediction mode of the target block is encoded. However, when the number indicating the intra-frame prediction mode of the target block is larger than the number of the reference mode information, encoding is performed after subtracting 1 from the number indicating the intra-screen prediction mode of the target block.

  On the reproduction side, first, the symbol is decoded in the target block predicted in the screen. When the symbol is 1, it means that the in-screen prediction mode of the target block is the same as the reference mode information. When the symbol is 0, information regarding the intra prediction mode is further decoded. However, when the decoded prediction mode number is the same as or larger than the reference mode information, 1 is added and the intra-screen prediction mode of the target block is set.

U.S. Patent Publication No. 6765964.

  By the way, it has been found that the accuracy of intra prediction is improved by providing more intra prediction modes than in the prior art. That is, in addition to the nine modes shown in FIG. 14, it is more effective to have an option of extrapolating the prediction signal from an intermediate angle (direction).

  However, when the number of prediction methods is increased, there is a problem that the encoding efficiency of identification information (prediction mode) for specifying the in-screen prediction method according to the prior art is lowered.

  This is because the probability that the prediction mode of the target block statistically matches the reference mode information (most probable mode) decreases as the number of intra prediction modes increases. In addition, the encoding of the prediction mode itself when it does not match the reference mode information also requires a larger number of bits because the number of intra prediction modes is large.

  The present invention solves the above-described problem, and efficiently encodes mode information for identifying an intra-screen prediction method of a target block when more prediction modes are provided in the intra-screen prediction signal generation method. It is an object to provide a method and apparatus. Another object of the present invention is to provide a method and apparatus for efficiently decoding coded mode information.

In order to achieve the above object, an image predictive encoding device according to the present invention includes:
Area dividing means for dividing the input image into a plurality of blocks;
Prediction signal generating means for determining a prediction mode having the smallest difference value from a plurality of prediction methods for a pixel signal included in a target block to be processed for the plurality of blocks, and generating a prediction signal according to the prediction mode; ,
Residual signal generation means for obtaining a residual signal between the pixel signal of the target block and the prediction signal;
Signal encoding means for encoding the residual signal and generating a compressed signal;
Prediction mode encoding means for encoding the determined prediction mode;
A storage means for restoring the compressed signal and storing it as a reproduced pixel signal,
The prediction mode encoding unit generates a candidate prediction mode list including, as components, prediction modes of a plurality of already-reproduced blocks adjacent to the target block, and a component corresponding to the determined prediction mode is the candidate prediction mode. A flag indicating whether or not the component is in the list is encoded, and if there is a corresponding component, an index to the component in the candidate prediction mode list is further encoded. If there is no corresponding component, the candidate prediction mode list is encoded. After each component is removed and the remaining prediction modes are ordered from the smallest component to the largest component, the determined prediction modes are encoded with a numbering according to the position after ordering of the prediction modes. It is characterized by that.

In order to achieve the above object, an image predictive encoding method according to the present invention includes:
An image predictive encoding method executed by an image predictive encoding device,
An area dividing step for dividing the input image into a plurality of blocks;
A prediction signal for determining a prediction mode having the smallest difference value from a plurality of prediction methods for a pixel signal included in a target block to be processed among the plurality of blocks, and generating a prediction signal according to the determined prediction mode Generation step;
A residual signal generation step for obtaining a residual signal representing a difference between the pixel signal of the target block and the prediction signal;
A signal encoding step of encoding the residual signal to generate a compressed signal;
A prediction mode encoding step for encoding the determined prediction mode;
Storing the decompressed compressed signal and storing it as a reproduced pixel signal;
An image predictive encoding method comprising:
In the prediction mode encoding step, a candidate prediction mode list including a prediction mode of a plurality of already reproduced blocks adjacent to the target block as a component is generated, and a component corresponding to the determined prediction mode is the candidate prediction mode A flag indicating whether or not the component is in the list is encoded. If there is a corresponding component, an index to the component in the candidate prediction mode list is further encoded. If there is no corresponding component, the candidate prediction mode list is encoded. After each component is removed and the remaining prediction modes are ordered from the smallest component to the largest component, the determined prediction modes are encoded with a numbering according to the position after ordering of the prediction modes. It is characterized by that.

In order to achieve the above object, an image predictive decoding apparatus according to the present invention includes:
Input compressed image data including a residual signal generated by predicting and encoding a target block by dividing an image into a plurality of blocks and encoding information regarding a prediction mode indicating a method of generating a prediction signal of the target block Input means;
A restoration means for extracting a residual signal of the target block from the compressed image data and restoring it to a reproduction residual signal;
A prediction mode decoding means for restoring the encoding information relating to the prediction mode and generating the prediction mode;
Prediction signal generation means for generating a prediction signal of the target block based on the prediction mode;
Image restoration means for restoring the pixel signal of the target block by adding the prediction signal to the reproduction residual signal;
Storage means for storing the restored pixel signal as a reproduction pixel signal;
An image predictive decoding device comprising:
The prediction mode decoding means includes
Generating a candidate prediction mode list including, as components, prediction modes of a plurality of already reproduced blocks adjacent to the target block;
Decoding a flag indicating whether a component corresponding to the prediction mode of the target block is in the candidate prediction mode list;
When the flag indicates that “there is a corresponding component”, the index indicating the corresponding component included in the candidate prediction mode list is further decoded, and the component indicated by the index is set as the prediction mode of the target block,
When the flag indicates “no corresponding component”,
Decrypt REM mode,
Substituting the decoded value of the REM mode into the variable of the intra prediction mode of the target block,
For each component in the candidate prediction mode list, the variable is compared with the smallest component of the components not yet compared in the candidate prediction mode list, and when the variable is greater than or equal to the minimum component Repetitively executing a process of adding 1 to the value of the variable, and setting the final value of the variable as the prediction mode of the target block;
It is characterized by that.

In order to achieve the above object, an image predictive decoding method according to the present invention includes:
An image predictive decoding method executed by an image predictive decoding device,
Input compressed image data including a residual signal generated by predicting and encoding a target block by dividing an image into a plurality of blocks and encoding information regarding a prediction mode indicating a method of generating a prediction signal of the target block An input step;
A restoration step of extracting a residual signal of the target block from the compressed image data and restoring it to a reproduction residual signal;
A prediction mode decoding step of restoring the encoding information related to the prediction mode and generating the prediction mode;
A prediction signal generation step of generating a prediction signal of the target block based on the prediction mode;
An image restoration step of restoring the pixel signal of the target block by adding the prediction signal to the reproduction residual signal;
A storage step of storing the reconstructed pixel signal as a reconstructed pixel signal, and an image predictive decoding method comprising:
The prediction mode decoding step includes:
A sub-step of generating a candidate prediction mode list including, as components, prediction modes of a plurality of already reproduced blocks adjacent to the target block;
A sub-step of decoding a flag indicating whether a component corresponding to the prediction mode of the target block is in the candidate prediction mode list;
When the flag indicates “there is a corresponding component”, an index indicating the corresponding component included in the candidate prediction mode list is further decoded, and the component indicated by the index is set as the prediction mode of the target block. Steps,
When the flag indicates “no corresponding component”,
Decrypt REM mode,
Substituting the decoded value of the REM mode into the variable of the intra prediction mode of the target block,
For each component in the candidate prediction mode list, the variable is compared with the smallest component of the components not yet compared in the candidate prediction mode list, and when the variable is greater than or equal to the minimum component Repeatedly executing a process of adding 1 to the value of the variable, and setting the final value of the variable as the prediction mode of the target block;
including,
It is characterized by that.

  The present invention may adopt the following aspects.

  In order to achieve the above object, an image predictive coding method according to the present invention includes an area dividing step for dividing an input image into a plurality of blocks, and pixel signals included in a target block that is a processing target for the plurality of blocks. On the other hand, an optimal prediction mode having the smallest difference value is determined from a plurality of prediction methods, a prediction signal generating step for generating a prediction signal according to the optimal prediction mode, and a residual for obtaining a residual signal between the pixel signal and the prediction signal of the target block A difference signal generation step, a signal encoding step that encodes the residual signal and generates a compressed signal, a prediction mode encoding step that encodes the optimal prediction mode, and a storage that restores the compressed signal and stores it as a reproduced pixel signal In the prediction mode encoding step, the optimum prediction mode of a plurality of already reproduced blocks adjacent to the target block is used as a component. A complementary prediction mode list is generated, and a flag indicating whether there is a component corresponding to the optimum prediction mode in the candidate prediction mode list is encoded. If there is a corresponding component, an index to the corresponding component in the candidate prediction mode list Is encoded, and when there is no corresponding component, the optimal prediction mode is encoded by numbering by removing each component in the candidate prediction mode list.

  Also, the image predictive decoding method according to the present invention is a code relating to a prediction mode indicating a method for generating a residual signal generated by dividing an image into a plurality of blocks and predictively encoding the target block and a prediction signal of the target block. An input step for inputting compressed image data including encoding information, a restoration step for extracting the residual signal of the target block from the compressed image data and restoring it to a reproduction residual signal, and restoring the encoding information related to the prediction mode A prediction mode decoding step for generating an optimal prediction mode, a prediction signal generation step for generating a prediction signal of the target block based on the optimal prediction mode, and adding the prediction signal and the reproduction residual signal, An image restoration step for restoring the pixel signal; and a storage step for storing the restored pixel signal as a reproduction pixel signal. In the prediction mode decoding step, a candidate prediction mode list including the optimum prediction mode of a plurality of already-reproduced blocks adjacent to the target block as a component is generated, and whether there is a component corresponding to the optimum prediction mode in the candidate prediction mode list When the flag indicating that “there is a corresponding component” is decoded, the index indicated in the candidate prediction mode list is further decoded, the component indicated by the index is set to the optimum prediction mode, and the flag is “corresponding” In the case where “there is no component”, the information regarding the REM mode is further decoded, and the value of the REM mode converted based on the candidate prediction mode list is set as the optimum prediction mode.

  According to the present invention, in an image predictive encoding method that performs intra prediction in more intra prediction modes than in the prior art, a candidate consisting of a plurality of prediction modes when encoding prediction mode information of a target block. Since the prediction mode list is prepared and the identifier of the component that matches the prediction mode of the target block is encoded, the appearance probability of the component that matches the prediction mode of the target block is increased, and the prediction mode is reduced with a smaller number of bits. Information can be encoded. In other words, while there is only one “most probable mode” in the past, a plurality of “most probable modes” are prepared in the present invention, so that the “most probable mode” that matches the prediction mode of the target block appears. This has the effect of increasing the probability of

  In addition, when the prediction mode of the target block does not exist in the candidate prediction mode list, the prediction mode itself of the target block is encoded, but the plurality of prediction modes included in the candidate prediction mode list are removed, and the remaining prediction modes Since the identification number is newly assigned and used, the prediction mode of the target block can be expressed by a smaller numerical value, and encoding can be performed with a shorter bit length.

  That is, according to the present invention, there is an effect that information about prediction modes can be encoded more efficiently when intra prediction is performed in more intra prediction modes than in the prior art.

It is a block diagram which shows the image predictive coding apparatus by embodiment of this invention. It is a schematic diagram which shows the pixel extrapolation direction corresponding to the prediction mode in a screen used for the image predictive coding apparatus by embodiment of this invention. 5 is a flowchart illustrating processing of an intra prediction mode encoder according to an embodiment of the present invention. It is a schematic diagram which shows the example for demonstrating the encoding process of the prediction mode in a screen by embodiment of this invention. It is a flowchart which shows the production | generation process (step 360) of the REM mode number in the process (FIG. 3) of the prediction mode encoder in a screen by embodiment of this invention. It is a schematic diagram for demonstrating the production | generation process of the REM mode number in the process (FIG. 3) of the prediction mode encoder in a screen by embodiment of this invention. It is a block diagram which shows the image prediction decoding apparatus by embodiment of this invention. 4 is a flowchart illustrating processing of an intra prediction mode decoder according to an embodiment of the present invention. It is a flowchart which shows the production | generation process (step 860) of the prediction mode of the object block in the process (FIG. 8) of the intra prediction mode decoder by embodiment of this invention. It is a schematic diagram for demonstrating the process of the prediction mode encoding method in a screen by embodiment of this invention about the case where there are two candidate prediction modes. It is a figure which shows the hardware constitutions of the computer for performing the program recorded on the recording medium. It is a perspective view of a computer for executing a program stored in a recording medium. It is a schematic diagram which shows the method of producing | generating the prediction signal of the target block by a prior art. It is a schematic diagram when the several method which produces | generates the prediction signal of the target block by a prior art is put together into one. It is a block diagram which shows the module of an image prediction encoding program. It is a block diagram which shows the module of an image prediction decoding program.

  Hereinafter, embodiments of the present invention will be described with reference to FIGS. 1 to 12.

  FIG. 1 is a block diagram showing an image predictive coding apparatus according to an embodiment of the present invention. The image predictive coding apparatus includes an input terminal 101, a block divider 102, an inter-screen prediction signal generation method determiner 103, an inter-screen prediction signal generator 104, an intra-screen prediction signal generation method determiner 105, and an intra-screen prediction signal generator. 106, changeover switch 109, subtractor 110, converter 111, quantizer 112, inverse quantizer 113, inverse transformer 114, adder 115, frame memory 116, intra prediction mode encoder 117, entropy encoding Device 118 and an output terminal 119.

  The operation of the image predictive coding apparatus configured as described above will be described below. A moving image signal composed of a plurality of images is input to the input terminal 101. An image to be encoded is divided into a plurality of regions by the block divider 102. In the embodiment according to the present invention, the block is divided into 8 × 8 pixels, but may be divided into other block sizes or shapes. Next, a prediction signal is generated for a region to be encoded (hereinafter referred to as “target block”). In the embodiment according to the present invention, two types of prediction methods, inter-screen prediction and intra-screen prediction, are used.

  In inter-screen prediction, unlike a target image, a playback image that has been encoded and restored in the past is used as a reference image, and motion information that gives a prediction signal with the smallest error for the target block is obtained from this reference image. Further, according to circumstances, the target block may be subdivided, and the inter-screen prediction method may be determined for the subdivided small area. In this case, the most efficient division method and corresponding motion information for the entire target block are determined from various division methods. In the embodiment according to the present invention, this processing is performed by the inter-screen prediction signal generation method determiner 103, and the target block is input via the line L102 and the reference image is input via L119. As the reference image, a plurality of images encoded and restored in the past are used as the reference image. For details, see MPEG-2, 4, H. H.264 is the same as one of the methods. The motion information and the small area dividing method determined in this way are sent to the inter-screen prediction signal generator 104 via the line L104. These pieces of information are sent to the entropy encoder 118 via the line L103, encoded, and sent from the output terminal 119. The inter-screen prediction signal generator 104 acquires a reference signal from the frame memory 116 (via the line L119) based on the subregion dividing method and the motion information corresponding to each subregion, and generates a prediction signal. The inter-screen prediction signal generated in this way is sent to the next processing block via the terminal 107.

  In intra-screen prediction, an intra-screen prediction signal is generated using already reproduced pixel values adjacent to the target block in the same screen. The method for generating the intra-screen prediction signal is determined by the intra-screen prediction signal generation method determiner 105. The process of the intra-screen prediction signal generation method determiner 105 will be described later. Information (prediction mode) related to the intra prediction method determined in this way is sent to the intra prediction signal generator 106 via the line L106. Information (prediction mode) regarding these intra prediction methods is sent to the intra prediction mode encoder 117 via line L105. The processing of the intra prediction mode encoder 117 will be described later. The processing result is sent to the entropy encoder 118, encoded, and sent out from the output terminal 119. The intra-screen prediction signal generator 106 acquires adjacent already reproduced pixel signals in the same screen from the frame memory 116 (via the line L116) based on information on the prediction method, and generates a prediction signal by a predetermined method. To do. The intra-screen prediction signal generated in this way is sent to the next processing block via the terminal 108.

  With respect to the inter-screen prediction signal and the intra-screen prediction signal obtained as described above, a signal having the smallest error is selected by the changeover switch 109 and sent to the subtractor 110. However, since there is no past image for the first image, all target blocks are processed by intra prediction. In this case, the switch 109 is always connected to the terminal 108 when processing the image. Note that the intra prediction method and intra prediction mode encoding method described below can also be applied to encoding / decoding of still images.

  The subtractor 110 subtracts the prediction signal (via line L109) from the signal of the target block (via line L102) to generate a residual signal. This residual signal is subjected to discrete cosine transform by a converter 111, and each coefficient thereof is quantized by a quantizer 112. Finally, the transform coefficient quantized by the entropy encoder 118 is encoded and transmitted from the output terminal 119 together with information on the prediction method (prediction mode) and the like.

  In order to perform intra prediction or inter prediction for a subsequent target block, it is necessary to reverse-process and restore the compressed signal of the target block. That is, the quantized transform coefficient is inversely quantized by the inverse quantizer 113 and then inverse discrete cosine transformed by the inverse transformer 114 to restore the residual signal. The residual signal restored by the adder 115 is added to the prediction signal sent from the line L109 to reproduce the pixel signal of the target block and store it in the frame memory 116.

  Next, the intra-screen prediction signal generation method determiner 105 used in the present invention will be described. FIG. 2 is a schematic diagram showing a pixel extrapolation direction corresponding to the intra prediction mode used in the embodiment of the present invention. In the present embodiment, the in-screen prediction signal is generated by a total of 16 methods. The numbers in FIG. 2 are identification numbers for identifying the in-screen prediction method and are referred to as prediction mode information or unit prediction mode. For each prediction mode (No. 0 to No. 15), the already reproduced pixel signal adjacent to the target block is extrapolated in the direction indicated by the arrow in FIG. 2 to generate an in-screen prediction signal. A specific extrapolation method for the prediction modes 0 to 8 is shown in FIG. 13, and the calculation method is described in Patent Document 1. Similarly, in the prediction modes 9 to 15, the surrounding already reproduced pixel signal is linearly interpolated, and the intra-screen prediction signal is generated by duplicating the value interpolated in the direction of the arrow. In this embodiment, 16 intra prediction methods are used, but the prediction mode encoding / decoding method according to the present invention is applied even when a method for generating other prediction modes and prediction signals is used. can do.

  The intra-screen prediction signal generation method determiner 105 generates 16 intra-screen prediction signals based on these 16 prediction modes, and calculates a difference from the pixel signal of the target block sent via the line L102 for each of them. Ask. The prediction mode that gives the smallest difference value is determined as the in-screen prediction mode of the target block.

  As described above, either the intra prediction or the inter prediction is selected for the target block (by the switch 109). When the intra prediction is selected, the intra prediction mode of the target block is: Processing is performed by the intra prediction mode encoder 117. In the encoding method of the intra prediction mode according to the present invention, it is necessary to use the intra prediction mode (identification number) of the block encoded in the past, and the intra prediction mode (identification number) of the block encoded in the past. ) Is stored, the intra prediction mode encoder 117 has a storage memory (not shown).

  FIG. 3 is a flowchart showing the processing of the intra prediction mode encoder 117 according to the embodiment of the present invention. In step 310, first, a list of candidate prediction modes is generated. As a component of this list, a prediction mode of a plurality of already reproduced blocks around the target block is used. In the present embodiment, the prediction mode possessed by the peripheral reproduced blocks 410 to 450 adjacent to the target block 400 shown in FIG. 4 is a component of the candidate prediction mode list. FIG. 6A is an example of the candidate prediction mode list, and the numerical value of each square indicates the identification number of the prediction mode corresponding to each peripheral block (410 to 450). In this example, neighboring blocks (410 to 450) have different prediction modes, but when the same prediction mode appears, it is handled as one component. For example, when the blocks 410 and 420 have the same prediction mode, the number of components in the candidate prediction mode list is four instead of five. That is, the number of components in the candidate prediction mode list is five at the maximum and may be at least one. In particular, when a neighboring block adjacent to the target block is predicted to be “between” the screen, there is no intra prediction mode. In the present embodiment, mode 2 (DC prediction) is the only component of the candidate prediction mode list. In FIG. 6A, the candidate prediction mode list is arranged in the order of increasing component values, but the candidate prediction mode list may be configured in ascending order. In order to construct a candidate prediction mode list for encoding the prediction mode of the subsequent block, the intra prediction mode encoder 117 stores the prediction mode of the current target block in the storage memory described above.

  Next, in step 320, the in-screen prediction mode of the target block is compared with each component in the candidate prediction mode list to check whether there is a matching component.

  If the in-screen prediction mode of the target block is found in the candidate prediction mode list, the process proceeds to step 330. Here, it is encoded as “1”. This “1” indicates that the in-screen prediction mode of the target block is included in the candidate prediction mode list. Next, an identifier (index) for a component in the candidate prediction mode list that matches the prediction mode of the target block is encoded (step 340). In the present embodiment, 0, 1, 2, 3, 4 are assigned to the indices of each cell from the left in FIG. 6A, and 2 is encoded as an index when the prediction mode of the target block is “8”. . These indexes are encoded with a binary code. That is, (0, 1, 2, 3, 4) is assigned with each code of (0, 01, 001, 0001, 00001). Note that the last bit of the code of the maximum index may be truncated. That is, the code “00001” of “4” may be “0000”. As another method, encoding may be performed using a fixed-length code. In this case, the code length of the fixed code may be changed depending on the size (number of components) of the candidate prediction mode list. For example, when the size of the candidate prediction mode list is 4, encoding is performed with 2 bits, and when the size is 2, encoding is performed with 1 bit. Index coding based on the size (number of components) of the candidate prediction mode list is efficient.

  When the candidate prediction mode list does not include the in-screen prediction mode of the target block in step 320, the process proceeds to step 350. Here, it is encoded as “0”. This “0” indicates that the in-screen prediction mode of the target block is not included in the candidate prediction mode list. In this case, it is necessary to encode the prediction mode of the target block. In the present embodiment, the prediction mode of the target block is encoded as “REM mode”. Since it is known that there is no prediction mode of the target block in the candidate prediction mode list, it is not the original identification number of the prediction mode, but the remaining prediction modes in which each component in the candidate prediction mode list is removed. The identification number newly assigned is encoded. This will be described with reference to FIG. FIG. 6A shows each component of the candidate prediction mode list, and none corresponds to the prediction mode of the target block. Therefore, the remaining prediction modes from which these prediction modes are removed are shown in FIG. 6B, 0, 1, 2, 3, 4,. . . The result of reassigning is shown in FIG. For example, when the prediction mode of the target block is “9”, instead of encoding “9”, “9” in FIG. 6B is reassigned to “6” in FIG. 6C. Therefore, “6” is encoded as the REM mode. That is, the prediction mode of the same target block can be encoded with a smaller numerical value, that is, with a smaller number of bits. The reassignment of the identification number of the prediction mode is performed in step 360 in FIG.

  Another implementation of step 360 is shown in FIG. In step 510, the identification number of the determined intra prediction mode of the target block is substituted as the REM mode. In step 520, X is the component with the largest number among the components not yet used in the comparison in the candidate prediction mode list. In step 530, the REM mode and X are compared. If the REM mode is greater than X, the value of the REM mode is decreased by 1 in step 540. In step 550, it is confirmed whether or not there is an uncompared component in the candidate prediction list.

  As a modification of the processing of FIG. 5, the same result can be obtained by setting X as the smallest component in step 520 and “in-screen prediction mode of target block ≦ X?” In step 530. In this case, if “in-screen prediction mode of the target block ≦ X?” Becomes negative, the processing ends immediately.

  The value of the REM mode generated in this way is encoded at step 370. In the present embodiment, encoding is performed using a fixed-length code, but encoding may be performed using a variable-length code. These code lengths may be based on the number of complementary set components of the candidate prediction mode list.

  In the present embodiment, the case where the size S (number of components) of the candidate prediction mode list is 5 at maximum has been described, but S may be an arbitrary numerical value. However, the encoding device and the decoding device need to generate this list by the same method. When the candidate prediction mode list is generated in the prediction mode of the upper (420) and left (440) blocks of the target block 400 in FIG. 4, S = 2. In this case, there are two types of candidate prediction mode lists. If two neighboring blocks are predicted in the screen, the list component is 2, and if only one of the neighboring blocks is predicted in the screen, the list component is 1. FIG. 10A shows the case where the list component is 1, and FIG. 10B shows the case where the list component is 2.

  Node 80 in FIG. 10A indicates whether there is any candidate prediction mode list that matches the prediction mode of the target block. If there is no match, the REM mode is encoded (82). If there is a match (81), there is no need to encode the index because there is only one component in the list. Similarly, FIG. 10B encodes the REM mode when there is no match in the list (94). If there is a match (91), since there are two components, an index indicating which of the first and second candidates matches is encoded.

  Next, an image predictive decoding method according to the present invention will be described. FIG. 7 is a block diagram illustrating an image predictive decoding apparatus according to an embodiment of the present invention. The image predictive decoding apparatus includes an input terminal 700, a data analyzer 701, an inverse quantizer 702, an inverse transformer 703, an adder 704, a prediction signal generator 705, a frame memory 706, an intra prediction mode restoration unit 707, and an output. A terminal 708 is provided.

  The operation of the image predictive decoding apparatus configured as described above will be described below. The compressed data compressed and encoded by the above-described method is input from the input terminal 700. This compressed data includes a residual signal obtained by predicting a target block obtained by dividing an image into a plurality of blocks, and mode information regarding a prediction method. In the data analyzer 701, from the compressed data, the residual signal of the target block, information on the prediction method, quantization parameter, motion information in the case of inter-screen prediction, and the above-described intra-screen prediction for the intra-screen predicted block Extract encoding information about the mode. The residual signal of the target block and the quantization parameter (via line L701) are sent to the inverse quantizer 702, and the inverse quantization process is performed. The result is subjected to inverse discrete cosine transform by an inverse transformer 703.

  When the data analyzer 701 determines that the target block has been predicted between screens, motion information is sent to the prediction signal generator 705 via the line L709. In this case, a prediction signal is acquired from an image reproduced in the past from the frame memory 706 based on the motion information. On the other hand, when the data analyzer 701 determines that the target block has been predicted in the screen, mode information related to the prediction in the screen is sent to the prediction mode restorer 707 in the screen via the line L710, and the prediction mode in the screen is displayed. Is restored and sent to the prediction signal generator 705. The prediction signal generator 705 acquires a pixel signal reproduced in the past in the same screen from the frame memory 706 based on the intra-screen prediction mode, and generates a prediction signal. A specific method for generating the intra prediction signal has already been described with reference to FIG. Details of the in-screen prediction mode restoring unit 707 will be described later.

  The prediction signal generated by the prediction signal generator 705 is added to the restored residual signal by the adder 704 via the line L705, and the pixel signal of the target block is reproduced. The reproduced image is output via the line L704 and simultaneously stored in the frame memory 706 via the line 708.

  Next, the process of the intra prediction mode restoring unit 707 according to this embodiment will be described. The output of the intra-screen prediction mode restorer 707 is the identification number of the intra-screen prediction mode of the target block, and is output via the line L707 and is necessary for restoring the prediction mode of the subsequent block. It is stored in a memory (not shown) in the in-screen prediction mode restoring unit 707.

  FIG. 8 is a flowchart showing the processing of the intra prediction mode decoder according to the embodiment of the present invention. In step 810, a list of candidate prediction modes is first generated. As a component of this list, a prediction mode of a plurality of previously reproduced blocks (410 to 450) around the target block 400 shown in FIG. 4 is used. The specific description is the same as step 301 in FIG. The encoding device and the decoding device need to generate this candidate prediction mode list in exactly the same way.

  Next, in step 820, one bit is decoded. When this one bit is transmitted from the data analyzer 701 via the line L710, the actual decoding process is performed by the data analyzer 701. This one bit indicates whether the in-screen prediction mode of the target block is included in the candidate prediction mode list. Therefore, in step 830, it is compared whether or not this one bit is “1”. If “1”, the process proceeds to step 840. Otherwise, go to step 850.

  In step 840, since the intra prediction mode of the target block is included in the candidate prediction mode list, an identifier (index) indicating which component in the candidate prediction mode list is the intra prediction mode of the target block. Decrypt. The component of the candidate prediction mode list pointed to by the index is the prediction mode of the target block. For example, if the index is “2”, the mode identification number “8” in the third cell from the left in FIG. 6A becomes the prediction mode of the target block. In the present embodiment, this index is decoded as a binary code. As another method, when an encoding method in which the bit length of the index is determined based on the size (number of components) of the candidate prediction mode list is used, the size (number of components) of the candidate prediction mode list is used as a data analyzer. Need to be sent to 701 (line L711).

  In step 850, since the intra prediction mode of the target block is not included in the candidate prediction mode list, the value of the REM mode is decoded. In this embodiment, it is restored to a numerical value as a fixed-length code. Since the value of this REM mode is different from the identification number of the actual prediction mode (as explained in FIG. 5), after re-mapping to the actual identification number in step 860, the in-screen prediction mode of the target block To do.

  FIG. 9 shows one implementation method for returning from the REM mode to the actual prediction mode identification number. In step 910, the decoded REM mode value is substituted into the PRED mode. This PRED mode is a variable for the intra-screen prediction mode of the target block.

  In step 920, X is the component with the smallest number among the components not yet used for comparison in the candidate prediction mode list. In step 930, the PRED mode is compared with X. If the PRED mode is greater than or equal to X, 1 is added to the value of the PRED mode in step 940. In step 950, it is confirmed whether or not there is an uncompared component in the candidate prediction list. The PRED mode after the end gives the actual identification number of the prediction mode of the target block.

  Instead of the processing of FIG. 9, the complement set of FIG. 6A is created as shown in FIG. 6B, and the N + 1 (N = value of REM mode) component from the left is the prediction mode of the target block. There is also a way to do it.

  In FIG. 7, the in-screen prediction mode restoring unit 707 is depicted as an independent functional block, but may be incorporated in the data analyzer 701. In this case, the line L710 is directly connected to the prediction signal generator 705, and the intra prediction mode is sent to the prediction signal generator 705.

  In the above embodiment, encoding of prediction mode information related to intra prediction is described. However, the same encoding / decoding method can be applied to inter prediction. Information regarding the prediction mode in the case of inter-screen prediction may also be encoded / decoded using a candidate prediction mode list. In this case, the candidate prediction mode list includes, as a component, information about the inter-screen prediction mode of surrounding reproduced blocks. Furthermore, motion information in the case of inter-screen prediction can be similarly encoded / decoded. In this case, the candidate prediction mode list includes, as a component, motion information of surrounding reproduced blocks.

  An image predictive coding program for causing a computer to execute the image predictive coding method according to the present invention is provided by being stored in a recording medium. An image predictive decoding program for causing a computer to execute the image predictive decoding method according to the present invention is provided by being stored in a recording medium. Examples of the recording medium include a recording medium such as a flexible disk, CD-ROM, DVD, or ROM, or a semiconductor memory.

  FIG. 15 is a block diagram illustrating modules of an image predictive coding program that can execute the image predictive coding method. The image predictive encoding program P100 includes a region dividing module P101, a predictive signal generating module P102, a residual signal generating module P103, a signal encoding module P104, a prediction mode encoding module P105, and a storage module P106. The functions realized by the above modules being executed by a computer are the same as the functions of the image predictive coding apparatus described above. FIG. 16 is a block diagram illustrating a module of an image predictive decoding program that can execute the image predictive decoding method. The image predictive decoding program P200 includes an input module P201, a restoration module P202, a prediction mode decoding module P203, a prediction signal generation module P204, an image restoration module P205, and a storage module P206. The functions realized by the above modules being executed by a computer are the same as the functions of the image predictive decoding apparatus described above. The thus configured image predictive encoding program P100 and image predictive decoding program P200 are stored in a recording medium and executed by a computer to be described later.

  FIG. 11 is a diagram illustrating a hardware configuration of a computer for executing a program recorded in a recording medium, and FIG. 12 is a perspective view of the computer for executing a program stored in the recording medium. Examples of the computer include a DVD player, a set-top box, a mobile phone, and the like that have a CPU and perform processing and control by software.

  As shown in FIG. 11, the computer 30 includes a reading device 12 such as a flexible disk drive device, a CD-ROM drive device, and a DVD drive device, a working memory (RAM) 14 in which an operating system is resident, and a recording medium 10. A memory 16 for storing programs stored therein, a display device 18 such as a display, a mouse 20 and a keyboard 22 as input devices, a communication device 24 for transmitting and receiving data and the like, and a CPU 26 for controlling execution of the programs. And. When the recording medium 10 storing the image predictive encoding program P100 is inserted into the reading device 12, the computer 30 can access the image predictive encoding program P100 stored in the recording medium 10 from the reading device 12. The image predictive coding program P100 makes it possible to operate as an image predictive coding apparatus according to the present invention. In addition, when the recording medium 10 storing the image predictive decoding program P200 is inserted into the reading device 12, the computer 30 can access the image predictive decoding program P200 stored in the recording medium 10 from the reading device 12. The image predictive decoding program P200 can operate as an image predictive decoding apparatus according to the present invention.

  DESCRIPTION OF SYMBOLS 101 ... Input terminal, 102 ... Block divider, 103 ... Inter-screen prediction signal generation method determiner, 104 ... Inter-screen prediction signal generator, 105 ... In-screen prediction signal generation method determiner, 106 ... In-screen prediction signal generator 109 ... changeover switch, 110 ... subtractor, 111 ... converter, 112 ... quantizer, 113 ... inverse quantizer, 114 ... inverse transformer, 115 ... adder, 116 ... frame memory, 117 ... intra prediction Mode encoder 118 ... Entropy encoder 119 ... Output terminal 700 ... Input terminal 701 ... Data analyzer 702 ... Inverse quantizer 703 ... Inverse transformer 704 ... Adder 705 ... Predictive signal Generator, 706 ... Frame memory, 707 ... In-screen prediction mode restorer, 708 ... Output terminal.

Claims (2)

Input compressed image data including a residual signal generated by predicting and encoding a target block by dividing an image into a plurality of blocks and encoding information regarding a prediction mode indicating a method of generating a prediction signal of the target block Input means;
A restoration means for extracting a residual signal of the target block from the compressed image data and restoring it to a reproduction residual signal;
A prediction mode decoding means for restoring the encoding information relating to the prediction mode and generating the prediction mode;
Prediction signal generation means for generating a prediction signal of the target block based on the prediction mode;
Image restoration means for restoring the pixel signal of the target block by adding the prediction signal to the reproduction residual signal;
Storage means for storing the restored pixel signal as a reproduction pixel signal;
An image predictive decoding device comprising:
The prediction mode decoding means includes
Generating a candidate prediction mode list including, as components, prediction modes of a plurality of already reproduced blocks adjacent to the target block;
Decoding a flag indicating whether a component corresponding to the prediction mode of the target block is in the candidate prediction mode list;
When the flag indicates that “there is a corresponding component”, the index indicating the corresponding component included in the candidate prediction mode list is further decoded, and the component indicated by the index is set as the prediction mode of the target block,
When the flag indicates “no corresponding component”,
Decrypt REM mode,
Substituting the decoded value of the REM mode into the variable of the intra prediction mode of the target block,
For each component in the candidate prediction mode list, the variable is compared with the smallest component of the components not yet compared in the candidate prediction mode list, and when the variable is greater than or equal to the minimum component Repetitively executing a process of adding 1 to the value of the variable, and setting the final value of the variable as the prediction mode of the target block;
An image predictive decoding apparatus characterized by the above. An image predictive decoding method executed by an image predictive decoding device,
Input compressed image data including a residual signal generated by predicting and encoding a target block by dividing an image into a plurality of blocks and encoding information regarding a prediction mode indicating a method of generating a prediction signal of the target block An input step;
A restoration step of extracting a residual signal of the target block from the compressed image data and restoring it to a reproduction residual signal;
A prediction mode decoding step of restoring the encoding information related to the prediction mode and generating the prediction mode;
A prediction signal generation step of generating a prediction signal of the target block based on the prediction mode;
An image restoration step of restoring the pixel signal of the target block by adding the prediction signal to the reproduction residual signal;
A storage step of storing the reconstructed pixel signal as a reconstructed pixel signal, and an image predictive decoding method comprising:
The prediction mode decoding step includes:
A sub-step of generating a candidate prediction mode list including, as components, prediction modes of a plurality of already reproduced blocks adjacent to the target block;
A sub-step of decoding a flag indicating whether a component corresponding to the prediction mode of the target block is in the candidate prediction mode list;
When the flag indicates “there is a corresponding component”, an index indicating the corresponding component included in the candidate prediction mode list is further decoded, and the component indicated by the index is set as the prediction mode of the target block. Steps,
When the flag indicates “no corresponding component”,
Decrypt REM mode,
Substituting the decoded value of the REM mode into the variable of the intra prediction mode of the target block,
For each component in the candidate prediction mode list, the variable is compared with the smallest component of the components not yet compared in the candidate prediction mode list, and when the variable is greater than or equal to the minimum component Repeatedly executing a process of adding 1 to the value of the variable, and setting the final value of the variable as the prediction mode of the target block;
including,
An image predictive decoding method characterized by the above.

Images