JP2008283662A - Image prediction/encoding device, image prediction/encoding method, image prediction/encoding program, image prediction/decoding device, image prediction/decoding method, and image prediction/decoding program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image prediction/encoding device, an image prediction/encoding method, an image prediction/encoding program, an image prediction/decoding device, an image prediction/decoding method, and an image prediction/decoding program which selects a plurality of candidate prediction signals without increasing an information amount. <P>SOLUTION: A weighting device 234 and an adder 235 perform a process, averaging, for example, of pixel signals extracted by a prediction adjacent region acquisition device 232 by using a predetermined synthesis method so as to generate a comparative signal for an adjacent pixel signal for each of combinations. A comparison/selection device 236 selects a combination having a high correlation between a comparison signal generated by the weighting device 234 and an adjacent pixel signal acquired by an object adjacent region acquisition unit 233. A prediction region acquisition unit 204, a weighting unit 205, and an adder 206 generate a candidate prediction signal and process it by using the predetermined synthesis method so as to generate a prediction signal. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an image predictive encoding device, an image predictive encoding method, an image predictive encoding program, an image predictive decoding device, an image predictive decoding method, and an image predictive decoding program. The present invention relates to an image predictive encoding device, an image predictive encoding method, an image predictive encoding program, an image predictive decoding device, an image predictive decoding method, and an image predictive decoding program.

  In order to efficiently transmit and store still images and moving image data, compression coding technology is used. In the case of moving images, MPEG-1 to 4 and ITU (International Telecommunication Union) H.264. 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. In predictive coding within a screen, a prediction signal is generated using an adjacent previously reproduced image signal (reconstructed compressed image data) in the same screen as the target block, and the generated A differential signal obtained by subtracting the prediction signal from the signal of the target block is encoded. In predictive coding between screens, the adjacent reproduced image signal in the screen different from the target block is referred to, the motion is corrected, the predicted signal is generated, and the predicted signal is subtracted from the signal of the target block. The difference signal is encoded.

  For example, H.M. 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. 14 shows ITUH. 2 is a schematic diagram for explaining an intra-screen prediction method used for H.264. In FIG. 14A, a target block 1702 is a block to be encoded, and a pixel group 1701 composed of pixels A to M adjacent to the boundary of the target block 1702 is an adjacent region. It is the reproduced image signal.

  In this case, a prediction signal is generated by extending a pixel group 1701 that is an adjacent pixel immediately above the target block 1702 downward. In FIG. 14B, the already reproduced pixels (I to L) on the left of the target block 1704 are stretched to the right to generate a prediction signal. Thus, the difference between each of the nine prediction signals generated by the method shown in FIGS. 14A to 14I and the pixel signal of the target block is taken, and the one with the smallest difference value is set as the optimum prediction signal. . As described above, a prediction signal can be generated by extrapolating pixels. The above contents are described in Patent Document 1 below.

  In normal inter-picture prediction encoding, a prediction signal is generated by a method in which a signal similar to the pixel signal is searched from an already reproduced screen for a block to be encoded. Then, a motion vector, which is a spatial displacement amount between the target block and a region formed by the searched signal, and a residual signal between the pixel signal and the prediction signal of the target block are encoded. Such a method of searching for a motion vector for each block is called block matching. FIG. 15 is a schematic diagram for explaining the block matching process. Here, a procedure for generating a prediction signal will be described using the target block 1402 on the encoding target screen 1401 as an example. A screen 1403 in FIG. 15A is a screen that has already been reproduced, and a region 1404 indicated by a broken line is a region in the same spatial position as the target block 1402. In block matching, a search range 1405 surrounding the region 1404 is set, and a region 1406 where the absolute value error sum with the pixel signal of the target block 1402 is minimized is detected from the pixel signals in this search range. A pixel signal in the region 1406 becomes a prediction signal, and a displacement amount from the region 1404 to the region 1406 is detected as a motion vector. ITUH. H.264 prepares a plurality of prediction types having different block sizes for encoding motion vectors in order to cope with changes in local features of an image. ITUH. H.264 prediction types are described in Patent Document 2, for example.

In the compression encoding of moving image data, the encoding order of each frame may be arbitrary. For this reason, there are three types of coding order methods for inter-screen prediction in which a prediction signal is generated with reference to a reproduced screen. The first method refers to forward prediction in which a prediction signal is generated with reference to a previously played screen in the playback order, and the second method refers to backward prediction with reference to a future played screen in the playback order. The method 3 is bidirectional prediction in which both forward prediction and backward prediction are performed and two prediction signals are averaged. About the kind of prediction between screens, it describes in patent document 3, for example.
US Pat. No. 6,765,964 US Patent Publication No. 7003035 US Pat. No. 6,259,739

  In the prior art, since a predicted signal for each pixel is generated by copying a reproduced pixel value including distortion due to encoding (for example, quantization noise), the predicted signal also includes encoding distortion. Thus, the prediction signal contaminated by the coding distortion becomes a factor of lowering the coding efficiency such as an increase in the code amount of the residual signal and deterioration in reproduction image quality.

  Although the influence of coding distortion can be suppressed by smoothing the prediction signal, additional information for indicating a plurality of candidate prediction signals to be smoothed is required, and the amount of information increases.

  Therefore, in the present invention, an image predictive encoding device, an image predictive encoding method, an image predictive encoding program, an image predictive decoding device, an image predictive decoding method, and an image that select a plurality of candidate prediction signals without increasing the amount of information An object is to provide a predictive decoding program.

  In order to solve the above-described problem, an image predictive coding apparatus according to the present invention includes an area dividing unit that divides an input image into a plurality of areas, and a processing target of the plurality of areas divided by the area dividing unit. Prediction signal generation means for generating a prediction signal for the target pixel signal in the target area, and residual signal generation means for generating a residual signal between the prediction signal generated by the prediction signal generation means and the target pixel signal; Encoding means for encoding the residual signal generated by the residual signal generating means, wherein the prediction signal generating means is an adjacent pixel that has already been reproduced and is adjacent to a target area composed of the target pixel signal. A plurality of predicted adjacent areas having a high correlation with a target adjacent area made of a signal are searched from a search area made of an already-reproduced image, and include at least one of the searched plurality of predicted adjacent areas Two or more combinations of predicted adjacent regions are derived, and pixel signals of predicted adjacent regions belonging to the combination are processed using a predetermined synthesis method, thereby generating a comparison signal for the adjacent pixel signal for each combination. Selecting a combination having a high correlation between the comparison signal and the adjacent pixel signal, and generating one or more candidate prediction signals of the target pixel signal based on a prediction adjacent region belonging to the selected combination, A configuration is provided in which the prediction signal is generated by processing the candidate prediction signal using a predetermined synthesis method.

  According to the present invention, a plurality of predicted adjacent areas having a high correlation with the target adjacent area composed of the already reproduced adjacent pixel signals adjacent to the target area composed of the target pixel signal are searched from the search area composed of the already reproduced image, By deriving two or more combinations of arbitrary predicted adjacent areas including at least one of the searched predicted adjacent areas, and processing the pixel signals of the predicted adjacent areas belonging to the combination using a predetermined synthesis method The comparison signal for the adjacent pixel signal is generated for each combination. Then, a combination having a high correlation between the comparison signal and the adjacent pixel signal is selected, and one or more candidate prediction signals of the target pixel signal are generated based on the prediction adjacent region belonging to the selected combination, and the candidate prediction signal is preliminarily generated. A prediction signal is generated by processing using a predetermined synthesis method. A residual signal between the prediction signal generated in this way and the target pixel signal is generated, and the generated residual signal is encoded.

  As a result, a combination of candidate prediction signals suitable for smoothing can be selected without increasing the amount of information using a target adjacent region composed of already-reproduced adjacent pixel signals adjacent to the target block. A prediction signal in consideration of noise characteristics can be generated.

  Moreover, it is preferable that the prediction signal production | generation means of the image predictive decoding apparatus of this invention selects the combination with a small absolute value sum of the difference of the said comparison signal and the said adjacent pixel signal.

  According to the present invention, it is possible to select a combination of candidate prediction signals that is more suitable for smoothing by selecting a combination having a small sum of absolute values of differences between the comparison signal and the adjacent pixel signal.

  In addition, it is preferable that the prediction signal generation unit of the image predictive decoding apparatus according to the present invention generates the comparison signal by weighted averaging pixel signals of prediction adjacent regions belonging to the combination.

  According to the present invention, an appropriate comparison signal for selecting a combination of candidate prediction signals more suitable for smoothing can be obtained by generating a comparison signal by weighted averaging pixel signals of predicted adjacent regions belonging to the combination. Can be generated.

  Moreover, in the prediction signal generation means of the image predictive decoding device of the present invention, it is preferable that the combination of the prediction adjacent regions includes 2 n prediction adjacent regions in descending order of correlation with the target adjacent region, Further, the value of n is preferably an integer of 0 or more.

  According to the present invention, the combination of predicted adjacent areas includes 2 n predicted adjacent areas in descending order of the correlation with the target adjacent area, so that only addition and shift operations can be performed, which is simple in implementation. Can be configured.

  In addition, the image predictive decoding apparatus of the present invention includes a data decoding unit that decodes encoded data of a residual signal related to a target region to be processed from compressed data, and a reproduction residual from the signal decoded by the data decoding unit. A residual signal restoring means for restoring the difference signal, a prediction signal generating means for generating a prediction signal for the target pixel signal in the target region, a prediction signal generated by the prediction signal generating means, and a residual signal restoring means Reconstructed image signal generating means for generating a reconstructed image signal by adding the regenerated residual signal restored in this way, and the prediction signal generating means is an existing signal adjacent to the target area composed of the target pixel signal. A plurality of predicted adjacent areas having a high correlation with a target adjacent area composed of adjacent pixel signals for reproduction are searched from a search area composed of already reproduced images, and the plurality of predicted adjacent areas searched for are searched. By deriving two or more combinations of arbitrary predicted adjacent areas including at least one of the areas, and processing pixel signals of the predicted adjacent areas belonging to the combination using a predetermined synthesis method, for each combination A comparison signal for each of the adjacent pixel signals is generated, a combination having a large correlation between the comparison signal and the adjacent pixel signal is selected, and the target pixel signal candidate is selected based on a predicted adjacent region belonging to the selected combination. One or more prediction signals are generated, and the prediction signal is generated by processing the candidate prediction signal using a predetermined synthesis method.

  According to the present invention, the encoded data of the residual signal related to the target region to be processed is decoded from the compressed data, and the reproduction residual signal is restored from the decoded signal. On the other hand, a prediction signal for the target pixel signal in the target region to be processed is generated, and a reproduced image signal is generated by adding the generated prediction signal and the restored reproduction residual signal. Then, a plurality of predicted adjacent areas having a high correlation with the target adjacent area composed of the already reproduced adjacent pixel signals adjacent to the target area composed of the target pixel signal are searched from the search area composed of the already reproduced image. Deriving two or more combinations of arbitrary predicted adjacent areas including at least one of the searched plurality of predicted adjacent areas, and processing pixel signals of the predicted adjacent areas belonging to the combination using a predetermined synthesis method Thus, a comparison signal for the adjacent pixel signal is generated for each combination, and a combination having a large correlation between the comparison signal and the adjacent pixel signal is selected. One or more candidate prediction signals of the target pixel signal are generated based on predicted adjacent regions belonging to the selected combination, and a prediction signal is generated by processing the candidate prediction signal using a predetermined synthesis method .

  As a result, a combination of candidate prediction signals suitable for smoothing can be selected without increasing the amount of information using a target adjacent region composed of already-reproduced adjacent pixel signals adjacent to the target block. A prediction signal in consideration of noise characteristics can be generated.

  In the prediction signal generating means of the image predictive decoding apparatus of the present invention, a comparison signal for the adjacent pixel signal is generated, and a combination having a small absolute value sum of differences between the comparison signal and the adjacent pixel signal is selected. Is preferred.

  According to the present invention, it is possible to select a combination of candidate prediction signals that is more suitable for smoothing by selecting a combination having a small sum of absolute values of differences between the comparison signal and the adjacent pixel signal.

  Moreover, it is preferable that the said prediction signal production | generation means of the image predictive decoding apparatus of this invention produces | generates the said comparison signal by carrying out the weighted average of the pixel signal of the prediction adjacent area | region which belongs to the said combination.

  According to the present invention, an appropriate comparison signal for selecting a combination of candidate prediction signals more suitable for smoothing can be obtained by generating a comparison signal by weighted averaging pixel signals of predicted adjacent regions belonging to the combination. Can be generated.

  In the prediction signal generation unit of the image predictive decoding device of the present invention, it is preferable that the combination of the prediction adjacent regions includes 2 n prediction adjacent regions in descending order of correlation with the target region, It is preferable that the value of n is an integer of 0 or more.

  According to the present invention, the combination of predicted adjacent areas includes 2 n predicted adjacent areas in descending order of the correlation with the target adjacent area, so that only addition and shift operations can be performed, which is simple in implementation. Can be configured.

  In addition, the image predictive coding apparatus according to the present invention includes a region dividing unit that divides an input image into a plurality of regions, and a target pixel of a target region that is a processing target among the plurality of regions divided by the region dividing unit. A prediction signal generating means for generating a prediction signal for the signal, a residual signal generating means for generating a residual signal between the prediction signal generated by the prediction signal generating means and the target pixel signal, and the residual signal generating means Encoding means for encoding the residual signal generated by the prediction signal generation means, and the prediction signal generation means includes a target adjacent area composed of already reproduced adjacent pixel signals adjacent to the target area composed of the target pixel signal; A plurality of predicted adjacent areas having a high correlation are searched from a search area composed of already-reproduced images, and the pixel signals of N predicted areas based on the target area among the searched predicted adjacent areas are searched. Alternatively, an evaluation value for evaluating a correlation between the N candidate prediction signals is previously determined using at least two signals from the pixel signals of the prediction adjacent region or both signals of the N predicted adjacent regions searched for. When the evaluation value is smaller than a prescribed threshold value, the prediction signal is generated by processing the N candidate prediction signals using a predetermined synthesis method.

  According to the present invention, a plurality of predicted adjacent areas having a high correlation with the target adjacent area composed of the already reproduced adjacent pixel signals adjacent to the target area are searched from the search area composed of the already reproduced image, and the plurality of searched predictions Among the adjacent regions, at least two signals are used from the pixel signals of the N predicted regions based on the target region or the pixel signals of the predicted adjacent regions of the N predicted adjacent regions searched, or both signals, An evaluation value for evaluating the correlation between the N candidate prediction signals is calculated by a predetermined method. When the evaluation value is smaller than a prescribed threshold value, a prediction signal is generated by processing the N candidate prediction signals using a predetermined synthesis method. A residual signal with the target pixel signal is generated using the generated prediction signal, and the residual signal is encoded.

  Thereby, an appropriate prediction signal can be generated based on a plurality of candidate prediction signals. This is particularly effective for signals that differ only in the noise component of the candidate prediction signal but have similar signal components.

  In the image predictive coding apparatus of the present invention, the prediction signal generation unit prepares a plurality of candidates as the value of N, calculates the evaluation value for the maximum value of the N value candidates, and the evaluation value defines A prediction signal is generated by processing the N candidate prediction signals by using a predetermined synthesis method, and when the evaluation value is greater than the threshold, the N value is subtracted. Thus, it is preferable to update to the next largest value and to compare the calculation of the evaluation value with a specified threshold value again.

  According to the present invention, when the evaluation value is larger than the threshold value, the N value is subtracted to be updated to the next larger value, and the evaluation value is calculated and compared with the specified threshold value again. The number of candidate prediction signals can be selected.

  Further, in the image predictive coding apparatus of the present invention, the prediction signal generation means includes a pixel signal of a prediction area adjacent to a prediction adjacent area having the highest correlation with the target adjacent area among the N predicted adjacent areas searched for. The pixel signal of the prediction region adjacent to the prediction adjacent region where the correlation between the target adjacent region is the Nth lowest, or the pixel signal of the prediction adjacent region having the highest correlation with the target adjacent region and the target adjacent region is Nth It is preferable to calculate the absolute value sum of the difference between the pixel signals in the prediction adjacent region that is very low or the signal that combines the pixel signals, and use the absolute value sum of the differences as the evaluation value.

  According to the present invention, of the N predicted adjacent areas searched for, the prediction is the Nth lowest correlation between the pixel signal of the predicted area adjacent to the predicted adjacent area having the highest correlation with the target adjacent area and the target adjacent area. The pixel signal of the prediction adjacent region adjacent to the adjacent region, the pixel signal of the prediction adjacent region having the highest correlation with the target adjacent region, and the pixel signal of the prediction adjacent region having the lowest correlation between the target adjacent region and the pixel signal, respectively Is calculated as an evaluation value. Thereby, an appropriate evaluation value can be calculated.

  In addition, the image predictive decoding apparatus of the present invention includes a data decoding unit that decodes encoded data of a residual signal related to a target region to be processed from compressed data, and a signal obtained by decoding by the data decoding unit. A residual signal restoring means for restoring a reproduction residual signal from the prediction signal, a prediction signal generating means for generating a prediction signal for the target pixel signal in the target region, a prediction signal generated by the prediction signal generating means, and the residual signal A reproduction image signal generation unit that generates a reproduction image signal by adding the reproduction residual signal restored by the restoration unit, and the prediction signal generation unit includes a target region including the target pixel signal. A plurality of predicted adjacent areas having a high correlation with a target adjacent area consisting of adjacent already-reproduced adjacent pixel signals are searched from a search area consisting of already-reproduced images. Among the measurement adjacent regions, at least two signals are obtained from the pixel signals of the N prediction regions based on the target region and / or the pixel signals of the prediction adjacent region among the searched N prediction adjacent regions. An evaluation value for evaluating the correlation between the N candidate prediction signals is calculated by a predetermined method, and when the evaluation value is smaller than a predetermined threshold, the N candidate prediction signals are determined in advance. A configuration is provided in which a prediction signal is generated by processing using a synthesis method.

  According to the present invention, the encoded data of the residual signal related to the target region to be processed is decoded from the compressed data, and the reproduction residual signal is restored from the decoded signal. On the other hand, a prediction signal for the target pixel signal in the target region to be processed is generated, and a reproduced image signal is generated by adding the generated prediction signal and the restored reproduction residual signal.

  On the other hand, when generating a prediction signal, the following processing is performed. A plurality of predicted adjacent areas having a high correlation with the target adjacent area composed of the already reproduced adjacent pixel signals adjacent to the target area are searched from the search area including the previously reproduced image. Then, at least 2 from the pixel signals of the N prediction regions based on the target region, the pixel signals of the prediction adjacent region among the searched N prediction adjacent regions, or both signals among the plurality of prediction adjacent regions searched. An evaluation value for evaluating the correlation between the N candidate prediction signals is calculated by using a predetermined method. When the evaluation value is smaller than a prescribed threshold value, a prediction signal is generated by processing the N candidate prediction signals using a predetermined synthesis method.

  Thereby, an appropriate prediction signal can be generated based on a plurality of candidate prediction signals. This is particularly effective for signals that differ only in the noise component of the candidate prediction signal but have similar signal components.

  In the image predictive decoding device of the present invention, the prediction signal generation unit prepares a plurality of candidates as the value of N, calculates the evaluation value for the maximum value of the N value candidates, and the evaluation value is defined. A prediction signal is generated by processing the N candidate prediction signals using a predetermined synthesis method when the threshold value is smaller than the threshold value, and the N value is subtracted when the evaluation value is larger than the threshold value. Preferably, the evaluation value is updated to the next largest value, and the evaluation value is compared with a prescribed threshold value again.

  According to the present invention, when the evaluation value is larger than the threshold value, the N value is subtracted to be updated to the next larger value, and the evaluation value is calculated and compared with the specified threshold value again. The number of candidate prediction signals can be selected.

  Further, in the image predictive decoding device of the present invention, the prediction signal generation means includes a pixel signal of a prediction area adjacent to a prediction adjacent area having the highest correlation with the target adjacent area among the N predicted adjacent areas searched for. Prediction of the pixel signal of the prediction region adjacent to the prediction adjacent region having the lowest correlation with the target adjacent region, or the prediction of the Nth lowest correlation between the pixel signal of the prediction adjacent region having the highest correlation with the target adjacent region and the target adjacent region It is preferable to calculate an absolute value sum of differences between pixel signals of adjacent regions or signals obtained by combining the pixel signals, and use the sum of absolute values of the differences as the evaluation value.

  According to the present invention, of the N predicted adjacent areas searched for, the prediction is the Nth lowest correlation between the pixel signal of the predicted area adjacent to the predicted adjacent area having the highest correlation with the target adjacent area and the target adjacent area. The pixel signal of the prediction adjacent region adjacent to the adjacent region, the pixel signal of the prediction adjacent region having the highest correlation with the target adjacent region, and the pixel signal of the prediction adjacent region having the lowest correlation between the target adjacent region and the pixel signal, respectively Is calculated as an evaluation value. Thereby, an appropriate evaluation value can be calculated.

  In addition, the image predictive coding apparatus according to the present invention includes a region dividing unit that divides an input image into a plurality of regions, and a target pixel of a target region that is a processing target among the plurality of regions divided by the region dividing unit. A prediction signal generating means for generating a prediction signal for the signal, a residual signal generating means for generating a residual signal between the prediction signal generated by the prediction signal generating means and the target pixel signal, and the residual signal generating means Encoding means for encoding the residual signal generated by the prediction signal generation means, wherein the prediction signal generation means includes a plurality of predicted adjacent areas having the same shape as the target adjacent area from a search area composed of already reproduced images. Obtaining a combination of arbitrary predicted adjacent areas including two or more of the predicted adjacent areas from among the plurality of acquired predicted adjacent areas, and the prediction adjacent areas belonging to the combination are derived. Deriving two or more sets of weighting factors for weighted average of the elementary signals, and for the combination, weighted averaging of the pixel signals of the prediction adjacent region belonging to the combination using the set of multiple weighting factors, Two or more comparison signals for the adjacent pixel signal are generated, a set of weight coefficients having a high correlation between the comparison signal and the adjacent pixel signal is selected, and based on the predicted adjacent region belonging to the combination, A configuration is provided in which two or more candidate prediction signals of the target pixel signal are generated, and a prediction signal is generated by weighted averaging the candidate prediction signals using the selected set of weighting coefficients.

  According to the present invention, a plurality of predicted adjacent areas having the same shape as the target adjacent area are acquired from the search area including the already-reproduced images, and two or more predicted adjacent areas are acquired from the acquired predicted adjacent areas. Including any predictive adjacent region combination. Then, two or more sets of weighting factors for weighted averaging of the pixel signals of the predicted adjacent regions belonging to the combination are derived, and for these combinations, the prediction adjacent regions belonging to these combinations are derived using the set of multiple weighting factors. Two or more comparison signals for adjacent pixel signals are generated by weighted averaging of the pixel signals. Next, a pair of weight coefficients having a high correlation between the comparison signal and the adjacent pixel signal is selected, and two or more candidate prediction signals for the target pixel signal are generated from the already reproduced image based on the prediction adjacent region belonging to the combination. The prediction signal is generated by performing weighted averaging of the candidate prediction signal using the selected set of weighting factors. Then, a residual signal with the target pixel signal is generated using the generated prediction signal, and the residual signal is encoded. This makes it possible to select a set of weighting factors without additional information for each target block.

  The image predictive coding apparatus according to the present invention includes a region dividing unit that divides an input image into a plurality of regions, and a target pixel of a target region that is a processing target of the plurality of regions divided by the region dividing unit. Prediction signal generation means for generating a prediction signal for the signal, residual signal generation means for generating a residual signal between the prediction signal generated by the prediction signal generation means and the target pixel signal, and the residual signal generation means Encoding means for encoding the residual signal generated by the prediction signal generation means, wherein the prediction signal generation means includes a plurality of predicted adjacent areas having the same shape as the target adjacent area from a search area composed of already reproduced images. Obtaining two or more combinations of arbitrary predicted adjacent areas including at least one of the plurality of acquired predicted adjacent areas, and regarding a combination including two or more predicted adjacent areas, By deriving two or more sets of weighting factors for weighted average of pixel signals of predicted adjacent regions belonging to the matching, and weighted average of pixel signals of predicted adjacent regions using the plurality of sets of weighting factors, Two or more comparison signals for the adjacent pixel signal are generated, a set of weight coefficients having a high correlation between the comparison signal and the adjacent pixel signal is selected, and the pixels in the predicted adjacent region belonging to the combination for the plurality of combinations Two or more comparison signals for the adjacent pixel signal are generated by performing weighted averaging of the pixel signals in the predicted adjacent region using the set of selected weighting factors, and the correlation between the comparison signal and the adjacent pixel signal A combination having a high value is selected, and one candidate prediction signal of the target pixel signal is selected from the already reproduced image based on a prediction adjacent region belonging to the selected combination. And above it generates, the candidate prediction signal has a structure for generating a predictive signal by averaging weighted with the selected set of weighting factors previously for the selected combination.

  According to the present invention, a plurality of predicted adjacent areas having the same shape as the target adjacent area are obtained from the search area including the already-reproduced images, and any predicted adjacent area including at least one acquired predicted adjacent area Two or more combinations are derived. Then, for a combination including two or more predicted adjacent areas, two or more weight coefficient sets for weighted averaging of pixel signals of the predicted adjacent areas belonging to the combination are derived, and the plurality of weight coefficient sets are used. Two or more comparison signals for the adjacent pixel signals are generated by weighted averaging of the pixel signals in the predicted adjacent region. Next, a pair of weighting factors having a high correlation between the comparison signal and the adjacent pixel signal is selected, and for a plurality of combinations, a prediction neighboring region is used by using the set of weighting factors selected from the pixel signals in the prediction adjacent region belonging to the combination. Two or more comparison signals for the adjacent pixel signal are generated by weighted averaging the pixel signals in the region, and a combination having a high correlation between the comparison signal and the adjacent pixel signal is selected. Then, based on the predicted adjacent region belonging to the selected combination, one or more candidate prediction signals of the target pixel signal are generated from the already-reproduced image, and the candidate prediction signal is weighted previously selected for the selected combination. A prediction signal is generated by weighted averaging using a set of coefficients. Then, a residual signal with the target pixel signal is generated using the generated prediction signal, and the residual signal is encoded. Thereby, for each target block, it is possible to select a combination of candidate prediction signals and a set of weighting coefficients that are effective for generating a prediction signal without additional information.

  In the image predictive coding device of the present invention, the prediction signal generating means may reduce the weighting coefficient as the sum of absolute values of the differences between the pixel signals of the plurality of prediction adjacent regions belonging to the combination and the adjacent pixel signals increases. Preferably at least one of the set of weighting factors is calculated so that it is set

  According to the present invention, an appropriate weighting factor can be obtained by calculating at least one of the weighting factor sets based on the sum of absolute values of differences between the pixel signals of the plurality of predicted adjacent regions belonging to the combination and the leading pixel signal. Can be calculated.

  Further, in the image predictive coding apparatus according to the present invention, the prediction signal generation unit prepares a set of weighting factors determined in advance according to the number of prediction adjacent regions belonging to the combination, and the set of prepared weighting factors It is preferable to derive at least one set of weighting factors by

  According to the present invention, a set of weighting factors determined according to the number of predicted adjacent regions belonging to the combination is prepared in advance, and at least one set of weighting factors is derived from the set of prepared weighting factors. An appropriate set of weighting factors can be calculated.

  Further, in the image predictive coding apparatus according to the present invention, the prediction signal generation means is configured to determine a set of weighting coefficients from a sum of absolute values of differences between pixel signals of a plurality of prediction adjacent regions belonging to the combination and the adjacent pixel signals. It is preferable to prepare a table and derive at least one set of weighting factors using the correspondence table.

  According to this invention, a correspondence table is prepared in which a set of weighting factors is determined based on the sum of absolute values of differences between pixel signals of a plurality of predicted adjacent regions belonging to a combination and adjacent pixel signals. By deriving a set of weighting factors, an appropriate set of weighting factors can be calculated.

  In addition, the image predictive decoding apparatus of the present invention includes a data decoding unit that decodes encoded data of a residual signal related to a target region to be processed from compressed data, and a signal obtained by decoding by the data decoding unit. A residual signal restoring means for restoring a reproduction residual signal from the prediction signal, a prediction signal generating means for generating a prediction signal for the target pixel signal in the target region, a prediction signal generated by the prediction signal generating means, and the residual signal Reproduction image signal generation means for generating a reproduction image signal by adding the reproduction residual signal restored by the restoration means, and the prediction signal generation means has a prediction neighboring shape having the same shape as the target neighboring region A plurality of regions are acquired from each of the search regions including the already-reproduced images, and an arbitrary combination of predicted adjacent regions including two or more of the acquired plurality of predicted adjacent regions is derived. Deriving two or more sets of weighting factors for weighted averaging pixel signals of predicted adjacent regions belonging to the combination, and using the plurality of weighting factor sets for the combinations, the pixel signals of the predicted adjacent regions belonging to the combination Is used to generate two or more comparison signals for the adjacent pixel signal, select a set of weighting coefficients having a high correlation between the comparison signal and the adjacent pixel signal, and generate a prediction adjacent region belonging to the combination. And generating a prediction signal by generating two or more candidate prediction signals of the target pixel signal from the already reproduced image and performing weighted averaging of the candidate prediction signals using the selected set of weighting factors. I have.

  According to the present invention, the encoded data of the residual signal related to the target region to be processed is decoded from the compressed data, and the reproduction residual signal is restored from the decoded signal. On the other hand, a prediction signal for the target pixel signal in the target region to be processed is generated, and a reproduced image signal is generated by adding the generated prediction signal and the restored reproduction residual signal.

  On the other hand, when generating a prediction signal, the following processing is performed. That is, in the present invention, a plurality of predicted adjacent areas having the same shape as the target adjacent area are obtained from the search area composed of the already-reproduced images, and two predicted adjacent areas are obtained from the acquired predicted adjacent areas. An arbitrary combination of predicted adjacent regions including the above is derived. Then, two or more sets of weighting factors for weighted averaging of the pixel signals of the predicted adjacent regions belonging to the combination are derived, and for these combinations, the prediction adjacent regions belonging to these combinations are derived using the set of multiple weighting factors. Two or more comparison signals for adjacent pixel signals are generated by weighted averaging of the pixel signals. Next, a pair of weight coefficients having a high correlation between the comparison signal and the adjacent pixel signal is selected, and two or more candidate prediction signals for the target pixel signal are generated from the already reproduced image based on the prediction adjacent region belonging to the combination. The prediction signal is generated by performing weighted averaging of the candidate prediction signal using the selected set of weighting factors. This makes it possible to select a set of weighting factors without additional information for each target block.

  In addition, the image predictive decoding apparatus according to the present invention includes a data decoding unit that decodes encoded data of a residual signal related to a target region to be processed from compressed data, and a signal obtained by decoding by the data decoding unit. A residual signal restoring means for restoring a reproduction residual signal from the prediction signal, a prediction signal generating means for generating a prediction signal for the target pixel signal in the target region, a prediction signal generated by the prediction signal generating means, and the residual signal A reproduction image signal generation unit that generates a reproduction image signal by adding the reproduction residual signal restored by the restoration unit, and the prediction signal generation unit has the same shape as that of the target adjacent region. A plurality of regions are obtained from each of the search regions made up of already reproduced images, and any combination of predicted adjacent regions including at least one of the acquired plurality of predicted adjacent regions Two or more sets of weight coefficients for deriving a weighted average of pixel signals in the predicted adjacent areas belonging to the combination, for a combination including two or more predicted adjacent areas. Is used to generate two or more comparison signals for the adjacent pixel signals by weighting and averaging the pixel signals of the predicted adjacent region, and select a set of weighting factors having a high correlation between the comparison signals and the adjacent pixel signals. For the plurality of combinations, the pixel signals of the predicted adjacent areas belonging to the combination are weighted and averaged using the selected set of weighting coefficients, and the comparison signal for the adjacent pixel signals is calculated as 2 Two or more generations are selected, a combination having a high correlation between the comparison signal and the adjacent pixel signal is selected, and a predicted adjacent region belonging to the selected combination is selected. Then, one or more candidate prediction signals of the target pixel signal are generated from the already reproduced image, and the candidate prediction signals are weighted and averaged using the previously selected weighting factor pair for the selected combination. Thus, a configuration for generating a prediction signal is provided.

  According to the present invention, the encoded data of the residual signal related to the target region to be processed is decoded from the compressed data, and the reproduction residual signal is restored from the decoded signal. On the other hand, a prediction signal for the target pixel signal in the target region to be processed is generated, and a reproduced image signal is generated by adding the generated prediction signal and the restored reproduction residual signal.

  On the other hand, when generating a prediction signal, the following processing is performed. In other words, in the present invention, an arbitrary predicted adjacent area that includes a plurality of predicted adjacent areas having the same shape as the target adjacent area from a search area that has been reproduced, and includes at least one acquired predicted adjacent area. Two or more combinations of regions are derived. Then, for a combination including two or more predicted adjacent areas, two or more weight coefficient sets for weighted averaging of pixel signals of the predicted adjacent areas belonging to the combination are derived, and the plurality of weight coefficient sets are used. Two or more comparison signals for the adjacent pixel signals are generated by weighted averaging of the pixel signals in the predicted adjacent region. Next, a pair of weighting factors having a high correlation between the comparison signal and the adjacent pixel signal is selected, and for a plurality of combinations, a prediction neighboring region is used by using the set of weighting factors selected from the pixel signals in the prediction adjacent region belonging to the combination. Two or more comparison signals for the adjacent pixel signal are generated by weighted averaging the pixel signals in the region, and a combination having a high correlation between the comparison signal and the adjacent pixel signal is selected. Then, based on the predicted adjacent region belonging to the selected combination, one or more candidate prediction signals of the target pixel signal are generated from the already-reproduced image, and the candidate prediction signal is weighted previously selected for the selected combination. A prediction signal is generated by weighted averaging using a set of coefficients. Thereby, for each target block, it is possible to select a combination of candidate prediction signals and a set of weighting coefficients that are effective for generating a prediction signal without additional information.

  Further, in the image predictive decoding device, the prediction signal generation means sets a smaller weighting coefficient as the absolute value sum of the differences between the pixel signals of the plurality of prediction adjacent regions belonging to the combination and the adjacent pixel signals increases. Preferably, at least one of the set of weighting factors is calculated.

  According to the present invention, an appropriate weight is obtained by calculating at least one of the sets of weighting factors based on the sum of absolute values of the differences between the pixel signals of the plurality of predicted adjacent regions belonging to the combination and the preceding pixel signal. A set of coefficients can be calculated.

  Further, in the image predictive decoding apparatus, the prediction signal generation unit prepares in advance a set of weighting factors determined according to the number of prediction adjacent regions belonging to the combination, and at least one set of weighting factors is provided. It is preferable to derive a set of weighting factors.

  According to the present invention, a set of weighting factors determined according to the number of predicted adjacent regions belonging to the combination is prepared in advance, and at least one set of weighting factors is derived from the set of prepared weighting factors. An appropriate set of weighting factors can be calculated.

  Further, in the image predictive decoding apparatus, the prediction signal generation means prepares a correspondence table in which a set of weighting coefficients is determined from a sum of absolute values of differences between pixel signals of a plurality of prediction adjacent regions belonging to a combination and the adjacent pixel signals, It is preferable to derive at least one set of weighting factors using the correspondence table.

  According to this invention, a correspondence table is prepared in which a set of weighting factors is determined based on the sum of absolute values of differences between the pixel signals of a plurality of predicted neighboring regions belonging to the combination and the neighboring pixel signals, and at least 1 is used using the correspondence table. By deriving one set of weighting factors, an appropriate set of weighting factors can be calculated.

  By the way, the present invention can be described as the invention of the image predictive encoding device or the image predictive decoding device as described above, and as described below, the image predictive encoding method, the image predictive decoding method, the image predictive encoding program, It can also be described as an invention of an image predictive decoding program. These are substantially the same inventions only in different categories, and exhibit the same actions and effects.

  That is, the image predictive coding method of the present invention includes a region dividing step for dividing an input image into a plurality of regions, and a target pixel of a target region that is a processing target of the plurality of regions divided by the region dividing step. A prediction signal generation step for generating a prediction signal for the signal, a residual signal generation step for generating a residual signal between the prediction signal generated by the prediction signal generation step and the target pixel signal, and the residual signal generation step An encoding step for encoding the residual signal generated by the step, wherein the prediction signal generating step includes a target adjacent region composed of an already reproduced adjacent pixel signal adjacent to the target region composed of the target pixel signal, and A plurality of predicted adjacent areas having a high correlation of Two or more combinations of arbitrary predicted adjacent regions including the same, and processing the pixel signals of the predicted adjacent regions belonging to the combination using a predetermined synthesis method, so that for each combination, A comparison signal is generated, a combination having a high correlation between the comparison signal and the adjacent pixel signal is selected, and one candidate prediction signal of the target pixel signal is selected based on a prediction adjacent region belonging to the selected combination. A configuration for generating a prediction signal by generating the above and processing the candidate prediction signal using a predetermined synthesis method is provided.

  The image predictive decoding method of the present invention also includes a data decoding step for decoding encoded data of a residual signal relating to a target region to be processed from compressed data, and a reproduction residual from the signal decoded by the data decoding step. A residual signal restoring step for restoring the difference signal, a prediction signal generating step for generating a prediction signal for the target pixel signal in the target region, a prediction signal generated by the prediction signal generating step, and the residual signal recovery step. A reproduction image signal generation step for generating a reproduction image signal by adding the reproduction residual signal restored in step S3, and the prediction signal generation step includes an existing region adjacent to the target region including the target pixel signal. A plurality of predicted adjacent areas having a high correlation with the target adjacent area composed of the adjacent pixel signals to be reproduced are searched from the search area composed of the already reproduced images. Then, two or more combinations of arbitrary predicted adjacent areas including at least one of the searched plurality of predicted adjacent areas are derived, and a pixel signal of the predicted adjacent area belonging to the combination is determined using a predetermined synthesis method By processing, a comparison signal for the adjacent pixel signal is generated for each combination, a combination having a large correlation between the comparison signal and the adjacent pixel signal is selected, and a predicted adjacent region belonging to the selected combination is selected. Based on this, one or more candidate prediction signals for the target pixel signal are generated, and the prediction signal is generated by processing the candidate prediction signal using a predetermined synthesis method.

  In addition, the image predictive coding program of the present invention includes a region dividing module that divides an input image into a plurality of regions, and a target pixel of a target region that is a processing target of the plurality of regions divided by the region dividing module. A prediction signal generation module that generates a prediction signal for the signal, a residual signal generation module that generates a residual signal between the prediction signal generated by the prediction signal generation module and the target pixel signal, and the residual signal generation module An encoding module that encodes the residual signal generated by the prediction signal generation module, and the prediction signal generation module includes a target adjacent region composed of an already reproduced adjacent pixel signal adjacent to the target region composed of the target pixel signal; A plurality of predicted adjacent areas having a high correlation of the search area are searched from a search area composed of already reproduced images. By deriving two or more combinations of arbitrary predicted adjacent regions including at least one of them, and processing the pixel signals of the predicted adjacent regions belonging to the combination using a predetermined synthesis method, for each combination A comparison signal for each adjacent pixel signal is generated, a combination having a high correlation between the comparison signal and the adjacent pixel signal is selected, and candidate prediction of the target pixel signal is performed based on a predicted adjacent region belonging to the selected combination. A configuration is provided in which one or more signals are generated, and the prediction signal is generated by processing the candidate prediction signal using a predetermined synthesis method.

  In addition, the image predictive decoding program of the present invention includes a data decoding module that decodes encoded data of a residual signal related to a target region to be processed from compressed data, and a reproduction residual from the signal decoded by the data decoding module. A residual signal restoration module that restores a difference signal, a prediction signal generation module that generates a prediction signal for a target pixel signal in the target region, a prediction signal generated by the prediction signal generation module, and the residual signal recovery module. A reconstructed image signal generating module that generates a reconstructed image signal by adding the reconstructed reproduction residual signal, and the prediction signal generating module includes an existing signal adjacent to the target region including the target pixel signal. A plurality of predicted adjacent areas having a high correlation with the target adjacent area composed of adjacent pixel signals to be reproduced are already reproduced images. Two or more combinations of arbitrary predicted adjacent areas including at least one of the searched plurality of predicted adjacent areas are derived, and pixel signals of the predicted adjacent areas belonging to the combination are determined in advance. By generating a comparison signal for the adjacent pixel signal for each combination, selecting a combination having a large correlation between the comparison signal and the adjacent pixel signal, and processing the selected combination. A prediction signal is generated by generating one or more candidate prediction signals of the target pixel signal based on a prediction adjacent region belonging to the target pixel signal and processing the candidate prediction signal using a predetermined synthesis method Yes.

  The image predictive coding method according to the present invention includes a region dividing step of dividing an input image into a plurality of regions, and a target pixel of a target region that is a processing target of the plurality of regions divided by the region dividing step. A prediction signal generation step for generating a prediction signal for the signal, a residual signal generation step for generating a residual signal between the prediction signal generated by the prediction signal generation step and the target pixel signal, and the residual signal generation step An encoding step for encoding the residual signal generated by the step, wherein the prediction signal generating step includes a target adjacent region composed of an already reproduced adjacent pixel signal adjacent to the target region composed of the target pixel signal, and A plurality of predicted adjacent areas having a high correlation between the target areas of the plurality of predicted adjacent areas searched for, The N candidate prediction signals are obtained by using at least two signals from the pixel signals of the N prediction regions based on the pixel signals of the prediction adjacent regions or both of the N prediction adjacent regions searched for. An evaluation value for evaluating a correlation between the N candidate prediction signals is calculated by a predetermined method, and when the evaluation value is smaller than a predetermined threshold, the N candidate prediction signals are processed by using a predetermined synthesis method. A configuration for generating a signal is provided.

  The image predictive decoding method of the present invention includes a data decoding step for decoding encoded data of a residual signal related to a target region to be processed from compressed data, and a signal obtained by decoding by the data decoding step. A residual signal restoration step for restoring a reproduction residual signal from the prediction signal, a prediction signal generation step for generating a prediction signal for the target pixel signal in the target region, a prediction signal generated by the prediction signal generation step, and the residual signal A reproduction image signal generation step of generating a reproduction image signal by adding the reproduction residual signal restored in the restoration step, and the prediction signal generation step is performed on a target region composed of the target pixel signal. A plurality of predicted adjacent regions having a high correlation with the target adjacent region made up of adjacent pre-reproduced adjacent pixel signals are searched for from the pre-reproduced image The pixel signals of N prediction regions based on the target region and / or the pixel signals of the prediction adjacent region among the searched N prediction adjacent regions among the plurality of predicted adjacent regions searched from An evaluation value for evaluating the correlation between the N candidate prediction signals is calculated by a predetermined method using at least two signals from the signals of N, and when the evaluation value is smaller than a predetermined threshold, the N A configuration is provided in which a prediction signal is generated by processing each candidate prediction signal using a predetermined synthesis method.

  The image predictive coding method according to the present invention includes a region dividing step of dividing an input image into a plurality of regions, and a target pixel of a target region that is a processing target of the plurality of regions divided by the region dividing step. A prediction signal generation step for generating a prediction signal for the signal, a residual signal generation step for generating a residual signal between the prediction signal generated by the prediction signal generation step and the target pixel signal, and the residual signal generation step An encoding step for encoding the residual signal generated by the prediction signal generation step, wherein the prediction signal generation step includes a predicted adjacent region having the same shape as the target adjacent region adjacent to the target region, which is an already reproduced image. Arbitrary combinations of predicted adjacent areas, each of which is acquired from a plurality of search areas and includes two or more predicted adjacent areas among the acquired predicted adjacent areas Two or more sets of weighting factors for deriving a weighted average of pixel signals of prediction adjacent regions belonging to the combination, and using the plurality of sets of weighting factors for the combination, Two or more comparison signals for the adjacent pixel signal are generated by performing weighted averaging of the pixel signals in the adjacent region, and a set of weight coefficients having a high correlation between the comparison signal and the adjacent pixel signal is selected, and the combination is selected. A prediction signal is generated by generating two or more candidate prediction signals of the target pixel signal from the already-reproduced image based on a predicted adjacent region to which the signal belongs, and performing weighted averaging of the candidate prediction signals using the selected set of weighting coefficients. The structure which produces | generates is provided.

  The image predictive coding method according to the present invention includes a region dividing step of dividing an input image into a plurality of regions, and a target pixel of a target region that is a processing target of the plurality of regions divided by the region dividing step. A prediction signal generation step for generating a prediction signal for the signal, a residual signal generation step for generating a residual signal between the prediction signal generated by the prediction signal generation step and the target pixel signal, and the residual signal generation step An encoding step for encoding the residual signal generated by the prediction signal generation step, wherein the prediction signal generation step includes a predicted adjacent region having the same shape as the target adjacent region adjacent to the target region, which is an already reproduced image. Two or more combinations of arbitrary predicted adjacent areas including at least one of the acquired multiple predicted adjacent areas are obtained from each search area. Then, for a combination including two or more predicted adjacent regions, two or more sets of weighting factors for weighted averaging pixel signals of the predicted adjacent regions belonging to the combination are derived, and the plurality of sets of weighting factors are used. Two or more comparison signals for the adjacent pixel signal are generated by performing weighted averaging of the pixel signals in the predicted adjacent region, a set of weight coefficients having a high correlation between the comparison signal and the adjacent pixel signal is selected, Two or more comparison signals for the adjacent pixel signals are obtained by weighted averaging the pixel signals of the predicted adjacent areas belonging to the combination using the set of weighting factors selected above. Generating and selecting a combination having a high correlation between the comparison signal and the adjacent pixel signal, and based on a predicted adjacent region belonging to the selected combination One or more candidate prediction signals of the target pixel signal are generated from the already-reproduced image, and the candidate prediction signals are predicted by weighted averaging using the previously selected combination of weight coefficients for the selected combination A configuration for generating a signal is provided.

  The image predictive decoding method of the present invention includes a data decoding step for decoding encoded data of a residual signal related to a target region to be processed from compressed data, and a signal obtained by decoding by the data decoding step. A residual signal restoration step for restoring a reproduction residual signal from the prediction signal, a prediction signal generation step for generating a prediction signal for the target pixel signal in the target region, a prediction signal generated by the prediction signal generation step, and the residual signal A reproduction image signal generation step of generating a reproduction image signal by adding the reproduction residual signal restored in the restoration step, and the prediction signal generation step includes a target adjacent region adjacent to the target region A plurality of predicted adjacent areas having the same shape as the search area including the already-reproduced images. Deriving a combination of arbitrary predicted adjacent regions including two or more, deriving two or more sets of weighting factors for weighted averaging pixel signals of predicted adjacent regions belonging to the combination, and for the combination, the plurality of weights Two or more comparison signals for the adjacent pixel signal are generated by performing weighted averaging of the pixel signals in the predicted adjacent region belonging to the combination using a set of coefficients, and the correlation between the comparison signal and the adjacent pixel signal is high A set of weighting factors is selected, two or more candidate prediction signals of the target pixel signal are generated from the already-reproduced image based on the prediction adjacent region belonging to the combination, and the candidate prediction signals are selected from the selected weighting factors. A configuration is provided in which a prediction signal is generated by weighted averaging using a set.

  The image predictive decoding method of the present invention includes a data decoding step for decoding encoded data of a residual signal related to a target region to be processed from compressed data, and a signal obtained by decoding by the data decoding step. A residual signal restoration step for restoring a reproduction residual signal from the prediction signal, a prediction signal generation step for generating a prediction signal for the target pixel signal in the target region, a prediction signal generated by the prediction signal generation step, and the residual signal A reproduction image signal generation step of generating a reproduction image signal by adding the reproduction residual signal restored in the restoration step, and the prediction signal generation step includes a target adjacent region adjacent to the target region A plurality of predicted adjacent areas having the same shape as the search area including the already-reproduced images. A set of weighting factors for deriving two or more combinations of arbitrary predicted adjacent areas including at least one and weighting and averaging pixel signals of predicted adjacent areas belonging to the combination of combinations including two or more predicted adjacent areas And two or more comparison signals for the adjacent pixel signal are generated by calculating a weighted average of the pixel signals in the prediction adjacent region using the plurality of sets of weighting factors, and the comparison signal and the adjacent signal are generated. Select a set of weighting factors having a high correlation with the pixel signal, and use the set of weighting factors selected above for the pixel signals of the prediction adjacent region belonging to the combination for the plurality of combinations, Two or more comparison signals for the adjacent pixel signal are generated by weighted averaging, and the correlation between the comparison signal and the adjacent pixel signal is high. And generating one or more candidate prediction signals of the target pixel signal from the already-reproduced image based on the prediction adjacent region belonging to the selected combination, and the candidate prediction signal is generated for the selected combination. A configuration is provided in which a prediction signal is generated by performing weighted averaging using a previously selected set of weighting factors.

  In addition, the image predictive coding program of the present invention includes a region dividing module that divides an input image into a plurality of regions, and a target pixel of a target region that is a processing target of the plurality of regions divided by the region dividing module. A prediction signal generation module that generates a prediction signal for the signal, a residual signal generation module that generates a residual signal between the prediction signal generated by the prediction signal generation module and the target pixel signal, and the residual signal generation module An encoding module that encodes the residual signal generated by the prediction signal generation module, and the prediction signal generation module includes a target adjacent region composed of an already reproduced adjacent pixel signal adjacent to the target region composed of the target pixel signal; A plurality of predicted adjacent areas having a high correlation of the search area are searched from a search area composed of already reproduced images. Among them, at least two signals are used from the pixel signals of the N prediction regions based on the target region or the pixel signals of the prediction adjacent region among the N predicted adjacent regions searched for, or both signals, An evaluation value for evaluating the correlation between the N candidate prediction signals is calculated by a predetermined method. When the evaluation value is smaller than a predetermined threshold, a synthesis method for determining the N candidate prediction signals is determined. It has a configuration for generating a prediction signal by processing it.

  Also, the image predictive code decoding program of the present invention is obtained by decoding the encoded data of the residual signal related to the target region to be processed from the compressed data, and the data decoding module decoding the data. A residual signal restoration module that restores a reproduction residual signal from the signal, a prediction signal generation module that generates a prediction signal for the target pixel signal in the target region, a prediction signal generated by the prediction signal generation module, and the residual A reproduction image signal generation module that generates a reproduction image signal by adding the reproduction residual signal restored by the signal restoration module, and the prediction signal generation module includes a target region including the target pixel signal. A plurality of predicted adjacent areas having a high correlation with the target adjacent area consisting of already reproduced adjacent pixel signals adjacent to Search from a search area composed of a reproduced image, and among the searched predicted adjacent areas, pixel signals of N predicted areas based on the target area or predicted adjacent areas of the searched N predicted adjacent areas An evaluation value for evaluating the correlation between the N candidate prediction signals is calculated by a predetermined method using at least two signals from the pixel signals or both of the pixel signals, and the evaluation value is calculated from a predetermined threshold value. When it is small, the N candidate prediction signals are processed by using a predetermined synthesis method to generate a prediction signal.

  The image predictive coding program according to the present invention includes an area division module that divides an input image into a plurality of areas, and a target pixel of a target area that is a processing target of the plurality of areas divided by the area division module. A prediction signal generation module that generates a prediction signal for the signal, a residual signal generation module that generates a residual signal between the prediction signal generated by the prediction signal generation module and the target pixel signal, and the residual signal generation module An encoding module that encodes the residual signal generated by the prediction signal generation module, wherein the prediction signal generation module includes a predicted adjacent region having the same shape as the target adjacent region adjacent to the target region, which is an already reproduced image. A plurality of each of the search areas are obtained, and two or more of the obtained prediction adjacent areas are included in the obtained prediction adjacent areas. A plurality of combinations of weighting factors for deriving two or more weight coefficient sets for weighted averaging of pixel signals of the prediction adjacent regions belonging to the combination. Two or more comparison signals for the adjacent pixel signal are generated by performing weighted averaging of the pixel signals in the prediction adjacent region belonging to the combination, and a set of weighting factors having a high correlation between the comparison signal and the adjacent pixel signal is selected. And generating two or more candidate prediction signals of the target pixel signal from the already-reproduced image based on the prediction adjacent region belonging to the combination, and weighting average the candidate prediction signals using the selected set of weight coefficients By doing so, a configuration for generating a prediction signal is provided.

  The image predictive coding program according to the present invention includes an area division module that divides an input image into a plurality of areas, and a target pixel of a target area that is a processing target of the plurality of areas divided by the area division module. A prediction signal generation module that generates a prediction signal for the signal, a residual signal generation module that generates a residual signal between the prediction signal generated by the prediction signal generation module and the target pixel signal, and the residual signal generation module An encoding module that encodes the residual signal generated by the prediction signal generation module, wherein the prediction signal generation module includes a predicted adjacent region having the same shape as the target adjacent region adjacent to the target region, which is an already reproduced image. Each of a plurality of search neighboring areas is acquired, and includes at least one of the obtained plurality of predicted neighboring areas. Two or more combinations are derived, and for a combination including two or more predicted adjacent regions, two or more sets of weighting factors for weighted averaging pixel signals of predicted adjacent regions belonging to the combination are derived, Two or more comparison signals for the adjacent pixel signal are generated by performing weighted averaging of the pixel signals in the predicted adjacent region using a set of weighting factors, and a weight coefficient having a high correlation between the comparison signal and the adjacent pixel signal is generated. For the plurality of combinations, the pixel signal of the predicted adjacent region belonging to the combination is weighted and averaged using the set of weighting factors selected above, thereby calculating the adjacent pixel signal. Generate two or more comparison signals for, select a combination having a high correlation between the comparison signal and the adjacent pixel signal, and belong to the selected combination Based on the measurement adjacent region, one or more candidate prediction signals of the target pixel signal are generated from the already-reproduced image, and the candidate prediction signal is used using a set of weight coefficients previously selected for the selected combination. And generating a prediction signal by weighted averaging.

  The image predictive decoding program according to the present invention includes a data decoding module that decodes encoded data of a residual signal related to a target region to be processed from compressed data, and a signal obtained by decoding by the data decoding module. A residual signal restoration module for restoring a reproduction residual signal from the prediction signal, a prediction signal generation module for generating a prediction signal for a target pixel signal in the target region, a prediction signal generated by the prediction signal generation module, and the residual signal A reproduction image signal generation module that generates a reproduction image signal by adding the reproduction residual signal restored by the restoration module, and the prediction signal generation module includes a target adjacent region adjacent to the target region A plurality of predicted adjacent areas having the same shape as the search area including the already reproduced images, Deriving a combination of arbitrary predicted adjacent areas including two or more predicted adjacent areas, and deriving two or more sets of weighting factors for weighted averaging of pixel signals of the predicted adjacent areas belonging to the combination, For the combination, two or more comparison signals for the adjacent pixel signal are generated by performing weighted averaging of the pixel signals in the prediction adjacent region belonging to the combination using the set of the plurality of weighting factors, and the comparison signal and the adjacent signal are generated. Selecting a set of weighting factors having a high correlation with the pixel signal, generating two or more candidate prediction signals of the target pixel signal from the already-reproduced image based on a prediction adjacent region belonging to the combination, and the candidate prediction signal Is configured to generate a prediction signal by performing weighted averaging using the set of selected weighting factors.

  The image predictive decoding program according to the present invention includes a data decoding module that decodes encoded data of a residual signal related to a target region to be processed from compressed data, and a signal obtained by decoding by the data decoding module. A residual signal restoration module for restoring a reproduction residual signal from the prediction signal, a prediction signal generation module for generating a prediction signal for a target pixel signal in the target region, a prediction signal generated by the prediction signal generation module, and the residual signal A reproduction image signal generation module that generates a reproduction image signal by adding the reproduction residual signal restored by the restoration module, and the prediction signal generation module includes a target adjacent region adjacent to the target region A plurality of predicted adjacent areas having the same shape as the search area including the already reproduced images, Two or more combinations of arbitrary predicted adjacent areas including at least one predicted adjacent area are derived, and pixel combinations of the predicted adjacent areas belonging to the combination are weighted average for combinations including two or more predicted adjacent areas. Two or more sets of weighting factors for deriving are derived, and two or more comparison signals for the neighboring pixel signals are generated by weighted averaging pixel signals in the predicted neighboring region using the plurality of sets of weighting factors. , Selecting a set of weighting factors having a high correlation between the comparison signal and the adjacent pixel signal, and using the set of weighting factors selected above for the pixel signals of the predicted adjacent region belonging to the combination for the plurality of combinations Two or more comparison signals for the adjacent pixel signal are generated by performing weighted averaging of the pixel signals in the predicted adjacent region, and the comparison signal and the adjacent pixel signal are generated. A combination having a high correlation with the selected combination, and generating one or more candidate prediction signals of the target pixel signal from the already-reproduced image based on a prediction adjacent region belonging to the selected combination, A configuration is provided in which a prediction signal is generated by performing weighted averaging on a selected combination using a previously selected set of weighting factors.

  According to the present invention, it is possible to select a combination of candidate prediction signals suitable for smoothing without increasing the amount of information, using a target adjacent region composed of already-reproduced adjacent pixel signals adjacent to the target region. A prediction signal that takes local noise characteristics into consideration can be generated.

  The present invention can be readily understood by considering the following detailed description with reference to the accompanying drawings shown for the embodiments. Subsequently, embodiments of the present invention will be described with reference to the accompanying drawings. Where possible, the same parts are denoted by the same reference numerals, and redundant description is omitted.

<First embodiment>
FIG. 1 is a block diagram illustrating an image predictive coding apparatus 100 that can execute an image predictive coding method according to an embodiment of the present invention. The image predictive coding apparatus 100 includes an input terminal 101, a block divider 102 (region dividing means), a prediction signal generator 103 (predicted signal generating means), a frame memory 104, a subtractor 105 (residual signal generating means), It includes a converter 106 (encoding means), a quantizer 107 (encoding means), an inverse quantizer 108, an inverse transformer 109, an adder 110, an entropy encoder 111, and an output terminal 112. Yes. The converter 106 and the quantizer 107 function as encoding means.

  The configuration of the image predictive encoding device 100 configured as described above will be described below.

  The input terminal 101 is a terminal for inputting a moving image signal composed of a plurality of images.

  The block divider 102 divides an image to be encoded, which is a moving image signal input from the input terminal 101, into a plurality of regions. 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.

  The prediction signal generator 103 is a part that generates a prediction signal for a target region (target block) to be encoded. Specific processing of the prediction signal generator 103 will be described later.

  The subtractor 105 uses the prediction signal generator 103 input via the line L103 from the pixel signal indicated by the target region obtained by the division by the block divider 102 input via the line L102. This is a part for generating a residual signal by subtracting the generated prediction signal. The subtractor 105 outputs the residual signal obtained by the subtraction to the converter 106 via the line L105.

  The converter 106 is a part that performs discrete cosine transform on the residual signal obtained by subtraction. The quantizer 107 is a part that quantizes the transform coefficient that has been discrete cosine transformed by the transformer 106. The entropy encoder 111 compresses and encodes the transform coefficient quantized by the quantizer 107, and outputs the compressed and compressed data via the line L111. The output terminal 112 outputs compressed data that is information input from the entropy encoder 111 to the outside.

  The inverse quantizer 108 inversely quantizes the quantized transform coefficient, and the inverse transformer 109 performs inverse discrete cosine transform to restore the encoded residual signal. The adder 110 adds the restored residual signal and the prediction signal sent from the line L103, reproduces the signal of the target block, obtains a reproduced image signal, and stores the reproduced image signal in the frame memory 104. Remember. In this embodiment, the converter 106 and the inverse converter 109 are used. However, other conversion processes may be used in place of these converters, and the converter 106 and the inverse converter 109 are not necessarily essential. Thus, in order to perform intra prediction or inter prediction for the subsequent target region, the compressed pixel signal of the target region is restored by inverse processing and stored in the frame memory 104.

  Next, the prediction signal generator 103 will be described. The prediction signal generator 103 is a part that generates a prediction signal for a target region (hereinafter referred to as a target block) to be encoded. In this embodiment, three types of prediction methods are used. That is, the prediction signal generator 103 generates a prediction signal using at least one or both of an inter-screen prediction method and an intra-screen prediction method described later.

  Processing of the prediction signal generator 103 in this embodiment will be described. FIG. 2 is a block diagram showing the prediction signal generator 103 used in the image predictive coding apparatus 100 according to the present embodiment. The template matching unit 201, the coordinate information memory 202, the candidate prediction signal combination determiner 203, and the prediction The area acquisition unit 204, the weighting unit 205, and the adder 206 are included.

  In the template matching unit 201, an image signal (reproduced image signal) that has already been reproduced in the past processing is input from the frame memory 104 via the line L104, and prediction signal candidates (candidate prediction signals) for the target pixel signal in the target region are obtained. A plurality of searches are performed by template matching described later, and coordinate information for accessing the searched candidate prediction signal is output to the coordinate information memory 202 via the line L201a. At the same time, difference value data indicating the relationship between the target region and each candidate prediction signal (corresponding to a sum of absolute error values (SAD (sumof absolute difference) described later)) is combined with a candidate prediction signal combination determiner 203 via a line L201b. Output to.

  The candidate prediction signal combination determiner 203 uses the difference value data input via the line L201b to set a plurality of candidate prediction signal combinations. Then, a candidate prediction signal combination is determined using the pixel signal input from the frame memory via the line L104 according to the coordinate information input from the coordinate information memory 202 via the line L202a, and the candidate prediction signal combination information is obtained. It outputs to the prediction area | region acquisition device 204 via the line L203.

  The prediction area acquisition unit 204 acquires coordinate information of candidate prediction signals belonging to this combination via line L202b according to the combination information of candidate prediction signals input via line L203. Then, the prediction region acquisition unit 204 acquires a candidate prediction signal corresponding to the acquired coordinate information from the frame memory 104 via the line L104 and outputs it to the weighting unit 205 as needed. The weighting unit 205 multiplies each candidate prediction signal input via the line L204 by a weighting coefficient, and outputs the result to the adder 206 via the line L205. The adder 206 sequentially adds the weighted candidate prediction signals and outputs the result as a prediction signal to the subtracter 105 in FIG. 1 via the line L103. As for the operation of the weighting unit, for example, there is a method of multiplying each candidate prediction signal by 1 / N when the number of candidate prediction signals is N, but other methods may be used.

  Furthermore, each configuration of the template matching unit 201, the candidate prediction signal combination determining unit 203, and the prediction region acquiring unit 204 will be described in detail. First, processing contents in the template matching unit 201 will be described. The template matching unit 201 accesses the reproduced image signal stored in the frame memory 104 via the line L104 and performs matching processing. Here, this matching process will be described. 3 and 4 are schematic diagrams for explaining the template matching process according to the embodiment of the present invention. Here, a process of generating a candidate prediction signal for the target block 402 will be described.

  First, a “target adjacent region” (also referred to as a template) and a “search region” are set for a target block by a predetermined method. In FIG. 3, a part (or all) of the reproduced images that are in contact with the target block 402 and are reproduced before and within the same screen are set as the search area 401. Here, the target block 402 is a small block of 4 × 4 pixels obtained by dividing an encoding target block made up of 8 × 8 pixels, but may be divided into other block sizes or shapes, or may not be divided. Also good.

  Further, in FIG. 4, a part of the reproduced image shown on the screen 411 different from the target block 402 is set as the search area 417. Further, a search area may be provided on each of a plurality of screens different from the target block 402. As the “target adjacent region”, the already reproduced pixel group (reverse L character region) 403 adjacent to the target block 402 is used. In this embodiment, the search area is set in two screens (FIGS. 3 and 4), but it may be set only in the same screen as the target block (FIG. 3). You may set only in a different screen (FIG. 4).

  As shown in FIG. 3, the search area 401 and the target block 402 do not need to touch each other, and the search area 401 may not touch the target block 402 at all. In addition, as shown in FIG. 4, it is not necessary to limit the search area to be set on one screen different from the target block (only the screen 411). Each of the search areas may be set in order to include future frames in order.

  In addition, the target adjacent area 403 may be in contact with the target block 402 even at least one pixel. Here, the shape of the target adjacent region is an inverted L character, but is not limited thereto. Therefore, if the target adjacent region is composed of reproduced pixels around the target block 402, the shape and the number of pixels may be arbitrary as long as they are determined in advance. The shape and size of the template in sequence units, frame units, or block units. (Number of pixels) may be encoded.

  The template matching unit 201 calculates the sum of absolute error values (SAD) between corresponding pixels between the search region 401 and the search region 417 or a pixel group having the same shape as the target adjacent region 403 in one of the search regions. ), And search for M regions having a small SAD and make them “predicted adjacent regions”. The search accuracy may be in units of integer pixels, or a pixel with decimal precision such as 1/2 pixel or 1/4 pixel may be generated and the search may be performed with decimal pixel precision. The value of M may be an arbitrary value as long as it is set in advance. As shown in FIGS. 3 and 4, M = 6, and the predicted adjacent regions 404a, 404b, 404c, 414a, 414b, and 414c are searched. Alternatively, the value of M may be determined by searching a region where the SAD is smaller than a certain threshold without determining the number of predicted adjacent regions to be searched, or the smaller of the number of regions smaller than the threshold and the set number may be set to M It may be a value. At this time, the threshold value may be applied not to the SAD value itself but to a difference value between the minimum SAD and the second and subsequent SADs. In the latter case, the template matching unit 201 can search many prediction adjacent regions without changing the threshold even when the minimum SAD is large. The upper limit value and threshold value of M may be set in advance, but an appropriate value may be encoded in sequence units, frame units, or block units.

  The regions 405a, 405b, 405c, 415a, 415b, and 415c adjacent to the prediction adjacent regions 404a, 404b, 404c, 414a, 414b, and 414c are determined as the prediction regions for the target block 402, and the pixel signals in the prediction region are candidate predictions Determined to be a signal. Here, the positional relationship between the prediction adjacent region and the prediction region indicating the candidate prediction signal is the same as the target region and the target adjacent region, but this need not be the case. In the present embodiment, as the access information for acquiring each predicted adjacent area (and predicted area) from the frame memory 104, the difference coordinates between the target adjacent area (and target area) and the predicted adjacent area (and predicted area) The identification numbers of the screens (reference images) to which 406a, 406b, 406c, 416a, 416b and 416c and the respective prediction adjacent areas (and prediction areas) belong are stored in the coordinate information memory 202 as coordinate information.

  A configuration of the template matching unit 201 for performing the operation as described above will be described. The template matching unit 201 includes a target adjacent region acquisition unit 211, a predicted adjacent region acquisition unit 212, a comparator 213, and a switch 214. First, the target adjacent area acquisition unit 211 acquires the target adjacent area 403 from the frame memory 104 via the line L104.

  The predicted adjacent region acquisition unit 212 sequentially acquires pixel signals of the same shape as the target adjacent region from the search region in the frame memory 104 via the line L104, and obtains from the target adjacent region acquisition unit 211 via the line L211. SAD is calculated between the pixel signal (adjacent pixel signal) of the target adjacent region to be obtained. The comparator 213 inputs the calculated SAD via the line L212b, and compares it with the Mth smallest SAD acquired so far. When the comparator 213 determines that the input SAD is smaller, the comparator 213 temporarily stores the SAD within the Mth and erases the M + 1th SAD. . Note that the comparator 213 sets a sufficiently large value as an initial value at the start of processing as an initial value of the SAD as compared with a normal SAD.

  In addition to performing this process, the predicted adjacent area acquisition unit 212 obtains coordinate information via the line L202a as information for accessing the predicted adjacent area (and the predicted area) via the line L202a as a memory for coordinate information. To 202. At this time, the coordinate information in which the SAD value is M + 1 is not necessary, and may be overwritten and stored in the coordinate information memory 202. When the search within the search region by the prediction adjacent region acquisition unit 212 is completed, the comparator 213 sends the temporarily stored M SAD values to the candidate prediction signal combination determiner 203 via the line L201b. Output.

  Next, the operation of the candidate prediction signal combination determiner 203 will be described. The candidate prediction signal combination determiner 203 includes a combination setting unit 231, a prediction adjacent region acquisition unit 232, a target adjacent region acquisition unit 233, a weighting unit 234, an adder 235, and a comparison / selection unit 236.

  The combination setter 231 sets a combination of a plurality of predicted adjacent areas based on the M SADs input via the line L201b. As will be described in detail in the description of FIG. 6 with respect to the processing in the predictive adjacent region combination determiner 203, N predictive adjacent regions having small SADs are set as one combination. The value of N is a multiplier of 2 smaller than M, and when M = 6, three combinations of N = 1, 2, and 4 are created.

  Thus, by using the candidate prediction signal combination processing of the present invention, it is possible to determine the smoothing strength of the candidate prediction signal appropriate for each target block, that is, the number of prediction candidate signals to be averaged. It becomes. In other words, N is defined such that the difference SAD between the comparison signal obtained by averaging the pixel signals of N predicted adjacent regions having a small difference SAD that is an absolute error value between the adjacent pixel signals and the adjacent pixel signal is the smallest. Thus, a candidate prediction signal suitable for generating a prediction signal can be selected from the M candidate prediction signals searched without additional information. The reason why the value of N is a multiplier of 2 is that it is considered that the signal averaging process is performed only by addition and shift calculation.

  Further, the combination of the M value, the N value, and the N prediction signal regions is not limited to this. The number of predicted adjacent regions included in one combination can be arbitrarily set from 1 to M. For example, when creating a combination composed of N predicted adjacent regions smaller than M, it is possible to select N from M and set all combinations. At this time, the value of N may be fixed, or a combination may be set by selecting two or more of 1 to M. However, in order to automatically select the same prediction adjacent region combination in the image predictive encoding device 100 which is an encoder and the image predictive decoding device 300 which will be described later as a decoder, the setting method of the combination of both is made to match. It is necessary to keep.

  When one piece of prediction adjacent region combination information is input via the line L231, the prediction adjacent region acquisition unit 232 acquires coordinate information for the prediction adjacent region included in the combination from the coordinate information memory 202 via the line L202a. . Then, the predicted adjacent area acquisition unit 232 acquires the pixel signal of the predicted adjacent area corresponding to the coordinate information via the line L104 and outputs it to the weighting unit 234 as needed.

  The weighter 234 multiplies the pixel signal of each prediction adjacent region input via the line L232 and outputs the result to the adder 235 via the line L234.

  The adder 235 generates a comparison signal to be compared with the pixel signal (adjacent pixel signal) of the target adjacent region by accumulating and accumulating the weighted pixel signals of the predicted adjacent region, and performing the generated comparison. The signal is output to the comparator / selector 236 via the line L235. Note that the operation of the weighter 234 includes, for example, a method of multiplying the pixel signal of each prediction adjacent region by 1 / N when the number of prediction adjacent regions is N, but other methods may be used. For example, a method of calculating a difference value (absolute error sum, square error sum, variance, etc.) between pixel signals of N predicted adjacent regions and adjacent pixel signals and determining a weighting coefficient for each predicted adjacent region based on the ratio Can be considered.

  Here, by making the weighting methods of the weighting unit 205 and the weighting unit 234 the same, an appropriate combination of candidate prediction signals (pixel signals of the prediction region) can be estimated using the prediction adjacent region. Since the predictive adjacent region can be shared by the encoder and the decoder, this is a decoder using this technique. There is an effect that a combination of prediction signals can be acquired without additional information. Note that the weighting methods of the weighter 205 and the weighter 234 are not necessarily the same. For example, in order to reduce the amount of calculation, it is also effective to apply a simple weighting process to the weighter 234 and to apply an adaptive weighting process according to local signal characteristics to the weighter 205. is there.

  The target adjacent area acquisition unit 233 acquires the pixel signal (adjacent pixel signal) of the target adjacent area from the frame memory 104 via the line L104 and outputs the acquired pixel signal to the comparison / selection unit 236.

  The comparator / selector 236 calculates the SAD between the comparison signal corresponding to the combination of the plurality of prediction adjacent regions and the adjacent pixel signal, and uses the combination of the target adjacent regions having the smallest value as the combination of the candidate prediction signals. select. The selected combination of candidate prediction signals (obtained via L231) is output to the prediction region acquisition unit 204 via line L203 as candidate prediction signal combination information.

  As described above, according to the present embodiment, for each target block, it is possible to select a combination of candidate prediction signals effective for prediction signal generation from a plurality of candidate prediction signals without additional information.

  FIG. 5 illustrates a search for a plurality (M) of candidate prediction signals for a pixel signal (target pixel signal) in a target region (target block) by the template matching unit 201 according to the present embodiment, and accesses the searched candidate prediction signals. It is a flowchart which shows the method of acquiring the coordinate information for doing. First, the target adjacent area acquisition unit 211 acquires the target adjacent area (template signal) for the target block from the frame memory 104 (S502).

  Next, the template matching unit 201 initializes a threshold value for selecting M candidate prediction signals to a sufficiently large value (S503). The predicted adjacent area acquisition unit 212 obtains a pixel group having the same shape as the target adjacent area and the target adjacent area in the search area and the sum of absolute differences (SAD) (S504). The SAD is compared with the threshold value by the comparator 213 in the template matching unit 201 (S505). If it is determined that the SAD value is smaller than the threshold value, the process proceeds to S506. If not, the process proceeds to S508.

  The comparator 213 in the template matching unit 201 compares the obtained SAD with the previous SAD, and when the obtained SAD is included by the Mth in ascending order, the searched pixel group is selected as a candidate prediction signal (and prediction adjacent). And the candidate signal is updated. In this embodiment, space-time coordinate information (spatial position and searched pixels) for accessing a candidate prediction signal (and a pixel signal of a prediction adjacent region) instead of the candidate prediction signal (and a pixel signal of the prediction adjacent region) The coordinate information is updated by storing the identification number of the screen including the group in the coordinate information memory 202 based on the switching control by the switch 214 (S506). At the same time, the template matching unit 201 updates the threshold value to the Mth smallest SAD value (S507).

  Thereafter, the predicted adjacent area acquisition unit 212 confirms whether or not all search areas have been searched (S508). If it is determined that not all have been searched, the process returns to S504, and the predicted adjacent area acquisition unit 212 uses the target adjacent area and the next pixel group having the same shape as the target adjacent area in the search area and the sum of absolute differences (SAD). Is required.

  When all the search has been completed, the template matching process for one target block ends (S509).

  In this way, the processing by the template matching unit 201 can identify the top M predicted adjacent areas including pixel signals having a high correlation with the pixel signal in the target adjacent area.

  Next, processing of the candidate prediction signal combination determiner 203 will be described with reference to the drawings. FIG. 6 selects a combination of N candidate prediction signals suitable for generating a prediction signal by smoothing (weighted average) a plurality of candidate prediction signals in the candidate prediction signal combination determiner 203 according to the present embodiment. 3 is a flowchart illustrating a method. First, the combination setter 231 of the combination determiner 203 sets the number N of candidate prediction signals to 1 (S602). Next, the target adjacent area acquisition unit 233 acquires the target adjacent area (template signal) for the target block from the frame memory 104 (S603).

  Then, the predicted adjacent area acquisition unit 232 acquires N predicted adjacent areas belonging to the combination set by the combination setting unit 231 via the line L104. In other words, the predicted adjacent area acquisition unit 232 calculates the pixel signal of the adjacent pixel signal of the target adjacent area for the target block and the pixel signal of the area having the same shape as the target adjacent area in the search area on the reference image (predicted adjacent area candidate). The coordinate information corresponding to the N predicted adjacent areas having a small SAD as the difference value is acquired from the coordinate information memory 202. Then, N predicted adjacent areas corresponding to the acquired coordinate information are acquired from the frame memory 104 (S604).

  Thereafter, the weighting unit 234 and the adder 235 generate a comparison signal by averaging the pixel signals of the N predicted adjacent regions (may be weighted average) (S605), and the comparison / selector 236 generates the comparison signal. SAD, which is a difference value between the comparison signal and the adjacent pixel signal, is calculated (S606). At the same time, the comparator / selector 236 compares the calculated SAD with the minimum SAD so far (S607). If the SAD is determined to be the minimum value, the process proceeds to S608. If not, the process proceeds to S609. move on. In S607, if the calculated SAD and the minimum SAD so far are the same, the process proceeds to S609, but the process may proceed to S608.

  If the calculated SAD is the smallest SAD so far, the comparison / selector 236 stores the combination of coordinate information acquired in S604 (S608).

  Here, the combination determiner 203 updates the value of N twice (S609). Then, the updated N is compared with the magnitude of M (S610). If the updated value of N is smaller than M, the process returns to S604. If the updated value of N is larger than M, the combination of coordinate information stored in S608 is determined as a candidate prediction signal combination, and the candidate prediction signal combination selection process is terminated (S611).

  As described above, by using the candidate prediction signal combination processing of this embodiment, it is possible to determine the smoothing strength of the candidate prediction signal appropriate for each target block, that is, the number of prediction candidate signals to be averaged. It becomes possible. In other words, by determining N with the smallest difference SAD between the comparison signal obtained by averaging the pixel signals of N predicted adjacent regions having a small SAD which is a difference value between the adjacent pixel signals and the adjacent pixel signal. From the searched M candidate prediction signals, a candidate prediction signal suitable for generating a prediction signal can be selected without additional information.

  The reason why the value of N is a multiplier of 2 is that it is considered that the signal averaging process is performed only by addition and shift calculation. In the present embodiment, the value of N is not limited to a multiplier of 2. Moreover, the method of setting the combination of a some prediction adjacent area | region is not limited to the method of FIG. The number of predicted adjacent regions included in one combination can be arbitrarily set from 1 to M. For example, when creating a combination composed of N predicted adjacent regions smaller than M, it is possible to set all combinations of selecting N from M. At this time, the value of N may be fixed, or a combination may be set by selecting two or more of 1 to M. In addition, in order to automatically select the same combination of prediction adjacent regions in the encoder and the decoder, it is necessary to match the setting method of the combination of both.

  FIG. 7 is a flowchart showing a prediction signal generation method in smoothing (weighted average) of a plurality of candidate prediction signals according to the present embodiment.

  The prediction region acquisition unit 204 acquires one or more candidate prediction signals corresponding to the target block from the frame memory 104 according to the selected coordinate information (S702). Then, the weighting unit 205 and the adder 206 perform weighted averaging of the obtained candidate prediction signals to generate a prediction signal for the target block (S703). This completes the process for one target block (S704).

  FIG. 8 is a flowchart showing an image predictive encoding method in the image predictive encoding device 100 according to the present embodiment. First, the prediction signal generator 103 of FIG. 2 generates a prediction signal of the target block (S102). Next, a residual signal indicating a difference between the signal of the target block and the prediction signal of the target block is encoded by the transformer 106, the quantizer 107, and the entropy encoder 111 (S103). The encoded residual signal is output via the output terminal 112 (S105).

  Thereafter, the encoded residual signal is decoded by an inverse quantizer 108 and an inverse transformer 109 in order to predictively encode a subsequent target block. Then, the prediction signal is added to the decoded residual signal by the adder 110, and the signal of the target block is reproduced and stored as a reference image in the frame memory 104 (S106). If all the target blocks have not been processed, the process returns to S102, the process for the next target block is performed, and if it has been completed, the process ends (S107).

  As described above, in the image predictive coding device 100 of the present embodiment, a prediction signal smoothed using a plurality of prediction signals can be obtained without using additional information.

  Next, an image predictive decoding method according to this embodiment will be described. FIG. 9 is a block diagram showing an image predictive decoding apparatus 300 according to this embodiment. The image predictive decoding apparatus 300 includes an input terminal 301, an entropy decoder 302 (data decoding means), an inverse quantizer 303 (residual signal restoration means), an inverse transformer 304 (residual signal restoration means), and an adder 305. (Reproduced image signal generation means), output terminal 306, frame memory 307, prediction signal generator 308 (prediction signal generation means). As the residual signal restoration means by the inverse quantizer 303 and the inverse transformer 304, other means may be used. Further, the inverse converter 304 may not be provided. Each configuration will be described below.

  The input terminal 301 inputs compressed data that has been compression-encoded by the above-described image predictive encoding method. The compressed data includes a residual signal obtained by predicting and encoding a target block obtained by dividing an image into a plurality of blocks.

  The entropy decoder 302 decodes the encoded data of the residual signal of the target block by entropy decoding the compressed data input at the input terminal 301.

  The inverse quantizer 303 inputs the residual signal of the decoded target block via the line L302 and performs inverse quantization. The inverse transformer 304 performs inverse discrete cosine transform on the inversely quantized data. Inverse quantizer 303 and inverse transformer 304 output signals obtained by inverse quantization and inverse discrete cosine transform, respectively, as difference signals (reproduced residual signals).

  The prediction signal generator 308 basically has the same configuration as that shown in FIG. 2 or a function corresponding thereto, and generates a prediction signal by the same processing as the prediction signal generator 103 of FIG. Since the prediction signal generator 308 generates a prediction signal only from the reproduced signal stored in the frame memory 307, by managing the input data to the frame memory 307 and the frame memory 104 in FIG. The same prediction signal as the prediction signal generator 103 in FIG. 1 can be generated. Details of the configuration of the prediction signal generator 308 are omitted because they have already been described with reference to FIG. The prediction signal generator 308 operating in this way outputs the generated prediction signal to the adder 305 via the line L308.

  The adder 305 adds the prediction signal generated by the prediction signal generator 308 to the difference signal (reproduction residual signal) restored by the inverse quantizer 303 and the inverse transformer 304, and reproduces the reproduced image of the target block. The signal is output to the output terminal 306 and the frame memory 307 via the line L305. The output terminal 306 outputs to the outside (for example, a display).

  The frame memory 307 stores the reproduction image output from the adder 305 as a reference image as a reference reproduction image for the next decoding process. At this time, the reproduced image is managed by the same method as that of the image predictive coding apparatus 100 of FIG.

  Next, the image predictive decoding method in the image predictive decoding apparatus 300 according to the present embodiment will be described with reference to FIG. First, the compressed compressed data is input via the input terminal 301 (S302). Then, the entropy decoder 302 performs entropy decoding on the compressed data and extracts quantized transform coefficients (S303). Here, a prediction signal is generated by the prediction signal generator 308 (S304). The process of S304 is basically the same as the process S102 of FIG. 8, and the processes of FIGS.

  On the other hand, the quantized transform coefficient is inversely quantized using the quantization parameter in the inverse quantizer 303, and inverse transform is performed in the inverse transformer 304, thereby generating a reproduction difference signal (S305). Then, the generated prediction signal and the reproduction difference signal are added to generate a reproduction signal, and the reproduction signal is stored in the frame memory to reproduce the next target block (S306). If there is next compressed data, this process is repeated again (S307), and all data is processed to the end (S308). If necessary, the process may return to S302 and take in compressed data.

  The image predictive encoding method and the image predictive decoding method according to the present embodiment can be provided by being stored in a recording medium as a program. Examples of the recording medium include a recording medium such as a floppy disk (registered trademark), CD-ROM, DVD, or ROM, or a semiconductor memory.

  FIG. 11 is a block diagram illustrating modules of a program that can execute the image predictive coding method. The image predictive coding program P100 includes a block division module P102, a prediction signal generation module P103, a storage module P104, a subtraction module P105, a transformation module P106, a quantization module P107, an inverse quantization module P108, an inverse transformation module P109, and an addition module P110. The entropy encoding module P111 is included. As shown in FIG. 12, the prediction signal generation module P103 includes a template matching module P201, a candidate prediction signal combination determination module P202, and a prediction signal synthesis module P203.

  The functions realized by executing the modules are the same as those of the constituent elements of the image predictive coding apparatus 100 described above. That is, the function of each module of the image predictive coding program P100 includes a block divider 102, a prediction signal generator 103, a frame memory 104, a subtractor 105, a converter 106, a quantizer 107, an inverse quantizer 108, and an inverse quantizer. The functions of the converter 109, the adder 110, and the entropy encoder 111 are the same, and the functions of each module of the prediction signal generation module P103 are the template matching unit 201, the candidate prediction signal combination determiner 203, and signal synthesis. This is similar to the functions of the prediction region acquisition unit 204 to the addition unit 206.

  FIG. 13 is a block diagram showing modules of a program that can execute the image predictive decoding method. The image predictive decoding program P300 includes an entropy decoding module P302, an inverse quantization module P303, an inverse transform module P304, an addition module P305, a storage module P307, and a prediction signal generation module P308.

  The functions realized by executing the modules are the same as those of the components of the image predictive decoding apparatus 300 described above. That is, the function of each module of the image predictive decoding program P300 is the same as the functions of the entropy decoder 302, the inverse quantizer 303, the inverse transformer 304, the adder 305, and the frame memory 307. The prediction signal generation module P308 has a function equivalent to that of the prediction signal generation module P103 in the image prediction encoding program P100, and includes a template matching unit 201, a candidate prediction signal combination determination unit 203, and signal synthesis. Functions of the prediction region acquisition unit 204 to the adder 206 are provided.

  The image predictive encoding program P100 or the image predictive decoding program P300 configured as described above is stored in the recording medium 10 and executed by a computer described later.

  FIG. 16 is a diagram illustrating a hardware configuration of a computer for executing a program recorded in a recording medium, and FIG. 17 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. 16, the computer 30 includes a reading device 12 such as a floppy disk drive device (floppy is a registered trademark), a CD-ROM drive device, a DVD drive device, and a working memory (RAM) in which an operating system is resident. 14, a memory 16 for storing a program stored in the recording medium 10, 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 a program CPU 26 for controlling the execution of the above. When the recording medium 10 is inserted into the reading device 12, the computer 30 can access the image predictive coding program P100 or the image predictive decoding program P300 stored in the recording medium 10 from the reading device 12, and image predictive coding. The program P100 or the image predictive decoding program P300 can operate as the image predictive encoding device 100 or the image predictive decoding device 300 of the present embodiment.

  As shown in FIG. 17, the image predictive coding program P100 or the image predictive decoding program P300 may be provided via a network as a computer data signal 40 superimposed on a carrier wave. In this case, the computer 30 stores the image predictive encoding program P100 or the image predictive decoding program P300 received by the communication device 24 in the memory 16, and executes the image predictive encoding program P100 or the image predictive decoding program P300. it can.

  The embodiment described so far may be modified as follows. In the candidate prediction signal combination determiner 203 in FIG. 2, the SAD (absolute error value) that is a difference value between the pixel signal of the target adjacent region of the target block and the comparison signal obtained by weighted averaging the pixel signals of the plurality of adjacent prediction regions. The optimal combination of candidate prediction signals is determined by calculating the sum of the two), but the combination can also be determined by using the sum of the square error of the difference signal (SSD) or variance (VAR) instead of SAD. It is. In the three evaluation criteria, the calculation amount increases in the order of SAD, SSD, and VAR. On the other hand, the evaluation accuracy is improved, and the effect that the code amount of the error signal can be reduced is expected.

  It is also effective to determine the final combination using SSD or VAR when a plurality of combinations having the same SAD as the difference value between the target adjacent signal and the comparison signal are obtained. That is, when the SAD calculated in the prediction adjacent region acquisition unit 232 matches the minimum value calculated so far, the comparator / selector 236 further selects any value using SSD or VAR as a comparison target. Compare what is small. Here, the combination determined to have a small SSD or VAR is stored in the comparator / selector 236 as a combination having the minimum value. In this case, the comparator / selector 236 calculates SSD or VAR together with SAD, and temporarily stores it.

  Furthermore, it is also possible to use a method for calculating the variance of combinations of a plurality of candidate prediction signals instead of the pixel signals in the adjacent prediction region and using them for determining the combination. Specifically, the prediction adjacent region acquisition unit 232, the weighting unit 234, and the adder 235 in FIG. 2 can be realized by replacing with the prediction region acquisition unit 204, the weighting unit 205, and the adder 206, respectively. In this modified example, the processing of the target adjacent region acquisition unit 233 is not necessary, and a prediction signal can be output from the comparison / selection unit 236 to the subtracter 105 in FIG. 1, so that the circuit scale can be reduced. There is an effect.

  Also in the comparator 213 of the template matching unit 201, the SAD that is the difference value is used for evaluating the pixel group having the same shape as the target adjacent area of the target block and the searched target adjacent area, but the SSD or VAR is changed. In other words, the same effect as in the case of the candidate prediction signal combination determiner 203 can be expected.

  The prediction signal generator 103 in FIG. 1 (the prediction signal generator 308 in FIG. 9) is not limited to the configuration in FIG. For example, in FIG. 2, coordinate information for accessing a plurality of candidate prediction signals searched by the template matcher is stored in the coordinate information memory 202. A configuration in which pixel signals are stored may be used. Although the amount of memory in FIG. 2 increases, there is an effect of reducing access to the frame memory. Further, the template matching unit 201 outputs the SAD that is the difference value between the target adjacent region and the M predicted adjacent regions to the candidate prediction signal combination determiner 203 via the line L201b. In addition, when the SAD that is the difference value is not used, the line L201b is unnecessary.

  In the present embodiment, the target adjacent area is composed of reproduced pixels, but may be composed of prediction signals of adjacent blocks. This is effective when it is desired to make the prediction area smaller than the coding block unit of the error signal, or when it is desired to generate the signal of the target adjacent area without performing the encoding / decoding process.

This embodiment can be applied to a method of searching for an indefinite number of candidate signals and selecting an appropriate number of candidate signals by template matching.

<Second Embodiment>
In the first embodiment, a combination of candidate prediction signals (prediction region pixel signals) for generating a prediction signal of a target region is generated from a plurality of candidates using a correlation between the prediction adjacent region and the target adjacent region. The method of determination was shown. In the second embodiment, a prediction signal generator 103a that determines a combination of candidate prediction signals for generating a prediction signal of a target region from a plurality of candidates using a correlation between the plurality of candidate prediction signals is shown. In the second embodiment, a method for determining the number of candidate prediction signals, not combinations of candidate prediction signals, will be described. This embodiment shows a case where a combination of candidate prediction signals is uniquely determined according to the number of candidate prediction signals, and is merely an example of a combination of candidate prediction signals. Similarly to the first embodiment, the second embodiment can also be applied to a method for determining one combination from a plurality of candidate combinations of candidate prediction signals.

  Subsequently, a second embodiment of the present invention will be described with reference to a modified attached drawing. Description of the drawings and description overlapping with those of the first embodiment will be omitted.

  Processing of the prediction signal generator 103a in the second embodiment will be described. Note that the prediction signal generator 308a applied to the image predictive decoding apparatus 300 performs the same configuration and processing as the prediction signal generator 103a. In addition, regarding the image predictive coding apparatus 100 of FIG. 1 and the image predictive decoding apparatus 300 of FIG. This is the same as in the first embodiment, and a description thereof is omitted. FIG. 18 is a block diagram showing the prediction signal generator 103a used in the image predictive coding apparatus 100 according to the present embodiment. The template matching unit 201, the coordinate information memory 202, the averaging number determiner 1800, and the prediction region acquisition. The unit 204 includes a weighter 205 and an adder 206.

  The processing of the template matching unit 201 is the same as that in FIG. That is, an image signal (reproduced image signal) that has already been reproduced in the past process is input from the frame memory 104 via the line L104, and a candidate for a prediction signal (candidate prediction signal) for the target pixel signal in the target region is template matching described later. And a plurality of coordinate information for accessing the searched candidate prediction signal is output to the coordinate information memory 202 via the line L201a. At the same time, difference value data (corresponding to the sum of absolute error values (corresponding to SAD (sumof absolute difference) described later) indicating the relationship between the target region and each candidate prediction signal is output to the averaging number determiner 1800 via the line L201b. In this embodiment, as in the first embodiment, the target area (target block) is a small block of 4 × 4 pixels obtained by dividing the encoding target block made up of 8 × 8 pixels, but the other blocks It may be divided into sizes or shapes, or may not be divided.

  The averaging number determiner 1800 obtains a plurality of candidate prediction signals based on the coordinate information for accessing the plurality of candidate prediction signals input via the line L202a, and considers the correlation between them. The number of candidate prediction signals used for generating the prediction signal is determined. The determined number of candidate prediction signals is output to the prediction region acquisition unit 204 via line L203 as information on the number of candidate prediction signal averages.

  The prediction area acquisition unit 204 acquires the coordinate information of the number of candidate prediction signals determined by the averaging number determiner 1800 via the line L202b according to the average number information of the candidate prediction signals input via the line L203. Then, the prediction region acquisition unit 204 acquires a candidate prediction signal corresponding to the acquired coordinate information from the frame memory 104 via the line L104 and outputs it to the weighting unit 205 as needed. The weighting unit 205 multiplies each candidate prediction signal input via the line L204 by a weighting coefficient, and outputs the result to the adder 206 via the line L205. The adder 206 sequentially adds the weighted candidate prediction signals and outputs the result as a prediction signal to the subtracter 105 in FIG. 1 via the line L103. As for the operation of the weighting unit, for example, there is a method of multiplying each candidate prediction signal by 1 / N when the number of candidate prediction signals is N, but other methods may be used.

  Since the operation of the template matching unit 201 in FIG. 18 is the same as that in FIG. 2, the detailed description is omitted, but between the search area in the frame memory 104 and the pixel signal (adjacent pixel signal) of the target adjacent area. M predicted adjacent areas having a small SAD are searched, and coordinate information, which is information for accessing the M predicted adjacent areas (and predicted areas), is output to the coordinate information memory 202 via the line L201. In addition, the calculated M SAD values are output to the averaging number determiner 1800 via the line L201b.

  Next, the operation of the averaging number determiner 1800 regarding the candidate prediction signal will be described. The averaging number determiner 1800 includes a counter 1801, a representative prediction region acquisition unit 1802, a correlation evaluation value calculator 1803 between prediction regions, and a threshold processing unit 1804.

  The counter 1801 sets an average number candidate N based on M input via the line L201b. In the present embodiment, the maximum value of N is M, and N is decreased until N = 2. In the case of M = 6, the value of N is output to the representative prediction region acquisition unit 1802 and the threshold processing unit 1804 via the line L1801 in the order of 6, 5, 4, 3, 2. Note that the value of N in the counter 1801 is not limited to the above. For example, the value of N may be a multiplier of 2 smaller than M, and when M = 6, the value of N may be output from the counter in the order of 4 and 2, or the value of N as N = 4. May be limited to one. However, in order to automatically select the same prediction adjacent region combination in the image predictive encoding device 100 which is an encoder and the image predictive decoding device 300 which will be described later as a decoder, the setting method of the combination of both is made to match. It is necessary to keep.

When one candidate for the value of N is input, the representative prediction region acquisition unit 1802 outputs the pixel signal of the prediction adjacent region and the pixel of the target adjacent region among the M prediction regions searched by the template matching unit 201. The coordinate information for accessing the N prediction regions having a small SAD between the signals (adjacent pixel signals) is acquired from the coordinate information memory 202 via the line L202a. Next, a pixel signal of a predetermined prediction area is acquired from the frame memory 104 using this coordinate information. In the present embodiment, the pixel signal p ₀ (i, j) in the prediction region having the smallest SAD between the pixel signal in the prediction adjacent region and the pixel signal (adjacent pixel signal) in the target adjacent region and the Nth smallest prediction region Pixel signal p _N-1 (i, j) is obtained and output to the correlation evaluation value calculator 1803 between the prediction regions via the line L1802. Here, (i, j) indicates the pixel position in the prediction region. i and j are values in the range of 0 to B-1, and B = 4 when the target region is 4 × 4.

When the pixel signals (candidate prediction signals) of the two prediction regions are input via the line L1802, the correlation evaluation value calculator 1803 between the prediction regions is calculated between the prediction regions for the N values according to the following equation (1). Correlation evaluation value EV _N is calculated and output to threshold value processor 1804 via line L1803.

Thresholding unit 1804, the correlation evaluation value between the prediction region for the N value via line L1803 is input, compares the predetermined threshold th and the correlation evaluation value EV _N (N). In the present embodiment, the value of th (N) is 32 regardless of the value of N. If the correlation evaluation value for N is smaller than a predetermined threshold th (N), the average number AN of the candidate prediction signal is determined to be N, and the value of N passes through line L203 as the average number information of the candidate prediction signal. Is output to the prediction region acquisition unit 204. When the correlation evaluation value for N is larger than a predetermined threshold th (N), the counter 1801 is notified via the line L1804 to update the value of N to a smaller value. However, when the value of N is 2, the value of AN is determined to be 1, and is output to the prediction region acquisition unit 204 via the line L203 as the average number information of the candidate prediction signal. Note that in the averaging number determiner in which the output value of the counter 1801 is a fixed value, even if the correlation evaluation value EV _{N for N} is larger than a predetermined threshold th (N), the value of N is not updated, and the AN The value is determined to be a predetermined default value (for example, 1), and is output to the prediction region acquisition unit 204 via the line L203 as the average number information of the candidate prediction signal.

  As described above, according to the present embodiment, it is possible to select the number of candidate prediction signals effective for prediction signal generation without additional information for each target region (target block). Noise removal by averaging a plurality of candidate prediction signals is effective when only the noise components of the candidate prediction signals are different and the signal components are similar. In the present invention, by evaluating the similarity between candidate prediction signals, the number of candidate prediction signals can be determined for each target region for generating a prediction signal.

Note that the method of calculating the correlation evaluation value EV _N between the prediction regions is not limited to Equation (1). For example, the following equation (2) or equation (3) may be used to mutually evaluate a difference value between N candidate prediction signals, and the added value may be used as a correlation evaluation value. However, in this case, the representative prediction region acquisition unit 1802 needs to acquire N candidate prediction signals from the frame memory 104.

  Also, the threshold th (N) is not limited to 32. For example, if the number of pixels in the target region changes, the threshold value needs to be changed. Further, when the correlation evaluation value between prediction regions is calculated using Expression (2) or Expression (3), it is necessary to adaptively change th (N) according to the value of N.

  The value of th (N) needs to be the same value in the encoder and the decoder. Although it may be determined in advance, a method of encoding in sequence units, frame units, or slices (composed of a plurality of blocks) and sending them to the decoder is also conceivable. Moreover, you may send in a block unit.

  Next, the processing of the candidate prediction signal averaging number determiner 1800 will be described with reference to the drawings. Since the processing of the template matching unit 201 according to the present embodiment has already been described with reference to FIG. 5, the description thereof will be omitted, but FIG. Is a flowchart showing a method of searching for a candidate prediction signal (and a pixel signal of a predicted adjacent region) and acquiring coordinate information for accessing the searched candidate prediction signal (and a pixel signal of the predicted adjacent region) A process of searching for pixel signals of M predicted adjacent areas having a small SAD with the adjacent pixel signal of the target adjacent area adjacent to the reference image from the search area on the reference image is performed.

  Next, the processing of the averaging number determiner 1800 regarding the candidate prediction signal will be described with reference to the drawings. FIG. 19 is a flowchart showing a method of selecting the number of candidate prediction signals suitable for generating a prediction signal by smoothing (weighted average) of candidate prediction signals in the averaging number determiner 1800 according to this embodiment. First, the counter 1801 of the averaging number determiner 1800 sets the number N of candidate prediction signals to M (1902).

  Next, pixel signals (candidate prediction signals) of N prediction regions set by the counter 1801 are acquired via the line L104 by the representative prediction region acquisition unit 1802. That is, the representative prediction area acquisition unit 1802 uses corresponding coordinate information for coordinate information for N prediction adjacent areas whose absolute value difference sum (SAD) is small with the adjacent pixel signal of the target adjacent area adjacent to the target block. Obtained from the memory 202. Then, N prediction regions corresponding to the acquired coordinate information are acquired from the frame memory 104 (S1903).

Then, a correlation evaluation value calculator 1803 between prediction regions calculates a correlation evaluation value EV _N between N prediction regions according to the equation (1). In Expression (1), since only pixel signals of two prediction areas are used for EV _N calculation, in this case, only two prediction areas may be acquired in step S1903.

Thereafter, the threshold value processor 1804 compares EV _N with a predetermined threshold value th (N) (32 in this embodiment) (S1905). If EV _N is smaller than th (N), the candidate prediction signal Is determined to be N (S1906), and the process is terminated (S1910). If EV _N is greater than th (N), the process proceeds to S1907. In S1905, when EV _N and th (N) are the same, the process proceeds to S1907. However, the process may proceed to S1906. In S1907, the counter 1801 updates the value of N (S1907). At this time, if the value of N is 1 (S1908), the averaging number determiner 1800 determines AN as 1 (S1909) and ends the process. If the value of N is 2 or more, the process returns to S1904. When the averaging number determiner 1800 fixes N candidates to one, the fixed value is set to N in S1902, and S1907 and S1908 are omitted.

  As described above, according to the averaging number determination process of the present embodiment, it is possible to select the number of candidate prediction signals effective for prediction signal generation without additional information for each target region (target block). . Noise removal by averaging a plurality of candidate prediction signals is effective when only the noise components of the candidate prediction signals are different and the signal components are similar. In the present invention, by evaluating the similarity between candidate prediction signals, the number of candidate prediction signals can be determined for each target region for generating a prediction signal.

  Note that the method of calculating the correlation evaluation value between the prediction regions is not limited to the equation (1). For example, a difference value between N candidate prediction signals may be evaluated mutually using Equation (2) or Equation (3), and the added value may be used as a correlation evaluation value. However, in this case, the representative prediction region acquisition unit 1802 needs to acquire N candidate prediction signals from the frame memory 104.

  Also, the threshold th (N) is not limited to 32. For example, if the number of pixels in the target region changes, the threshold value needs to be changed. Further, when the correlation evaluation value between prediction regions is calculated using Expression (2) or Expression (3), it is necessary to adaptively change th (N) according to the value of N.

  The value of th (N) needs to be the same value in the encoder and the decoder. Although it may be determined in advance, a method of encoding and sending in sequence units or frame units is also conceivable.

  In the present embodiment, when M = 6, N candidates are set to 6, 5, 4, 3, 2, but the present invention is not limited to this setting. For example, N value candidates may be 4 and 2, which are multipliers of 2. This process can be realized by setting the value of N to 4 in S1902 and changing S1907 to a process of halving the value of N (N = N / 2) (when M = 6). However, it is necessary to match the N candidate setting methods in the encoding process and the decoding process.

  FIG. 20 is a flowchart showing a prediction signal generation method in smoothing (weighted average) of a plurality of candidate prediction signals according to the present embodiment.

  The prediction region acquisition unit 204 acquires AN candidate prediction signals corresponding to the target block from the frame memory 104 according to the selected coordinate information and candidate prediction signal averaging number information (S2002). Then, the weighting unit 205 and the adder 206 perform weighted averaging of the obtained candidate prediction signals to generate a prediction signal for the target block (S2003). This completes the processing for one target block (S2004).

  The image predictive encoding method and the image predictive decoding method in the image predictive encoding device 100 according to the second embodiment are the same as the flowcharts shown in FIGS. 8 and 9 and will not be described.

  The image predictive encoding method and the image predictive decoding method according to the second embodiment can be provided by being stored in a recording medium as a program. Examples of the recording medium include a recording medium such as a floppy disk (registered trademark), CD-ROM, DVD, or ROM, or a semiconductor memory. The configuration of the program module capable of executing the image predictive coding method and the program module capable of executing the image predictive decoding method are the same as those in the block diagrams shown in FIGS. To do. As shown in FIG. 21, the predicted signal generation module P103 includes a template matching module P201, an average signal number determination module P2102, and a predicted signal synthesis module P203. The hardware and computer for executing the program recorded on the recording medium have also been described in FIGS. 16 and 17 and will not be described.

  The embodiment described so far may be modified as follows.

  In the above description, the correlation evaluation value used to determine the combination of candidate prediction signals is calculated using the pixel signal (candidate prediction signal) in the prediction region, but the pixel in the prediction adjacent region adjacent to the prediction region instead of the prediction region A correlation evaluation value may be calculated using the signal. Furthermore, the correlation evaluation value may be calculated using a pixel signal of a region that combines the prediction region and the prediction adjacent region (for example, a region including the prediction adjacent region 404b and the prediction region 405b in FIG. 3).

In the above, the AN value is determined by the averaging number determiner and the averaging number determination process. However, the threshold value processing of the present invention is performed only for one or a few N candidates, and the prediction region is determined from the threshold value. When the correlation evaluation value EV is large, a combination that determines AN by other methods is also possible. For example, EV _N is calculated for one N candidate and threshold processing in the second embodiment is performed. When EV _N is smaller than th (N), AN = N and EV _N is equal to th (N). If larger, a method of determining the AN value by applying the candidate prediction signal combination determination process in the first embodiment is conceivable. That is, for a plurality of AN candidates n, a method may be considered in which n is the smallest difference SAD between the comparison signal obtained by averaging the pixel signals of n predicted adjacent regions and the adjacent pixel signal.

  In the averaging number determiner 1800 of FIG. 2, the correlation evaluation value of the prediction region is calculated using SAD (sum of absolute error values) between the pixel signals (candidate prediction signals) of the prediction region. It is also possible to use the sum (SSD) or variance (VAR) of the square error of the difference signal. In the three evaluation criteria, the calculation amount increases in the order of SAD, SSD, and VAR. On the other hand, the evaluation accuracy is improved, and the effect that the code amount of the error signal can be reduced is expected. Also, instead of SAD or SSD, absolute value error average (MAD) or mean square error (MSE) may be used.

The modification regarding template matching in the first embodiment also applies to the second embodiment.

<Third embodiment>
The first and second embodiments have been described on the assumption that the weighted average when generating the prediction signal of the target region from a plurality of candidate prediction signals is a predetermined method. However, if the candidate prediction signal combination selection method according to the first embodiment of the present invention is used, it is possible to select from a plurality of prepared weighting factors for each appropriate target region without additional information. This embodiment can be understood by considering the modifications of the drawings used in the above description and the following description. Subsequently, a third embodiment of the present invention will be described with reference to a modified attached drawing. Description of the drawings and description overlapping with those of the first embodiment will be omitted.

  The processing of the prediction signal generator 103b in the third embodiment will be described. Note that the prediction signal generator 308b applied to the image predictive decoding apparatus 300 also performs the same configuration and processing as the prediction signal generator 103b. As for the image predictive coding apparatus 100 in FIG. 1 and the image predictive decoding apparatus 300 in FIG. This embodiment is the same as the embodiment of FIG. FIG. 22 is a block diagram showing the prediction signal generator 103b used in the image predictive coding apparatus 100 according to the present embodiment. The template matching unit 201, the coordinate information memory 202, the candidate prediction signal combination determiner 2200, and the prediction A region acquisition unit 204 and a weighted averager 2204 are included.

  The processing of the template matching unit 201 is the same as that in FIG. That is, an image signal (reproduced image signal) that has already been reproduced in the past process is input from the frame memory 104 via the line L104, and a candidate for a prediction signal (candidate prediction signal) for the target pixel signal in the target region is template matching described later. And a plurality of coordinate information for accessing the searched candidate prediction signal is output to the coordinate information memory 202 via the line L201a. At the same time, difference value data indicating the relationship between the target region and each candidate prediction signal (corresponding to a sum of absolute error values (SAD (sumof absolute difference) described later)) is combined with a candidate prediction signal combination determiner 2200 via a line L201b. In this embodiment, as in the first embodiment, the target region (target block) is a small block of 4 × 4 pixels obtained by dividing the encoding target block made up of 8 × 8 pixels. It may be divided into block sizes or shapes, or may not be divided.

  The candidate prediction signal combination determiner 2200 sets a plurality of candidate prediction signal combinations using the difference value data input via the line L201b. Furthermore, using the same difference value data, a plurality of weighting factors for weighted averaging of candidate prediction signals are set. Then, using the pixel signal input from the frame memory via the line L104 according to the coordinate information input from the coordinate information memory 202 via the line L202a, the combination of candidate prediction signals and the candidate prediction signals are weighted averaged The combination information of the candidate prediction signals is output to the prediction region acquisition unit 204 via the line L203, and the weight coefficient information of the candidate prediction signals is weighted via the L2203. It is output to the averager 2204.

  The prediction area acquisition unit 204 acquires coordinate information of candidate prediction signals belonging to this combination via line L202b according to the combination information of candidate prediction signals input via line L203. Then, the prediction region acquisition unit 204 acquires a candidate prediction signal corresponding to the acquired coordinate information from the frame memory 104 via the line L104, and outputs it to the weighted averager 2204 as needed. The weighted averager 2204 multiplies each candidate prediction signal input via the line L204 by the weighting coefficient input via the line L2203 and sequentially adds them to calculate a weighted addition value. At the same time, the weighted averager 2204 adds the weighting coefficient input via L2203. Then, the weighted addition value is divided by the addition value of the weighting coefficient, and is output to the subtracter 105 in FIG.

式（４）において、NはラインＬ２０３経由で予測領域取得器２０４に入力される候補予測信号の組み合わせ情報を示しており、予測領域取得器２０４はＮ個の候補予測信号p_n(i,j) (n=0〜N-1)をフレームメモリ１０４から取得する。w_nはラインL２２０３経由で重み付け平均化器２２０４に入力される重み係数の組（Ｎ個）を示している。n番目の重み係数w_nは、重み付け平均化器２２０４にてn番目の候補予測信号p_n(i,j)内の各画素と掛け合わされ、重み付けされたＮ個の候補予測信号の各画素は累積加算される。このように重み付け加算された候補予測信号の重み付け加算値は、重み係数の加算値で除算され、各画素の予測信号Pb(i,j)が算出される。 The procedure for generating the prediction signal Pb (i, j) of the target region, which is performed by the prediction region acquisition unit 204 and the weighted averager 2204, is organized in Expression (4). Here, (i, j) indicates the coordinates of each pixel in the target region.

In Equation (4), N indicates combination information of candidate prediction signals input to the prediction region acquisition unit 204 via the line L203, and the prediction region acquisition unit 204 includes N candidate prediction signals p _n (i, j ) (n = 0 to N−1) is acquired from the frame memory 104. w _n indicates a set (N) of weighting factors input to the weighted averager 2204 via the line L2203. n-th weighting coefficient w _n is, n-th candidate prediction signal p _n (i, j) in the weighted averaging unit 2204 is multiplied with each pixel in each pixel of N weighted candidate prediction signals Cumulative addition. The weighted addition value of the candidate prediction signal thus weighted and added is divided by the addition value of the weighting coefficient to calculate the prediction signal Pb (i, j) of each pixel.

  Since the operation of the template matching unit 201 in FIG. 22 is the same as that in FIG. 2, the detailed description is omitted, but SAD between the search area in the frame memory 104 and the pixel signal (adjacent pixel signal) in the target adjacent area is performed. M prediction adjacent regions having a small size are searched, and coordinate information, which is information for accessing the M predicted adjacent regions (and prediction regions), is output to the coordinate information memory 202 via the line L202a. Further, the calculated M SAD values are output to the candidate prediction signal combination determiner 2200 via the line L201b.

  Next, the operation of the candidate prediction signal combination determiner 2200 will be described. The candidate prediction signal combination determiner 203 includes a combination setter 231, a weight coefficient setter 2201, a prediction adjacent region acquisition unit 232, a target adjacent region acquisition unit 233, a weighted averager 2202, and a comparison / selector 2203. Is done.

  In the third embodiment, when generating the prediction signal of the target region, first, the combination setting unit 231 and the weighting factor setting unit 2201 operate for the combination of the candidate prediction signals and the weighted averaging.

  The combination setter 231 sets a combination of a plurality of predicted adjacent areas based on the M SADs input via the line L201b. In the present embodiment, the N predicted adjacent areas with small input SADs are set as one combination. The set combination is output to the predicted adjacent region acquisition unit 232, the comparison / selection unit 2203, and the weighting factor setting unit 2201 via the line L231. The value of N is a multiplier of 2 smaller than M, and when M = 6, three combinations of N = 1, 2, and 4 are created. In addition, the combination of the value of M, the value of N, and N prediction signal area | regions is not limited to this. The number of predicted adjacent regions included in one combination can be arbitrarily set from 1 to M. For example, when creating a combination composed of N predicted adjacent regions smaller than M, it is possible to select N from M and set all combinations. At this time, the value of N may be fixed, or a combination may be set by selecting two or more of 1 to M. Moreover, the number will not be limited if the number of combinations is one or more. However, in order to automatically select the same prediction adjacent region combination in the image predictive encoding device 100 which is an encoder and the image predictive decoding device 300 which will be described later as a decoder, the setting method of the combination of both is made to match. It is necessary to keep.

（２）q=2、式（８）、対象隣接領域t(r) (rは対象隣接領域内のR個の画素の位置)とＮ個の予測隣接領域s_n(r)(n=0〜N-1)との間のＳＡＤ（LSAD_n、式（７））に基づいて算出。

（３）q=3、式（９）、ＳＡＤ値の小さい対象隣接領域の画素信号に大きな値の重み係数を割り当てる

（４）q=4、図２６に示す対応表に基づいてLSAD₀とLSAD_n(n=0〜N-1)の関係からw_nの値を設定。 The weighting factor setting unit 2201 is configured to calculate N predicted adjacent regions based on the combination information of the predicted adjacent regions input from the line L231 and the SAD between the target adjacent region and the pixel signals of the N predicted adjacent regions. A plurality of sets of weighting coefficient sets for weighted averaging of pixel signals are set. In the present embodiment, four types are prepared as candidates for a set of weighting factors. In the following, q and n indicate the weighting coefficient of type q (q = 1-4) for the pixel signal of the nth (n = 0 to N-1) predicted adjacent region (1) q = 1, formula (6) ), Set N weighting factors to 1 / N each

(2) q = 2, the formula (8), the target adjacent region t (r) (r is the position of R pixels in the target adjacent region) prediction adjacent regions and the N _{s n (r) (n =} 0 To N-1) based on SAD (LSAD _n , Equation (7)).

(3) q = 3, Expression (9), assigning a large weighting factor to the pixel signal in the target adjacent region having a small SAD value

(4) q = 4, and the value of w _n is set from the relationship between LSAD ₀ and LSAD _n (n = ₀ to _N −1) based on the correspondence table shown in FIG.

FIG. 26 is an explanatory diagram illustrating an example of a correspondence table in which a set of weighting factors is set based on a difference value between a pixel signal in the target adjacent region and a pixel signal in the predicted adjacent region. LSAD _n is between target adjacent region t (r) (r is the position of R pixels in the target adjacent region) and the N prediction adjacent regions _{s n (r) (n =} 0~N-1) LSAD ₀ indicates SAD when n is 0. By using the correspondence table shown in FIG. 26, the weighting coefficient is uniquely determined based on each SAD. For example, when LSAD _n is 6 and LSAD ₀ is 17, the weighting coefficient is set to 6.

  Note that the number of sets of weighting factors and the method of setting candidate values are not limited to these four types. The number of sets of weighting coefficients may be one. The method using the correspondence table (table) in FIG. 26 is not limited to this table. Not only thresholds and weight values in the table, but also information that can be shared between the encoding side and decoding side, such as block size, quantization accuracy, and the number of predicted adjacent areas (N) for weighted averaging, etc. A table added as the above condition may be created, or a plurality of tables classified according to these conditions may be prepared. However, in order to automatically select the same set of weight coefficients in the image predictive coding apparatus 100 that is an encoder and the image predictive decoding apparatus 300 that is a decoder, which will be described later, a setting method for both sets of weight coefficients is used. Must match.

  These set of weighting factors are output to the weighted averager 2202 and the comparator / selector 2203 via the line L2201.

  When one piece of prediction adjacent region combination information (N) is input via the line L231, the prediction adjacent region acquisition unit 232 receives coordinate information about the N prediction adjacent regions included in the combination via the line L202a. Obtained from the memory 202. Then, the predicted adjacent region acquisition unit 232 acquires pixel signals of N predicted adjacent regions corresponding to the coordinate information via the line L104, and outputs them to the weighted averager 2202 as needed.

  The weighted averager 2202 includes N pieces of pixel signals of N predicted adjacent regions input via the line L232, and N pieces of N weighting factors included in one set of weighting coefficients input from the weighting coefficient setting unit 2201 via the line L2201. Multiply the weighting factors and add them cumulatively. Then, an addition value of N weighting factors is calculated, and a cumulative addition value of the weighted prediction adjacent region pixel signals is divided by the addition value of the weighting factor, thereby obtaining a pixel signal (adjacent pixel signal) of the target adjacent region. A comparison signal for comparison with is generated.

  The comparison signal is generated for each of a plurality of sets of weighting factors. That is, for the combination information (N) of one prediction adjacent region set by the combination setting unit 231, the processing of the weighted averager 2202 for the four types of weighting factor sets set by the weighting factor setting unit 2201 is repeated. To be implemented.

式（５）は、予測隣接領域の組み合わせ情報Ｎに対して用意されたタイプqの重み係数の組について、比較信号の生成方法を示している。第１に、n番目の予測隣接領域s_n(r)内のR個の画素に、それぞれタイプqの重み係数の組におけるn番目の重み係数w_q,nを掛ける。第２に、重み付けされたＮ個の予測隣接領域w_q,nx s_n(r)を足し合わせ、対象隣接領域内のR個の各画素について、Ｎ個の重み係数w_q,nの加算値で除算する。 The comparison signal generating means will be described using the following equation (5).

Equation (5) shows a method for generating a comparison signal for a set of weighting factors of type q prepared for the combination information N of predicted adjacent regions. First, the R pixels in the _nth predicted adjacent region s _n (r) are each multiplied _{by the nth} weighting factor w _{q, n} in the set of weighting factors of type q. Second, the weighted N predicted adjacent regions w _{q, n} xs _n (r) are added together, _and the added value of N weighting factors w _{q, n} for each of R pixels in the target adjacent region Divide by.

  The target adjacent region acquisition unit 233 acquires the pixel signal (adjacent pixel signal) of the target adjacent region from the frame memory 104 via the line L104 and outputs the pixel signal to the comparison / selector 2203.

  The comparator / selector 2203 inputs the pixel signal of the target adjacent area via the line L233, and inputs a plurality of comparison signals generated for a plurality of sets of weighting factors for one combination of predicted adjacent areas via the line L2203. To do. Then, the SAD is calculated between the comparison signal corresponding to a plurality of sets of weighting factors and the adjacent pixel signal, and the set of weighting factors having the smallest value is selected and temporarily stored.

  Processing for selecting a set of weighting factors (weighting factor setting unit, prediction adjacent region acquisition unit, adder, weight divider, target adjacent region acquisition unit for combination information of one prediction adjacent region output from the combination setting unit The processing of the comparator / selector) is performed on a combination of a plurality of prediction adjacent regions (except when there is one combination of prediction adjacent regions), and the SAD value calculated between adjacent pixel signals is The smallest combination of predicted adjacent regions and the combination of weighting factors are selected. The selected combination of prediction adjacent regions (acquired via L231) is output as combination information of candidate prediction signals to the prediction region acquisition unit 204 via line L203, and the selected combination of weighting factors is the weight of the candidate prediction signal. The coefficient information is output to the weighted averager 2204 via L2203.

  As described above, according to the present embodiment, it is possible to select a combination of candidate prediction signals and a combination of weighting factors that are effective for prediction signal generation without additional information for each target block. If combinations of candidate prediction signals effective for prediction signal generation are determined in advance, there is an effect on selecting an appropriate combination of weighting factors.

  Next, processing of the candidate prediction signal combination determiner 1800 will be described with reference to the drawings. Since the processing of the template matching unit 201 according to the present embodiment has already been described with reference to FIG. 5, description thereof will be omitted, but FIG. 5 illustrates a plurality (M) of pixel signals (target pixel signals) in a target region (target block). Is a flowchart showing a method of searching for a candidate prediction signal (and a pixel signal of a predicted adjacent region) and acquiring coordinate information for accessing the searched candidate prediction signal (and a pixel signal of the predicted adjacent region) A process of searching for pixel signals of M predicted adjacent areas having a small SAD with the adjacent pixel signal of the target adjacent area adjacent to the reference image from the search area on the reference image is performed.

  FIG. 23 shows a combination of N candidate prediction signals suitable for generating a prediction signal in the target region and a weighted average of the N candidate prediction signals in the candidate prediction signal combination determiner 2200 according to the third embodiment. It is a flowchart which shows the method of selecting the group of the weighting factor for doing. First, the combination setter 231 of the candidate prediction signal combination determiner 2200 sets the number N of candidate prediction signals to 1 (S602). Next, the target adjacent area acquisition unit 233 acquires the target adjacent area (template area) for the target block from the frame memory 104 (S603).

  Then, the predicted adjacent area acquisition unit 232 acquires N predicted adjacent areas belonging to the combination set by the combination setting unit 231 via the line L104. That is, the predicted adjacent area acquisition unit 232 receives the adjacent pixel signal of the target adjacent area for the target block and the pixel signal of the area having the same shape as the target adjacent area in the search area on the reference image (predicted adjacent area candidate). Coordinate information corresponding to N predicted adjacent areas having a small SAD, which is the difference value of, is acquired from the coordinate information memory 202. Then, N predicted adjacent areas corresponding to the acquired coordinate information are acquired from the frame memory 104 (S604).

  Next, the weighting factor setting unit 2201 sets a plurality of sets of weighting factors for the N predicted adjacent regions based on the SAD between the adjacent pixel signal and the N predicted adjacent regions, and a combination of candidate prediction signals The determiner 2200 determines weighting factors for the N predicted adjacent regions in a process described later (explained in FIG. 24). Based on the determined weight coefficient, the weighted averager 2202 generates a comparison signal by weighted average of the pixel groups of the N predicted adjacent regions (S2301).

  Thereafter, the comparator / selector 2203 calculates SAD, which is a difference value between the generated comparison signal and the adjacent pixel signal (S606). At the same time, the comparator / selector 2203 compares the calculated SAD with the smallest SAD so far (S607). If the SAD is determined to be the minimum value, the process proceeds to S2302, and if not, the process proceeds to S609. move on. In S607, when the calculated SAD and the minimum SAD so far are the same, the process proceeds to S2302, but the process may proceed to S608.

  If the calculated SAD is the smallest SAD so far, the comparison / selector 2203 stores the combination of the coordinate information acquired in S604 and the combination of the weighting factors determined in S2301 (S2302).

  Here, the combination determiner 2200 updates the value of N by a factor of 2 (S609). Then, the updated N is compared with the magnitude of M (S610). If the updated value of N is smaller than M, the process returns to S604. If the updated value of N is greater than M, the combination of coordinate information stored in S2302 is determined as a combination of candidate prediction signals. Further, the set of weighting coefficients stored in S2302 is determined as “weighting coefficient for weighted average of candidate prediction signals”, and the candidate prediction signal combination selection processing is terminated (S611).

  FIG. 24 is a flowchart showing a method of selecting a set of weighting factors for weighted averaging of N candidate prediction signals in the candidate prediction signal combination determiner 2200 according to the third embodiment. If the value of N is 1, the candidate prediction signal combination determiner 2200 skips this process.

First, the candidate predictive signal combination determiner 2200, for the N predicted adjacent regions searched by the template matching unit 201, SAD (LSADn, n = 0 to N−1) between the adjacent pixel signals of the target adjacent region. ) Is acquired (S2402). Next, the weighting factor setting unit 2201 sets the weighting factor type q to 1 (S2403). Thereafter, the weighting factor setting unit 2201 determines a set of weighting factors according to the set type (q) and the value of N (S2404). In the present embodiment, four types of weight coefficient set candidates are prepared as described below, but the number of weight coefficient sets and the method of setting candidate values are not limited to these four types.
(1) q = 1, equation (6), N weighting factors are set to 1 / N, respectively (2) q = 2, equation 8, target adjacent region t (r) (r is R in the target adjacent region (SAD (LSAD _n between r) (n = 0~N-1 ), formula (7) pieces of the position of the pixel) and the n prediction adjacent regions s _n calculated based on).
(3) q = 3, equation (9), assigning a large weighting factor to the pixel signal of the target adjacent region having a small SAD value (4) q = 4, LSAD ₀ and LSAD _n based on the table shown in FIG. Set the value of w _n from the relationship (n = 0 to N-1).

When the set of weighting factors is set, a comparison signal is generated according to Equation (5) using the pixel signals of the N predicted adjacent regions acquired in S604 and the set of weighting factors (S2405). Equation (5) shows a method for generating a comparison signal for a set of weighting factors of type q prepared for the combination information N of predicted adjacent regions. First, the R pixels in the _nth predicted adjacent region s _n (r) are each multiplied by the nth weighting factor w _q , n in the type q weighting factor set. Second, the weighted N predicted adjacent regions w _{q, n} xs _n (r) are added together, _and the added value of N weighting factors w _{q, n} for each of R pixels in the target adjacent region Divide by.

  Thereafter, the comparator / selector 2203 calculates SAD which is a difference value between the generated comparison signal and the adjacent pixel signal of the target adjacent region (template region) with respect to the target block acquired in S603 (S2406). At the same time, the comparator / selector 2203 compares the calculated SAD with the minimum SAD so far (S2407). If the SAD is determined to be the minimum value, the process proceeds to S2408. If not, the process proceeds to S2409. move on. In S2407, when the calculated SAD and the minimum SAD so far are the same, the process proceeds to S2408, but the process may proceed to S2409.

  When the calculated SAD is the smallest SAD so far, the minimum value is updated to SAD, and the weighting coefficient for the N predicted adjacent regions is updated to the value of type q (S2408).

  Next, the weighting factor setting unit 2201 updates q by adding 1 (S2409). Then, the magnitudes of the updated q and Q (the set number of sets of weighting factors, 4 in the present embodiment) are compared (S2410). If the updated q value is less than or equal to M, the process returns to S2404. When the updated value of q is larger than Q, the set of weighting coefficients for the combination of adjacent prediction areas set by the combination setting unit 231 (composed of N prediction adjacent areas) is used as the combination of the weighting coefficients stored in S2302. And the weighting coefficient selection process ends (S2411). The value of Q is not limited to 4, but may be 1 or more.

  As described above, the weighting coefficient for averaging the pixel signals of the N prediction adjacent regions (and the pixel signal of the target region, that is, the candidate prediction signal) by using the combination processing of the candidate prediction signals of the present embodiment. Can be determined without additional information.

  FIG. 25 is a flowchart showing a method of generating a prediction signal by smoothing (weighted average) candidate prediction signals according to this embodiment.

The prediction region acquisition unit 204 acquires one or more candidate prediction signals corresponding to the target block from the frame memory 104 according to the selected coordinate information (S702). Then, the weighted averager 2204 performs weighted average processing of the candidate prediction signals using the set of weighting coefficients selected from the acquired candidate prediction signals, and generates a prediction signal for the target block (S2501). . This completes the process for one target block (S704). At this time, the generation of the prediction signal Pb (i, j) is calculated based on Expression (4). Here, (i, j) indicates the coordinates of each pixel in the target region. In Equation (4), N indicates combination information of candidate prediction signals input to the prediction region acquisition unit 204 via the line L203, and the prediction region acquisition unit 204 includes N candidate prediction signals p _n (i, j ) (n = 0 to N−1) is acquired from the frame memory 104. w _n indicates a set (N) of weighting factors input to the weighted averager 2204 via the line L2203. n-th weighting coefficient w _n, at the weighted averaging unit 2204, the n-th candidate prediction signal p _n (i, j) multiplied by the respective pixels in each pixel of N weighted candidate prediction signals Are cumulatively added. The cumulative addition value of the candidate prediction signals weighted and added in this way is divided by the addition value of the weighting coefficient to calculate the prediction value Pb (i, j) of each pixel.

  The image predictive encoding method and the image predictive decoding method in the image predictive encoding device 100 according to the third embodiment are the same as the flowcharts shown in FIGS. 8 and 9 and will not be described.

  The image predictive encoding method and the image predictive decoding method according to the third embodiment can be provided by being stored in a recording medium as a program. Examples of the recording medium include a recording medium such as a floppy disk (registered trademark), CD-ROM, DVD, or ROM, or a semiconductor memory. Since the configuration of the program module capable of executing the image predictive encoding method and the module of the program capable of executing the image predictive decoding method are the same as the block diagrams designated in FIGS. To do. The hardware and computer for executing the program recorded on the recording medium have also been described in FIGS. 16 and 17 and will not be described. Note that the prediction signal synthesis module P203 has a function including a prediction region acquisition unit 204 and a weighted averager 2204.

  The third embodiment described so far may be modified as follows.

  In the weighting processing described in Equation (4) and Equation (5), the calculation accuracy of the weighting coefficient is real number accuracy. However, in this embodiment, the weighted averaging process may be performed after each weighting coefficient is converted to an integer. When the weighting factor is a real value, different arithmetic devices may lead to different arithmetic results. However, this problem can be avoided by determining the integerization rule between the encoder and the decoder. At this time, if each weighting coefficient is multiplied by a certain value and then converted into an integer, the calculation accuracy of the weighting process can be kept high.

  In the candidate prediction signal combination determiner 2200 of FIG. 22, the SAD (absolute error value) that is a difference value between the pixel signal of the target adjacent region of the target block and the comparison signal obtained by weighted averaging the pixel signals of the plurality of adjacent prediction regions. The optimal combination of candidate prediction signals is determined by calculating the sum of the two), but the combination can also be determined by using the sum of the square error of the difference signal (SSD) or variance (VAR) instead of SAD. It is. In the three evaluation criteria, the calculation amount increases in the order of SAD, SSD, and VAR. On the other hand, the evaluation accuracy is improved, and the effect that the code amount of the error signal can be reduced is expected.

  It is also effective to determine the final combination using SSD or VAR when a plurality of combinations having the same SAD as the difference value between the target adjacent signal and the comparison signal are obtained. That is, when the SAD calculated in the predicted adjacent region acquisition unit 232 matches the minimum value calculated so far, the comparator / selector 2203 further selects any value using SSD or VAR as a comparison target. Compare what is small. Here, the combination determined to have a small SSD or VAR is stored in the comparator / selector 2203 as a combination having the minimum value. In this case, the comparator / selector 2203 calculates SSD or VAR together with SAD, and temporarily stores it.

The configuration of the prediction signal generator 103b (prediction signal generator 308b) is not limited to the configuration of FIG. For example, in FIG. 22, for a plurality of candidate prediction signals searched by the template matcher, coordinate information for accessing the signals is stored in the coordinate information memory 202. A configuration in which pixel signals are stored may be used. Although the amount of memory in FIG. 22 increases, there is an effect of reducing access to the frame memory.

<Fourth embodiment>
In the third embodiment, the decoder searches for a prediction region and a prediction adjacent region adjacent thereto based on information (for example, a target adjacent signal) added to the target adjacent region. On the other hand, coordinate information (for example, coordinate information 416 in FIG. 4) for accessing the prediction area and the prediction adjacent area, that is, information (motion vector or reference image identification number) added to the target area is encoded by the encoder. Based on this additional information, even when the decoder acquires the prediction area and the prediction adjacent area, the weighting coefficient to be applied when generating the prediction signal of each target area is determined from a plurality of candidates without additional information. The present invention is applicable.

  The encoding device and the decoding device according to the third embodiment can be realized by changing the template matching unit in the prediction generator of FIG. 2 to the motion compensator of FIG. The encoding device and the decoding device are changed as shown in FIG. 27 and FIG. 30 for encoding and decoding the additional information of the target region. The encoding process and the decoding process according to this modification can be realized by changing the template matching process in FIG. 5 to the motion compensation process in FIGS. In the fourth embodiment, the number of candidate prediction signals is fixed to M, and a set of M weighting factors is selected from a plurality of candidates.

  FIG. 28 is a block diagram showing an image predictive encoding device 100b according to the fourth embodiment. The input / output configuration of the prediction signal generator 103 c is different from that of the image prediction encoding device 100 in the block diagram of the image prediction encoding device 100. That is, the pixel signal (target pixel signal) of the target region (target block) is input via L102, and additional information of the target region including coordinate information for accessing each of two or more predicted regions via L2703 is It is output to the entropy encoder 111. The additional information of the target region output from the prediction signal generator 103 c is encoded by the entropy encoder 111 together with the transform coefficient quantized by the quantizer 107.

  Next, the prediction signal generator 103c in the image predictive coding device 100b according to the fourth embodiment will be described. A difference from the prediction signal generator of FIG. 22 is that the template matching unit 201 is changed to a motion compensator.

  FIG. 28 is a block diagram of the motion compensator 201b in the prediction signal generator 103c in the image predictive encoding device 100b according to the fourth embodiment.

  The motion compensator 201b accesses the reproduced image signal stored in the frame memory 104 via the line L104 and performs block matching processing. Here, this block matching processing will be described with reference to FIGS. Here, a process of generating a candidate prediction signal for the target block 402 will be described.

  First, a “search area” is set based on the information added to the target block and the encoded adjacent block. In FIG. 3, a part (or all) of the reproduced images that are in contact with the target block 402 and are reproduced before and within the same screen are set as the search area 401. Here, the target block 402 is a small block of 4 × 4 pixels obtained by dividing an encoding target block made up of 8 × 8 pixels, but may be divided into other block sizes or shapes, or may not be divided. Also good.

  Further, in FIG. 4, a part of the reproduced image shown on the screen 411 different from the target block 402 is set as the search area 417. In this embodiment, the search area is set in two screens (FIGS. 3 and 4), but it may be set only in the same screen as the target block (FIG. 3). You may set only in a different screen (FIG. 4).

  As shown in FIG. 3, the search area 401 and the target block 402 do not need to touch each other, and the search area 401 may not touch the target block 402 at all. Also, as shown in FIG. 4, it is not necessary to limit the search area to be set on one screen (only the screen 411) different from the target block. And may include future frames), and search areas may be set respectively.

  The motion compensator 201b includes a pixel in the target area 402 and a pixel group having the same shape as the target area 402 in the search area 401 and the search area 417 (for each reference frame). In the meantime, a sum (SAD) of absolute error values between corresponding pixels is obtained, and M regions having a small SAD are searched and set as “predicted regions”. Then, an inverted L-shaped region adjacent to the prediction region is referred to as a “prediction adjacent region”. The search accuracy may be in units of integer pixels, or a pixel with decimal precision such as 1/2 pixel or 1/4 pixel may be generated and the search may be performed with decimal pixel precision. The value of M is set in advance, but in FIGS. 3 and 4, M = 6, and the prediction regions 405a, 405b, 405c, 415a, 415b, and 415c are searched. Predicted adjacent areas accompanying these are areas 404a, 404b, 404c, 414a, 414b, and 414c, and additional information of the target area, that is, coordinate information (motion vector and reference image) for accessing each predicted area and predicted adjacent area Are identified by vectors 406a, 406b, 406c, 416a, 416b and 416c.

  Furthermore, the motion compensator 201b calculates SAD between the pixel signal (target adjacent signal) of the “target adjacent region 403” adjacent to the target block 402 and the pixel signals of the M predicted adjacent regions 404 and 414 searched. .

  A configuration of the motion compensator 201b for performing the operation as described above will be described. The motion compensator 201b includes a target adjacent region acquisition unit 211, a predicted adjacent region acquisition unit 212, and a prediction region acquisition unit 2801.

  The prediction region acquisition unit 2801 sequentially acquires pixel signals of a region (prediction region) having the same shape as the target region 402 (target block) from the search region in the frame memory 104 via the line L104, and inputs it via the line L102. The SAD is calculated between the pixel signal (target pixel signal) of the target region 402 thus obtained and the pixel signal of the acquired prediction region (including 405 and 415). Then, M prediction regions having a small SAD are searched from the image group within the search range. Coordinate information (406 and 416, motion vector and reference image identification number) for accessing the searched M prediction regions (405 and 415) is entropy code via line L2703 as additional information of the target region. Is output to the generator 111. At the same time, the M pieces of coordinate information are output to the coordinate information memory 202 and the predicted adjacent area acquisition unit 212 via the line L201a.

  The target adjacent region acquisition unit 211 acquires the target adjacent region 403 from the frame memory 104 via the line L104 and outputs the target adjacent region 403 to the predicted adjacent region acquisition unit 212 via the line L211.

  When the coordinate information of the M prediction regions is input via the line L201a, the prediction adjacent region acquisition unit 212 receives the region having the same shape as the target adjacent region 403 from the search region in the frame memory 104 based on the coordinate information. M pixel signals of the predicted adjacent area are acquired. Then, SAD is calculated between the pixel signal (adjacent pixel signal) of the target adjacent area 403 obtained from the target adjacent area acquisition unit 211 via the line L211 and the acquired pixel signals of the M predicted adjacent areas, and the line L201b To the candidate prediction signal combination determiner 2200.

  In this way, the motion compensator 201b searches for M prediction regions, and outputs coordinate information for accessing these prediction regions to the entropy encoder as additional information of the target region. Then, for the M prediction areas searched, SAD between the pixel signal of the prediction adjacent area adjacent to each prediction area and the pixel signal of the target adjacent area and coordinate information for accessing each prediction adjacent area are generated. And output to the candidate prediction signal combination determiner 2200. Based on the input information, the candidate prediction signal combination determiner 2200 generates a plurality of weight coefficient sets for weighted averaging pixel signals of M prediction regions, as shown in the third embodiment. Select from candidates.

  In FIG. 29, the motion compensator 201b according to the present embodiment searches a plurality (M) of candidate prediction signals for the pixel signal (target pixel signal) in the target region (target block), and accesses the searched candidate prediction signal. It is a flowchart which shows the method of acquiring the coordinate information for doing as additional information of an object area | region.

  First, the target adjacent area acquisition unit 211 acquires the target adjacent area (template signal) for the target block from the frame memory 104. At the same time, the motion compensator 201b acquires the image signal of the target region from the frame memory that stores the input image signal (S2902).

  Next, the motion compensator 201b initializes a threshold value for selecting M candidate prediction signals to a sufficiently large value (S503). The prediction region acquisition unit 2801 acquires a prediction region from the search region, and obtains an absolute value difference sum (SAD) between the pixel signals of the acquired prediction region and the target region (S2903). Further, the SAD value is compared with the threshold value (S505), and if it is determined that the SAD value is smaller than the threshold value, the process proceeds to S2904, and if not, the process proceeds to S508.

  In the process S2904, the coordinate information for accessing the M prediction areas having a small value among the SADs calculated in the process S2903 is stored / updated. At the same time, the motion compensator 201b updates the threshold value to the Mth smallest SAD value (S507).

  Thereafter, the prediction area acquisition unit 2801 checks whether or not all search areas have been searched (S508). When it is determined that not all have been searched, the process returns to S2903, and when all have been searched, the motion compensator 201b performs the process S2905 and ends the motion compensation process (S509). In process S2905, the detected M pieces of coordinate information are encoded as additional information of the target area. Further, the predicted adjacent area acquisition unit 212 calculates the SAD of the pixel signal of the target adjacent area acquired in step S2902 and the pixel signals of M predicted adjacent areas acquired based on the M pieces of coordinate information. .

  FIG. 30 is a block diagram showing an image predictive decoding device 300b according to the fourth embodiment. The input configuration of the prediction signal generator 308c is different from that of the image prediction decoding device 300 in the block diagram of the image prediction decoding device 300b. That is, additional information of the target region including coordinate information for accessing each of two or more prediction regions decoded by the entropy decoder 302 is input to the prediction signal generator 308c.

  Next, the prediction signal generator 103c in the image predictive coding device 100b according to the fourth embodiment will be described. The difference from the prediction signal generator of FIG. 22 is that the template matching unit 201 is changed to a target adjacent region motion compensator 201b.

  FIG. 31 is a block diagram of the target adjacent region motion compensator 201c in the prediction signal generator 308c in the image predictive decoding device 300b according to the fourth embodiment.

  The target adjacent region motion compensator 201c acquires pixel signals of M predicted adjacent regions 404 and 414 based on the M pieces of coordinate information input via the line L309, and acquires the “target adjacent” adjacent to the target block 402. The SAD is calculated with respect to the pixel signal (target adjacent signal) in the “region 403”.

  A configuration of the target adjacent region motion compensator 201c will be described. The motion compensator 201c includes a target adjacent region acquisition unit 211 and a predicted adjacent region acquisition unit 212.

  First, coordinate information for accessing the M prediction regions input via the line L309 is input to the prediction adjacent region acquisition unit 212 and also output to the candidate prediction signal combination determiner 2200.

  The target adjacent region acquisition unit 211 acquires the target adjacent region 403 from the frame memory 104 via the line L104 and outputs the target adjacent region 403 to the predicted adjacent region acquisition unit 212 via the line L211.

  When the coordinate information of M prediction regions is input via the line L309, the prediction adjacent region acquisition unit 212 receives the region having the same shape as the target adjacent region 403 from the search region in the frame memory 104 based on the coordinate information. M pixel signals of the predicted adjacent area are acquired. Then, SAD is calculated between the pixel signal (adjacent pixel signal) of the target adjacent area 403 obtained from the target adjacent area acquisition unit 211 via the line L211 and the acquired pixel signals of the M predicted adjacent areas, and the line L201b To the candidate prediction signal combination determiner 2200.

  As described above, the target adjacent region motion compensator 201c calculates the SAD between the pixel signal of the prediction adjacent region adjacent to each of the M prediction regions and the pixel signal of the target adjacent region, and accesses each prediction adjacent region. Together with the coordinate information for output to the candidate determination signal combination determiner 2200. Based on the input information, the candidate prediction signal combination determiner 2200 generates a plurality of weight coefficient sets for weighted averaging pixel signals of M prediction regions, as shown in the third embodiment. Select from candidates.

  FIG. 32 is a flowchart illustrating processing performed by the target adjacent region motion compensator 201c according to the present embodiment.

  First, the entropy decoder 302 of the decoder decodes the additional information of the target region and decodes the coordinate information for accessing the M prediction regions (S3202). The target adjacent area acquisition unit 211 acquires the target adjacent area (template signal) for the target block from the frame memory 307 (S3203).

  Second, the prediction adjacent region acquisition unit 212 calculates the pixel signal of the target adjacent region acquired in step S3203 and the pixel signals of the M predicted adjacent regions acquired based on the decoded M pieces of coordinate information. SAD is calculated (S3204), and the process ends (S3205).

  Thus, even when a plurality of prediction regions are generated based on information added to the target region, a set of weighting factors for generating a prediction signal of the target region by weighting and averaging these prediction regions. Can be determined from a plurality of candidates without additional information.

  Note that, similarly to the first to third embodiments, the image predictive encoding device 100 and the image predictive decoding device 300 of the fourth embodiment can be configured as a program or a recording medium including the program. The specific module configuration is the same as that shown in FIGS. The module configuration corresponding to the prediction signal generator 103c is the same as that in FIG. 12, but it is necessary to replace the template matching module P201 with a motion compensation module having the function of the motion compensator 201b. The module configuration corresponding to the prediction signal generator 308c has the same configuration as that in FIG. 12, but instead of the template matching module P201, the target adjacent region motion compensation module corresponding to the target adjacent region motion compensator 201c is used. There is a need.

  Next, the operational effects of the image predictive encoding device 100 and the image predictive decoding device 300 of this embodiment will be described.

  In the image predictive encoding device 100 according to the first embodiment, the template matching unit 201 correlates with the target adjacent region 403 including the already-reproduced adjacent pixel signal adjacent to the target region (target block) 402 including the target pixel signal. A plurality of predicted adjacent regions 404a to 404c and 414a to 414c having a high value are searched from the search regions 401 and 417 made of the already reproduced images. The combination setting unit 231 in the candidate prediction signal combination determiner 203 derives two or more combinations of arbitrary predicted adjacent areas including at least one of the searched predicted adjacent areas 404a to 404c and 414a to 414c. Then, the prediction adjacent region acquisition unit 232 extracts the pixel signal of the derived prediction adjacent region, and the weighter 234 and the adder 235 use, for example, an average of the extracted pixel signals using a predetermined synthesis method. As a result, the comparison signal for the adjacent pixel signal is generated for each combination. Then, the comparison / selector 236 selects a combination having a high correlation between the comparison signal generated by the weighting unit 234 and the like and the adjacent pixel signal acquired by the target adjacent region acquisition unit 233. The prediction region acquisition unit 204 generates one or more candidate prediction signals of the target pixel signal based on the prediction adjacent region belonging to the selected combination, and the weighter 205 and the adder 206 determine the candidate prediction signal in advance. A prediction signal is generated by processing using a synthesis method. The subtractor 105 subtracts the prediction signal generated in this way from the target pixel signal acquired via the block divider 102 to generate a residual signal, and the converter 106, the quantizer 107, and the entropy coding The unit 111 encodes the generated residual signal.

  As a result, a combination of candidate prediction signals suitable for smoothing can be selected without increasing the amount of information using a target adjacent region composed of already-reproduced adjacent pixel signals adjacent to the target block. A prediction signal in consideration of noise characteristics can be generated.

  Further, the comparison / selector 236 in the image predictive coding device 100 of the present embodiment is more suitable for smoothing by selecting a combination having a small SAD that is the sum of absolute values of differences between the comparison signal and the adjacent pixel signal. A combination of candidate prediction signals can be selected.

  In addition, the weighter 234 and the adder 235 in the image predictive encoding device 100 according to the present embodiment generate a comparison signal by performing weighted averaging of the pixel signals in the prediction adjacent region belonging to the combination set in the combination setting unit 231. Thus, it is possible to generate an appropriate comparison signal for selecting a combination of candidate prediction signals more suitable for smoothing.

  In addition, the combination of prediction adjacent regions set by the combination setting unit 231 in the image predictive encoding device 100 of the present embodiment includes 2 n predicted adjacent regions in descending order of correlation with the target adjacent region. As a result, it can be performed only by addition and shift operation, and a simple configuration can be achieved. Here, the value of n is preferably an integer of 0 or more.

  Further, in the image predictive decoding device 300 of the present embodiment, the template matching unit 201 has a correlation with the target adjacent region 403 including the already reproduced adjacent pixel signal adjacent to the target region (target block) 402 including the target pixel signal. A plurality of high predicted adjacent areas 404a to 404c and 414a to 414c are searched from the search areas 401 and 417 made of the already reproduced images. The combination setting unit 231 in the candidate prediction signal combination determiner 203 derives two or more combinations of arbitrary predicted adjacent areas including at least one of the searched predicted adjacent areas 404a to 404c and 414a to 414c. Then, the prediction adjacent region acquisition unit 232 extracts the pixel signal of the derived prediction adjacent region, and the weighter 234 and the adder 235 use, for example, an average of the extracted pixel signals using a predetermined synthesis method. As a result, the comparison signal for the adjacent pixel signal is generated for each combination. Then, the comparison / selector 236 selects a combination having a high correlation between the comparison signal generated by the weighting unit 234 and the like and the adjacent pixel signal acquired by the target adjacent region acquisition unit 233. The prediction region acquisition unit 204 generates one or more candidate prediction signals of the target pixel signal based on the prediction adjacent region belonging to the selected combination, and the weighter 205 and the adder 206 determine the candidate prediction signal in advance. A prediction signal is generated by processing using a synthesis method.

  Then, the entropy decoder 302, the inverse quantizer 303, and the inverse transformer 304 restore the difference signal from the compressed data input through the input terminal 301, and the adder 305 includes the predicted signal generated as described above. The restored difference signal is added to generate a reproduced image signal.

  As a result, a combination of candidate prediction signals suitable for smoothing can be selected without increasing the amount of information using a target adjacent region composed of already-reproduced adjacent pixel signals adjacent to the target block. A prediction signal in consideration of noise characteristics can be generated.

  Further, the comparison / selector 236 in the image predictive decoding device 300 of the present embodiment is more suitable for smoothing by selecting a combination having a small SAD that is the sum of absolute values of differences between the comparison signal and the adjacent pixel signal. A combination of candidate prediction signals can be selected.

  In addition, the weighting unit 234 and the adder 235 in the image predictive decoding device 300 according to the present embodiment generate a comparison signal by performing weighted averaging of the pixel signals of the prediction adjacent regions belonging to the combination set by the combination setting unit 231. Thus, it is possible to generate an appropriate comparison signal for selecting a combination of candidate prediction signals more suitable for smoothing.

  In addition, the combination of prediction adjacent regions set by the combination setting unit 231 in the image predictive decoding device 300 of the present embodiment includes 2 n predicted adjacent regions in descending order of correlation with the target adjacent region. Therefore, it can be performed only by addition and shift operation, and a simple configuration in mounting can be achieved. Here, the value of n is preferably an integer of 0 or more.

  Further, in the image predictive coding program P100 of the present embodiment, the template matching module P201 correlates with the target adjacent region 403 including the already reproduced adjacent pixel signal adjacent to the target region (target block) 402 including the target pixel signal. A plurality of predicted adjacent regions 404a to 404c and 414a to 414c having a high value are searched from the search regions 401 and 417 made of the already reproduced images. The candidate prediction signal combination determination module P202 derives two or more combinations of arbitrary predicted adjacent areas including at least one of the searched predicted adjacent areas 404a to 404c and 414a to 414c. Then, the candidate prediction signal combination determination module P202 extracts the pixel signals of the derived prediction adjacent regions, and processes the extracted pixel signals by, for example, averaging using a predetermined synthesis method. The comparison signal for the adjacent pixel signal is generated for each combination. Then, the predicted signal synthesis module P203 selects a combination having a high correlation between the generated comparison signal and the acquired adjacent pixel signal. The candidate prediction signal combination determination module P202 generates one or more candidate prediction signals of the target pixel signal based on the prediction adjacent region belonging to the selected combination, and processes the candidate prediction signals using a predetermined synthesis method. To generate a prediction signal. The subtraction module P105 subtracts the prediction signal generated in this way from the target pixel signal acquired via the block division module P102 to generate a residual signal, and transform module P106, quantization module P107, and entropy coding Module P111 encodes the generated residual signal.

  Further, in the image predictive decoding program P300 of the present embodiment, the template matching module P201 has a correlation with the target adjacent region 403 composed of the already reproduced adjacent pixel signal adjacent to the target region (target block) 402 composed of the target pixel signal. A plurality of high predicted adjacent areas 404a to 404c and 414a to 414c are searched from the search areas 401 and 417 made of the already reproduced images. The predicted signal synthesis module P203 derives two or more combinations of arbitrary predicted adjacent areas including at least one of the searched predicted adjacent areas 404a to 404c and 414a to 414c. Then, the prediction signal synthesis module P203 extracts the pixel signal of the derived prediction adjacent region, and processes the extracted pixel signal by, for example, averaging using a predetermined synthesis method, for each combination. The comparison signals for the adjacent pixel signals are respectively generated. Then, the predicted signal synthesis module P203 selects a combination having a high correlation between the generated comparison signal and the acquired adjacent pixel signal. The prediction signal synthesis module P203 generates one or more candidate prediction signals of the target pixel signal based on the prediction adjacent regions belonging to the selected combination, and the prediction signal synthesis module P203 combines the candidate prediction signals in advance. Is used to generate a prediction signal.

  Then, the entropy decoding module P302, the inverse quantization module P303, and the inverse transform module P304 restore the difference signal from the input compressed data, and the addition module P305 outputs the prediction signal generated as described above and the restored difference signal. Addition to generate a reproduction image signal.

  The second embodiment to which the prediction signal generator 103a is applied in the image predictive encoding device 100 has the following operational effects. In other words, in the prediction signal generator 103a, the prediction adjacent region acquisition unit 212 includes a plurality of prediction adjacent regions that have a high correlation with the target adjacent region including the already reproduced adjacent pixel signals that are adjacent to the target region. Search from the search area. Then, the representative prediction region acquisition unit 1802 outputs pixel signals of N prediction regions based on the target region among the plurality of prediction adjacent regions searched for or pixel signals of the prediction adjacent region among the N prediction adjacent regions searched for. Alternatively, two signals are acquired from both signals.

  Then, a correlation evaluation value calculator 1803 between prediction regions calculates an evaluation value for evaluating the correlation between N candidate prediction signals by a predetermined method. The prediction area acquisition unit 204 acquires N candidate prediction signals when the evaluation value is smaller than a predetermined threshold value, and the weighter 205 and the adder 206 combine predetermined combinations for the N candidate prediction signals. A prediction signal is generated by processing using the method. The subtractor 105 subtracts the generated prediction signal from the target pixel signal to calculate a residual signal, and the converter 106, the quantizer 107, and the entropy encoder 111 encode the residual signal and output it. The terminal 112 outputs.

  Thereby, an appropriate prediction signal can be generated based on a plurality of candidate prediction signals. This is particularly effective for signals that differ only in the noise component of the candidate prediction signal but have similar signal components.

  Also, in the prediction signal generator 103a, when the threshold value processor 1804 determines that the evaluation value is larger than the threshold value, the counter 1801 updates the value to the next largest value by subtracting the N value, and the representative prediction region acquisition unit 1802 re-acquires a combination of pixel signals, a correlation evaluation value calculator 1803 between prediction regions calculates an evaluation value again, and a threshold value processor 1804 recalculates the evaluation value and compares it with a prescribed threshold value. The prediction region acquisition unit 204 can select the number of valid candidate prediction signals.

  Further, in the prediction signal generator 103a, the correlation evaluation value calculator 1803 between prediction regions sets the prediction adjacent region having the highest correlation with the target adjacent region among the N prediction adjacent regions searched by the template matching unit 201. The pixel signal of the prediction adjacent region adjacent to the prediction adjacent region having the Nth lowest correlation between the pixel signal of the adjacent prediction region and the target adjacent region, or the pixel signal of the prediction adjacent region having the highest correlation with the target adjacent region and the target adjacent The absolute value sum of the pixel signals of the prediction adjacent region having the Nth lowest correlation with the region, or the difference between the signals combining the pixel signals is calculated, and the absolute value sum of the differences is used as the evaluation value. Thereby, an appropriate evaluation value can be calculated.

  Further, when the prediction signal generator 308a is applied to the image predictive decoding device 300, the following operational effects are obtained. The predicted adjacent area acquisition unit 212 searches a plurality of predicted adjacent areas having a high correlation with the target adjacent area including the already-reproduced adjacent pixel signals adjacent to the target area from the search area including the already-reproduced image. Then, the representative prediction region acquisition unit 1802 outputs pixel signals of N prediction regions based on the target region among the plurality of prediction adjacent regions searched for or pixel signals of the prediction adjacent region among the N prediction adjacent regions searched for. Alternatively, two signals are acquired from both signals.

  Then, a correlation evaluation value calculator 1803 between prediction regions calculates an evaluation value for evaluating the correlation between N candidate prediction signals by a predetermined method. The prediction area acquisition unit 204 acquires N candidate prediction signals when the evaluation value is smaller than a predetermined threshold value, and the weighter 205 and the adder 206 combine predetermined combinations for the N candidate prediction signals. A prediction signal is generated by processing using the method.

  On the other hand, the entropy decoder 302, the inverse quantizer 303, and the inverse transformer 304 restore the differential signal from the compressed data input through the input terminal 301, and the adder 305 includes the prediction signal generated as described above. The restored difference signal is added to generate a reproduced image signal.

  Thereby, an appropriate prediction signal can be generated based on a plurality of candidate prediction signals. This is particularly effective for signals that differ only in the noise component of the candidate prediction signal but have similar signal components.

  Also, in the prediction signal generator 103a, when the threshold value processor 1804 determines that the evaluation value is larger than the threshold value, the counter 1801 updates the value to the next largest value by subtracting the N value, and the representative prediction region acquisition unit 1802 re-acquires a combination of pixel signals, a correlation evaluation value calculator 1803 between prediction regions calculates an evaluation value again, and a threshold value processor 1804 recalculates the evaluation value and compares it with a prescribed threshold value. The prediction region acquisition unit 204 can select the number of valid candidate prediction signals.

  Further, in the prediction signal generator 103a, the correlation evaluation value calculator 1803 between prediction regions sets the prediction adjacent region having the highest correlation with the target adjacent region among the N prediction adjacent regions searched by the template matching unit 201. The pixel signal of the prediction adjacent region adjacent to the prediction adjacent region having the Nth lowest correlation between the pixel signal of the adjacent prediction region and the target adjacent region, or the pixel signal of the prediction adjacent region having the highest correlation with the target adjacent region and the target adjacent The absolute value sum of the pixel signals of the prediction adjacent region having the Nth lowest correlation with the region, or the difference between the signals combining the pixel signals is calculated, and the absolute value sum of the differences is used as the evaluation value. Thereby, an appropriate evaluation value can be calculated.

  The third embodiment in which the prediction signal generator 103b is applied to the image predictive encoding device 100 has the following operational effects. Based on the target area or the target adjacent area adjacent to the target area, the template matching unit 201 acquires a plurality of predicted adjacent areas having the same shape as the target adjacent area from the search area including the already reproduced images.

  Then, the combination setter 231 derives an arbitrary combination of predicted adjacent areas including two or more of the predicted adjacent areas among the plurality of acquired predicted adjacent areas. Next, the weighting factor setting unit 2201 derives two or more sets of weighting factors for performing weighted averaging of pixel signals in the prediction adjacent region belonging to the combination, and the weighting averager 2202 calculates the plurality of combinations for the combinations. Two or more comparison signals for the adjacent pixel signals are generated by performing weighted averaging of the pixel signals of the predicted adjacent regions belonging to these combinations using a set of weighting factors.

  Next, the comparator / selector 2203 selects a set of weighting coefficients having a high correlation between the comparison signal and the adjacent pixel signal, and the prediction area acquisition unit 204 determines the already reproduced image based on the prediction adjacent area belonging to the combination. Two or more candidate prediction signals of the target pixel signal are generated from the above, and the weighted averager 2204 generates a prediction signal by performing weighted averaging of the candidate prediction signals using a selected set of weighting coefficients. Then, the subtractor 105 generates a residual signal by subtracting the generated prediction signal from the target pixel signal, and the converter 106, the quantizer 107, and the entropy encoder 111 encode the residual signal. . This makes it possible to select a set of weighting factors without additional information for each target block.

  Further, in the image encoding device 100 according to the third embodiment, the template matching unit 201 uses a target adjacent region or a target adjacent region adjacent to the target region to generate a predicted adjacent region having the same shape as the target adjacent region. A plurality of images are obtained from the search area made up of reproduced images.

  Then, the combination setter 231 derives a combination of arbitrary predicted adjacent areas including one or more of the predicted adjacent areas among the plurality of acquired predicted adjacent areas.

  Then, the weighting factor setting unit 2201 derives two or more sets of weighting factors for weighted average of the pixel signals of the predicted adjacent regions belonging to the combination of combinations including two or more predicted adjacent regions, and performs weighted averaging. The unit 2202 generates two or more comparison signals for adjacent pixel signals by performing weighted averaging of the pixel signals in the predicted adjacent region using the plurality of sets of weighting factors.

  Next, the comparator / selector 2203 selects a set of weighting coefficients having a high correlation between the comparison signal and the adjacent pixel signal, and the weighted averager 2202 again predicts adjacent regions belonging to the selected combinations. Are compared with each other using a set of selected weighting factors, and two or more comparison signals for the adjacent pixel signal are generated. Again, the comparison / selector 2203 A combination having a high correlation between the comparison signal and the adjacent pixel signal is selected.

  Then, the prediction region acquisition unit 204 generates one or more candidate prediction signals of the target pixel signal from the already reproduced image based on the prediction adjacent region belonging to the selected combination, and the weighted averager 2204 includes the candidate prediction signal. A predicted signal is generated by weighted averaging the signal using the previously selected combination of weighting factors for the selected combination. Then, a residual signal with the target pixel signal is generated using the generated prediction signal, and the residual signal is encoded. Thereby, for each target block, it is possible to select a combination of candidate prediction signals and a set of weighting coefficients that are effective for generating a prediction signal without additional information.

  Further, in the prediction signal generator 103b, the weighting coefficient setting unit 2201 sets a smaller weighting coefficient as the absolute value sum of the differences between the pixel signals of the plurality of prediction adjacent regions belonging to the combination and the preceding pixel signal increases. As shown (for example, as shown in equation (7)), an appropriate set of weighting factors can be calculated by calculating at least one set of weighting factors.

  In the prediction signal generator 103b, the weighting factor setting unit 2201 prepares in advance a set of weighting factors (for example, as shown in Expression (6)) determined according to the number of prediction adjacent regions belonging to the combination. Thus, by deriving at least one set of weight coefficients from the prepared set of weight coefficients, an appropriate set of weight coefficients can be calculated.

  Further, in the prediction signal generator 103b, the weighting factor setting unit 2201 is a correspondence table in which a set of weighting factors is determined based on the sum of absolute values of differences between the pixel signals of a plurality of prediction adjacent regions belonging to the combination and the adjacent pixel signals (for example, As shown in FIG. 26, an appropriate weight coefficient set can be calculated by deriving at least one set of weight coefficients using the correspondence table.

  Further, when the prediction signal generator 308b is applied to the image predictive decoding device 300, the following operational effects are obtained. Based on the target area or the target adjacent area adjacent to the target area, the template matching unit 201 acquires a plurality of predicted adjacent areas having the same shape as the target adjacent area from the search area including the already reproduced images.

  Then, the combination setter 231 derives an arbitrary combination of predicted adjacent areas including two or more of the predicted adjacent areas among the plurality of acquired predicted adjacent areas. Next, the weighting factor setting unit 2201 derives two or more sets of weighting factors for performing weighted averaging of pixel signals in the prediction adjacent region belonging to the combination, and the weighting averager 2202 calculates the plurality of combinations for the combinations. Two or more comparison signals for the adjacent pixel signals are generated by performing weighted averaging of the pixel signals of the predicted adjacent regions belonging to these combinations using a set of weighting factors.

  Next, the comparator / selector 2203 selects a set of weighting coefficients having a high correlation between the comparison signal and the adjacent pixel signal, and the prediction area acquisition unit 204 determines the already reproduced image based on the prediction adjacent area belonging to the combination. Two or more candidate prediction signals of the target pixel signal are generated from the above, and the weighted averager 2204 generates a prediction signal by performing weighted averaging of the candidate prediction signals using a selected set of weighting coefficients. Then, the subtractor 105 generates a residual signal by subtracting the generated prediction signal from the target pixel signal, and the converter 106, the quantizer 107, and the entropy encoder 111 encode the residual signal. . This makes it possible to select a set of weighting factors without additional information for each target block.

  Further, in the image encoding device 100 according to the third embodiment, the template matching unit 201 uses a target adjacent region or a target adjacent region adjacent to the target region to generate a predicted adjacent region having the same shape as the target adjacent region. A plurality of images are obtained from the search area made up of reproduced images.

  Then, the combination setter 231 derives a combination of arbitrary predicted adjacent areas including one or more of the predicted adjacent areas among the plurality of acquired predicted adjacent areas.

  Then, the weighting factor setting unit 2201 derives two or more sets of weighting factors for weighted average of the pixel signals of the predicted adjacent regions belonging to the combination of combinations including two or more predicted adjacent regions, and performs weighted averaging. The unit 2202 generates two or more comparison signals for adjacent pixel signals by performing weighted averaging of the pixel signals in the predicted adjacent region using the plurality of sets of weighting factors.

  Next, the comparator / selector 2203 selects a set of weighting coefficients having a high correlation between the comparison signal and the adjacent pixel signal, and the weighted averager 2202 again predicts adjacent regions belonging to the selected combinations. Are compared with each other using a set of selected weighting factors, and two or more comparison signals for the adjacent pixel signal are generated. Again, the comparison / selector 2203 A combination having a high correlation between the comparison signal and the adjacent pixel signal is selected.

  Then, the prediction region acquisition unit 204 generates one or more candidate prediction signals of the target pixel signal from the already reproduced image based on the prediction adjacent region belonging to the selected combination, and the weighted averager 2204 includes the candidate prediction signal. A predicted signal is generated by weighted averaging the signal using the previously selected combination of weighting factors for the selected combination.

  On the other hand, the entropy decoder 302, the inverse quantizer 303, and the inverse transformer 304 restore the differential signal from the compressed data input through the input terminal 301, and the adder 305 includes the prediction signal generated as described above. The restored difference signal is added to generate a reproduced image signal. Thereby, for each target block, it is possible to select a combination of candidate prediction signals and a set of weighting coefficients that are effective for generating a prediction signal without additional information.

  Further, in the prediction signal generator 103b, the weighting coefficient setting unit 2201 sets a smaller weighting coefficient as the absolute value sum of the differences between the pixel signals of the plurality of prediction adjacent regions belonging to the combination and the preceding pixel signal increases. As shown (for example, as shown in equation (7)), an appropriate set of weighting factors can be calculated by calculating at least one set of weighting factors.

  Further, in the prediction signal generator 103b, the weighting factor setting unit 2201 prepares in advance a set of weighting factors (as shown in the equation (6)) determined according to the number of prediction adjacent regions belonging to the combination, By deriving at least one set of weighting factors from the prepared set of weighting factors, an appropriate set of weighting factors can be calculated.

  In the prediction signal generator 103b, the weighting factor setting unit 2201 is a correspondence table in which a set of weighting factors is determined based on the sum of absolute values of differences between pixel signals in a plurality of prediction adjacent regions belonging to the combination and adjacent pixel signals (FIG. 26). And a suitable weight coefficient set can be calculated by deriving at least one set of weight coefficients using the correspondence table.

  As the fourth embodiment, the third embodiment is configured to perform motion compensation. For example, in the third embodiment, the template matching unit 201 extracts a predicted adjacent area having the same shape as the target adjacent area based on the target area or the target adjacent area adjacent to the target area from the search area including the already reproduced image. However, the present invention is not limited to this. For example, the predicted adjacent region is acquired based on additional information added to the target region, such as a motion vector or a reference image identification number. Also good.

It is a block diagram which shows the image predictive coding apparatus by embodiment of this invention. It is a block diagram which shows the prediction signal generator 103 used for an image predictive coding apparatus. FIG. 10 is a schematic diagram for explaining template matching processing and processing for searching for a prediction adjacent region and a prediction region candidate in the template matching unit 201. FIG. 11 is a second schematic diagram for explaining template matching processing and processing for searching for a prediction adjacent region and a prediction region candidate in the template matching unit 201. 10 is a flowchart for explaining a template matching in template matching unit 201 and a method of searching for a prediction adjacent region and a prediction region candidate. It is a flowchart for demonstrating the combination determination method of the candidate prediction signal in the combination determination unit 203 of a candidate prediction signal. It is a flowchart for demonstrating the method of synthesize | combining a candidate prediction signal and producing | generating a prediction signal. 3 is a flowchart illustrating an image predictive encoding method in the image predictive encoding device 100. It is a block diagram which shows the image prediction decoding apparatus. It is a flowchart which shows the image prediction decoding method of the image prediction decoding apparatus 300. It is a block diagram which shows the module of the program which can perform the image predictive coding method. It is a block diagram which shows the module of the prediction signal generation module P103. It is a block diagram which shows the module of the program which can perform the image prediction decoding method. It is a schematic diagram for demonstrating the prediction method in a screen. It is a schematic diagram for demonstrating a block matching process. 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 block diagram of the prediction signal generator 103 used for the image predictive coding apparatus in 2nd embodiment. 18 is a flowchart for explaining a method of determining a combination of candidate prediction signals in an averaging number determiner 1800. It is a flowchart which shows the 2nd method which synthesize | combines a candidate prediction signal and produces | generates a prediction signal. It is a block diagram which shows the 2nd module of the prediction signal generation module P103. It is a block diagram which shows the prediction signal generator 103 used for the image predictive coding apparatus in 3rd embodiment. 12 is a flowchart for explaining a weight coefficient determination method for candidate prediction signals in a weight coefficient setting unit 2201. It is a flowchart for demonstrating the 3rd combination determination method of the candidate prediction signal in the combination determination unit 2200 of a candidate prediction signal. It is a flowchart which shows the 3rd method which synthesize | combines a candidate prediction signal and produces | generates a prediction signal. It is explanatory drawing which shows the correspondence table which sets the group of a weighting factor from the difference value between the pixel signal of a target adjacent area | region, and the pixel signal of a prediction adjacent area | region. It is a block diagram which shows the image predictive coding apparatus by 4th embodiment of this invention. It is a block diagram which shows the motion compensator 201b in the prediction signal generator 103c of the image predictive coding apparatus by 4th embodiment of this invention. It is a flowchart which shows the motion compensation method in the motion compensator 201b. It is a block diagram which shows the image prediction decoding apparatus by 4th embodiment of this invention. It is a block diagram which shows the object adjacent area | region motion compensator 201c in the prediction signal generator 308c of the image prediction decoding apparatus by 4th embodiment of this invention. It is a flowchart which shows the motion compensation method in the object adjacent area | region motion compensator 201c.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 ... Image predictive coding apparatus, 101 ... Input terminal, 102 ... Block divider, 103 ... Prediction signal generator, 104 ... Frame memory, 105 ... Subtractor, 106 ... Converter, 107 ... Quantizer, 108 ... Inverse Quantizer, 109 ... Inverse transformer, 110 ... Adder, 111 ... Entropy encoder, 112 ... Output terminal, 201 ... Template matching unit, 202 ... Memory for coordinate information, 203 ... Combination determiner of candidate prediction signals, 204 ... Predictive region acquisition unit, 205 ... Weighter, 206 ... Adder, 211 ... Target adjacent region acquisition unit, 212 ... Predictive adjacent region acquisition unit, 213 ... Comparator, 214 ... Switch, 231 ... Combination setting unit, 232 ... Predictive adjacent region acquisition unit, 233... Target adjacent region acquisition unit, 234... Weighting unit, 235... Adder, 236. Image predictive decoding apparatus, 301 ... input terminal, 302 ... entropy decoder, 303 ... inverse quantizer, 304 ... inverse transformer, 305 ... adder, 306 ... output terminal, 307 ... frame memory, 308 ... prediction signal generator , P100: Image predictive coding program, P102: Block division module, P103: Prediction signal generation module, P104: Storage module, P105: Subtraction module P106: Conversion module, P108: Inverse quantization module, P109: Inverse conversion module, P110 ... addition module, P111 ... entropy encoding module, P201 ... template matching module, P202 ... decision module, P203 ... predictive signal synthesis module, P300 ... image prediction decoding program, P302 ... entropy decoding module, 303 ... Inverse quantization module, P304 ... Inverse transformation module, P305 ... Addition module, P307 ... Storage module, P308 ... Prediction signal generation module, 1800 ... Averaging number determiner, 1801 ... Counter, 1802 ... Representative prediction region acquisition unit, 1803 ... correlation evaluation value calculator, 1804 ... threshold value processor, 2200 ... candidate prediction signal combination determiner, 2201 ... weighting coefficient setting unit, 2202 ... weighted averager, 2203 ... comparison / selector, 2204 ... weighted average , 2801 ... Predictive region acquisition unit.

Claims (42)

Area dividing means for dividing the input image into a plurality of areas;
A prediction signal generation unit that generates a prediction signal for a target pixel signal of a target region that is a processing target among the plurality of regions divided by the region dividing unit;
Residual signal generation means for generating a residual signal between the prediction signal generated by the prediction signal generation means and the target pixel signal;
Encoding means for encoding the residual signal generated by the residual signal generating means,
The predicted signal generation means searches a plurality of predicted adjacent areas having a high correlation with a target adjacent area consisting of an already reproduced adjacent pixel signal adjacent to the target area consisting of the target pixel signal from a search area consisting of an already reproduced image. And
Two or more combinations of arbitrary predicted adjacent areas including at least one of the searched plurality of predicted adjacent areas are derived, and pixel signals of the predicted adjacent areas belonging to the combination are processed using a predetermined synthesis method. Thus, a comparison signal for the adjacent pixel signal is generated for each combination, and a combination having a high correlation between the comparison signal and the adjacent pixel signal is selected.
Generate one or more candidate prediction signals of the target pixel signal based on the prediction adjacent region belonging to the selected combination, and generate the prediction signal by processing the candidate prediction signal using a predetermined synthesis method An image predictive coding apparatus characterized by:   The image predictive coding apparatus according to claim 1, wherein the prediction signal generation unit selects a combination having a small sum of absolute values of differences between the comparison signal and the adjacent pixel signal.   The image predictive coding apparatus according to claim 1, wherein the prediction signal generation unit generates the comparison signal by performing weighted averaging of pixel signals of prediction adjacent regions belonging to the combination.   4. The prediction signal generation unit according to claim 1, wherein the combination of the predicted adjacent areas includes 2 n predicted adjacent areas in descending order of correlation with the target adjacent area. The image predictive encoding device according to the item.   The image predictive coding apparatus according to claim 4, wherein the value of n is an integer of 0 or more. Data decoding means for decoding encoded data of a residual signal related to a target region to be processed from among compressed data;
Residual signal restoration means for restoring a reproduction residual signal from a signal obtained by decoding by the data decoding means;
Prediction signal generating means for generating a prediction signal for the target pixel signal of the target region;
Reproduction image signal generation means for generating a reproduction image signal by adding the prediction signal generated by the prediction signal generation means and the reproduction residual signal restored by the residual signal restoration means, and
The predicted signal generation means searches a plurality of predicted adjacent areas having a high correlation with a target adjacent area consisting of an already reproduced adjacent pixel signal adjacent to the target area consisting of the target pixel signal from a search area consisting of an already reproduced image. And
Two or more combinations of arbitrary predicted adjacent areas including at least one of the searched plurality of predicted adjacent areas are derived, and pixel signals of the predicted adjacent areas belonging to the combination are processed using a predetermined synthesis method To generate a comparison signal for the adjacent pixel signal for each combination, and select a combination having a large correlation between the comparison signal and the adjacent pixel signal,
One or more candidate prediction signals of the target pixel signal are generated based on the prediction adjacent region belonging to the selected combination, and a prediction signal is generated by processing the candidate prediction signal using a predetermined synthesis method An image predictive decoding apparatus characterized by:   The said prediction signal production | generation means produces | generates the comparison signal with respect to the said adjacent pixel signal, and selects the combination with a small absolute value sum of the difference of this comparison signal and the said adjacent pixel signal. Image predictive decoding apparatus.   The image predictive decoding apparatus according to claim 6 or 7, wherein the prediction signal generation unit generates the comparison signal by performing weighted averaging of pixel signals of prediction adjacent regions belonging to the combination.   9. The prediction signal generation unit according to claim 6, wherein the combination of predicted adjacent areas includes 2 n predicted adjacent areas in descending order of correlation with the target area. The image predictive decoding apparatus described in 1.   The image predictive decoding apparatus according to claim 9, wherein the value of n is an integer of 0 or more. A region dividing step for dividing the input image into a plurality of regions;
A prediction signal generating step for generating a prediction signal for a target pixel signal of a target region that is a processing target among the plurality of regions divided by the region dividing step;
A residual signal generation step for generating a residual signal between the prediction signal generated by the prediction signal generation step and the target pixel signal;
An encoding step for encoding the residual signal generated by the residual signal generation step,
The predicted signal generation step searches a plurality of predicted adjacent areas having a high correlation with a target adjacent area composed of already reproduced adjacent pixel signals adjacent to the target area composed of the target pixel signal from a search area composed of a previously reproduced image. And
Two or more combinations of arbitrary predicted adjacent areas including at least one of the searched plurality of predicted adjacent areas are derived, and pixel signals of the predicted adjacent areas belonging to the combination are processed using a predetermined synthesis method. Thus, a comparison signal for the adjacent pixel signal is generated for each combination, and a combination having a high correlation between the comparison signal and the adjacent pixel signal is selected.
Generate one or more candidate prediction signals of the target pixel signal based on the prediction adjacent region belonging to the selected combination, and generate the prediction signal by processing the candidate prediction signal using a predetermined synthesis method An image predictive encoding method characterized by: A data decoding step for decoding encoded data of a residual signal related to a target region to be processed from the compressed data;
A residual signal restoration step for restoring a reproduction residual signal from the signal decoded by the data decoding step;
A prediction signal generation step of generating a prediction signal for the target pixel signal of the target region;
A reproduction image signal generation step of generating a reproduction image signal by adding the prediction signal generated in the prediction signal generation step and the reproduction residual signal restored in the residual signal restoration step, and
The predicted signal generation step searches a plurality of predicted adjacent areas having a high correlation with a target adjacent area composed of already reproduced adjacent pixel signals adjacent to the target area composed of the target pixel signal from a search area composed of a previously reproduced image. And
Two or more combinations of arbitrary predicted adjacent areas including at least one of the searched plurality of predicted adjacent areas are derived, and pixel signals of the predicted adjacent areas belonging to the combination are processed using a predetermined synthesis method. Accordingly, a comparison signal for the adjacent pixel signal is generated for each combination, and a combination having a large correlation between the comparison signal and the adjacent pixel signal is selected.
One or more candidate prediction signals of the target pixel signal are generated based on the prediction adjacent region belonging to the selected combination, and a prediction signal is generated by processing the candidate prediction signal using a predetermined synthesis method An image predictive decoding method characterized by: An area division module for dividing an input image into a plurality of areas;
A prediction signal generation module that generates a prediction signal for a target pixel signal of a target region that is a processing target among the plurality of regions divided by the region division module;
A residual signal generation module that generates a residual signal between the prediction signal generated by the prediction signal generation module and the target pixel signal;
An encoding module that encodes the residual signal generated by the residual signal generation module;
The predicted signal generation module searches a plurality of predicted adjacent areas having a high correlation with a target adjacent area composed of already reproduced adjacent pixel signals adjacent to the target area composed of the target pixel signal from a search area composed of a previously reproduced image. And
Two or more combinations of arbitrary predicted adjacent areas including at least one of the searched plurality of predicted adjacent areas are derived, and pixel signals of the predicted adjacent areas belonging to the combination are processed using a predetermined synthesis method. Thus, a comparison signal for the adjacent pixel signal is generated for each combination, and a combination having a high correlation between the comparison signal and the adjacent pixel signal is selected.
Generate one or more candidate prediction signals of the target pixel signal based on the prediction adjacent region belonging to the selected combination, and generate the prediction signal by processing the candidate prediction signal using a predetermined synthesis method An image predictive encoding program characterized by: A data decoding module that decodes encoded data of a residual signal related to a target region to be processed from the compressed data;
A residual signal restoration module for restoring a reproduction residual signal from the signal decoded by the data decoding module;
A prediction signal generation module that generates a prediction signal for a target pixel signal in the target region;
A reproduction image signal generation module that generates a reproduction image signal by adding the prediction signal generated by the prediction signal generation module and the reproduction residual signal restored by the residual signal restoration module;
The predicted signal generation module searches a plurality of predicted adjacent areas having a high correlation with a target adjacent area composed of already reproduced adjacent pixel signals adjacent to the target area composed of the target pixel signal from a search area composed of a previously reproduced image. And
Two or more combinations of arbitrary predicted adjacent areas including at least one of the searched plurality of predicted adjacent areas are derived, and pixel signals of the predicted adjacent areas belonging to the combination are processed using a predetermined synthesis method. Accordingly, a comparison signal for the adjacent pixel signal is generated for each combination, and a combination having a large correlation between the comparison signal and the adjacent pixel signal is selected.
One or more candidate prediction signals of the target pixel signal are generated based on the prediction adjacent region belonging to the selected combination, and a prediction signal is generated by processing the candidate prediction signal using a predetermined synthesis method An image predictive decoding program characterized by: Area dividing means for dividing the input image into a plurality of areas;
A prediction signal generation unit that generates a prediction signal for a target pixel signal of a target region that is a processing target among the plurality of regions divided by the region dividing unit;
Residual signal generation means for generating a residual signal between the prediction signal generated by the prediction signal generation means and the target pixel signal;
Encoding means for encoding the residual signal generated by the residual signal generating means,
The prediction signal generating means includes
Search for a plurality of predicted adjacent areas having a high correlation with the target adjacent area consisting of the already reproduced adjacent pixel signal adjacent to the target area consisting of the target pixel signal from the search area consisting of the already reproduced image,
At least from the pixel signals of the N prediction regions based on the target region, the pixel signals of the prediction adjacent region among the N prediction adjacent regions searched for among the plurality of predicted adjacent regions searched, or the signals of both Using two signals, an evaluation value for evaluating the correlation between the N candidate prediction signals is calculated by a predetermined method,
An image predictive coding apparatus, wherein when the evaluation value is smaller than a predetermined threshold value, a prediction signal is generated by processing the N candidate prediction signals using a predetermined synthesis method. The prediction signal generation means prepares a plurality of candidates as the value of N, calculates the evaluation value for the maximum value of the N value candidates, and when the evaluation value is smaller than a prescribed threshold, Generating a prediction signal by processing the candidate prediction signal using a predetermined synthesis method;
16. When the evaluation value is larger than the threshold value, the N value is subtracted to be updated to the next largest value, and the calculation of the evaluation value is compared with a prescribed threshold value again. The image predictive coding apparatus described in 1.   The prediction signal generation means has the Nth correlation between the pixel signal of the prediction area adjacent to the prediction adjacent area having the highest correlation with the target adjacent area and the target adjacent area among the N predicted adjacent areas searched for. A pixel signal of a prediction region adjacent to a low prediction adjacent region, a pixel signal of a prediction adjacent region having the highest correlation with the target adjacent region, and a pixel signal of a prediction adjacent region having the Nth lowest correlation between the target adjacent region, or 17. The image predictive coding apparatus according to claim 15, wherein an absolute value sum of a difference between signals including the pixel signals is calculated, and the absolute value sum of the differences is used as the evaluation value. Data decoding means for decoding encoded data of a residual signal related to a target region to be processed from among compressed data;
Residual signal restoration means for restoring a reproduction residual signal from a signal obtained by decoding by the data decoding means;
Prediction signal generating means for generating a prediction signal for the target pixel signal of the target region;
Reproduction image signal generation means for generating a reproduction image signal by adding the prediction signal generated by the prediction signal generation means and the reproduction residual signal restored by the residual signal restoration means, and
The predicted signal generation means searches a plurality of predicted adjacent areas having a high correlation with a target adjacent area consisting of an already reproduced adjacent pixel signal adjacent to the target area consisting of the target pixel signal from a search area consisting of an already reproduced image. And
At least from the pixel signals of the N prediction regions based on the target region, the pixel signals of the prediction adjacent region among the N prediction adjacent regions searched for among the plurality of predicted adjacent regions searched, or the signals of both Using two signals, an evaluation value for evaluating the correlation between the N candidate prediction signals is calculated by a predetermined method,
An image predictive decoding apparatus that generates a prediction signal by processing the N candidate prediction signals using a predetermined synthesis method when the evaluation value is smaller than a predetermined threshold. The prediction signal generation means prepares a plurality of candidates as the value of N, calculates the evaluation value for the maximum value of the N value candidates, and when the evaluation value is smaller than a prescribed threshold, Generating a prediction signal by processing the candidate prediction signal using a predetermined synthesis method;
19. When the evaluation value is larger than the threshold value, the N value is subtracted to be updated to the next largest value, and the evaluation value is calculated and compared with a specified threshold value again. The image predictive decoding apparatus described in 1.   The prediction signal generation means predicts the correlation between the pixel signal of the prediction area adjacent to the prediction adjacent area having the highest correlation with the target adjacent area and the target adjacent area among the N predicted adjacent areas searched for. The pixel signal of the prediction adjacent region adjacent to the adjacent region, the pixel signal of the prediction adjacent region having the highest correlation with the target adjacent region, and the pixel signal of the prediction adjacent region having the Nth lowest correlation between the target adjacent region, or the pixel signal The image predictive decoding apparatus according to claim 18 or 19, wherein the sum of absolute values of the differences of the signals combined with each other is calculated, and the sum of the absolute values of the differences is used as the evaluation value. Area dividing means for dividing the input image into a plurality of areas;
A prediction signal generation unit that generates a prediction signal for a target pixel signal of a target region that is a processing target among the plurality of regions divided by the region dividing unit;
Residual signal generation means for generating a residual signal between the prediction signal generated by the prediction signal generation means and the target pixel signal;
Encoding means for encoding the residual signal generated by the residual signal generating means,
The predicted signal generation means obtains a plurality of predicted adjacent areas having the same shape as the target adjacent area adjacent to the target area from the search area composed of already reproduced images,
A weighting factor for deriving a combination of arbitrary prediction adjacent areas including two or more of the obtained prediction adjacent areas from among the plurality of obtained prediction adjacent areas, and weighted averaging pixel signals of the prediction adjacent areas belonging to the combination Deriving two or more sets of
For the combination, by generating a weighted average of the pixel signals of the prediction adjacent region belonging to the combination using the set of the plurality of weighting factors, two or more comparison signals for the adjacent pixel signal are generated,
Selecting a set of weighting coefficients having a high correlation between the comparison signal and the adjacent pixel signal;
Two or more candidate prediction signals of the target pixel signal are generated from the already reproduced image based on the prediction adjacent region belonging to the combination, and the candidate prediction signals are weighted and averaged using the selected set of weighting factors. An image predictive coding apparatus characterized by generating a prediction signal by means of Area dividing means for dividing the input image into a plurality of areas;
A prediction signal generation unit that generates a prediction signal for a target pixel signal of a target region that is a processing target among the plurality of regions divided by the region dividing unit;
Residual signal generation means for generating a residual signal between the prediction signal generated by the prediction signal generation means and the target pixel signal;
Encoding means for encoding the residual signal generated by the residual signal generating means,
The predicted signal generation means obtains a plurality of predicted adjacent areas having the same shape as the target adjacent area adjacent to the target area from the search area composed of already reproduced images,
Deriving two or more combinations of arbitrary predicted adjacent regions including at least one of the obtained plurality of predicted adjacent regions,
For a combination including two or more predicted adjacent areas, two or more sets of weighting factors for weighted averaging pixel signals of the predicted adjacent areas belonging to the combination are derived, and predicted adjacents are generated using the plurality of sets of weighting coefficients. Two or more comparison signals for the adjacent pixel signal are generated by weighted averaging the pixel signals in the region, and a set of weighting coefficients having a high correlation between the comparison signal and the adjacent pixel signal is selected.
For the plurality of combinations, the pixel signal of the predicted adjacent area belonging to the combination is weighted and averaged using the set of weighting factors selected above, so that the comparison signal for the adjacent pixel signal is 2 One or more, and select a combination having a high correlation between the comparison signal and the adjacent pixel signal,
Based on the predicted adjacent region belonging to the selected combination, one or more candidate prediction signals of the target pixel signal are generated from the already-reproduced image, and the candidate prediction signal is previously selected with respect to the selected combination An image predictive coding apparatus, characterized in that a prediction signal is generated by weighted averaging using a set of weighting coefficients.   The prediction signal generating means sets at least the set of weighting factors so that a smaller weighting factor is set as the sum of absolute values of differences between pixel signals of a plurality of prediction adjacent regions belonging to the combination and the adjacent pixel signals increases. The image predictive coding apparatus according to claim 21 or 22, wherein one of the following is calculated.   The prediction signal generation means prepares a set of weighting factors determined in advance according to the number of prediction adjacent regions belonging to the combination, and derives at least one set of weighting factors from the set of weighting factors prepared. The image predictive coding apparatus according to claim 21 or 22,   The prediction signal generation means prepares a correspondence table in which a set of weighting coefficients is determined from the sum of absolute values of differences between pixel signals of a plurality of prediction adjacent regions belonging to the combination and the adjacent pixel signal, and uses the correspondence table The image predictive coding apparatus according to claim 21 or 22, wherein a set of at least one weighting factor is derived. Data decoding means for decoding encoded data of a residual signal related to a target region to be processed from among compressed data;
Residual signal restoration means for restoring a reproduction residual signal from a signal obtained by decoding by the data decoding means;
Prediction signal generating means for generating a prediction signal for the target pixel signal of the target region;
Reproduction image signal generation means for generating a reproduction image signal by adding the prediction signal generated by the prediction signal generation means and the reproduction residual signal restored by the residual signal restoration means, and
The predicted signal generation means obtains a plurality of predicted adjacent areas having the same shape as the target adjacent area adjacent to the target area from the search area composed of already reproduced images,
Deriving a combination of arbitrary predicted adjacent areas including two or more of the obtained plurality of predicted adjacent areas, and deriving two or more sets of weighting coefficients for weighted averaging pixel signals of the predicted adjacent areas belonging to the combination. And
For the combination, by generating a weighted average of the pixel signals of the prediction adjacent region belonging to the combination using the set of the plurality of weighting factors, two or more comparison signals for the adjacent pixel signal are generated,
Selecting a set of weighting coefficients having a high correlation between the comparison signal and the adjacent pixel signal;
Two or more candidate prediction signals of the target pixel signal are generated from the already reproduced image based on the prediction adjacent region belonging to the combination, and the candidate prediction signals are weighted and averaged using the selected set of weighting factors. An image predictive decoding apparatus, characterized in that a prediction signal is generated by: Data decoding means for decoding encoded data of a residual signal related to a target region to be processed from among compressed data;
Residual signal restoration means for restoring a reproduction residual signal from a signal obtained by decoding by the data decoding means;
Prediction signal generating means for generating a prediction signal for the target pixel signal of the target region;
Reproduction image signal generation means for generating a reproduction image signal by adding the prediction signal generated by the prediction signal generation means and the reproduction residual signal restored by the residual signal restoration means, and
The predicted signal generation means obtains a plurality of predicted adjacent areas having the same shape as the target adjacent area adjacent to the target area from the search area composed of already reproduced images,
Deriving two or more combinations of arbitrary predicted adjacent regions including at least one of the obtained plurality of predicted adjacent regions,
For a combination including two or more predicted adjacent areas, two or more sets of weighting factors for weighted averaging pixel signals of the predicted adjacent areas belonging to the combination are derived, and predicted adjacents are generated using the plurality of sets of weighting coefficients. Two or more comparison signals for the adjacent pixel signal are generated by weighted averaging the pixel signals in the region, and a set of weighting coefficients having a high correlation between the comparison signal and the adjacent pixel signal is selected.
For the plurality of combinations, the pixel signal of the predicted adjacent area belonging to the combination is weighted and averaged using the set of weighting factors selected above, so that the comparison signal for the adjacent pixel signal is 2 One or more, and select a combination having a high correlation between the comparison signal and the adjacent pixel signal,
Based on the predicted adjacent region belonging to the selected combination, one or more candidate prediction signals of the target pixel signal are generated from the already-reproduced image, and the candidate prediction signal is previously selected with respect to the selected combination An image predictive decoding apparatus that generates a prediction signal by performing weighted averaging using a set of weighting coefficients.   The prediction signal generating means sets at least the set of weighting factors so that a smaller weighting factor is set as the sum of absolute values of differences between pixel signals of a plurality of prediction adjacent regions belonging to the combination and the adjacent pixel signals increases. 28. The image predictive decoding apparatus according to claim 26, wherein one of the following is calculated.   The prediction signal generation means prepares a set of weighting factors determined in advance according to the number of prediction adjacent regions belonging to the combination, and derives at least one set of weighting factors from the set of weighting factors prepared. 28. The image predictive decoding apparatus according to claim 26 or 27.   The prediction signal generation means prepares a correspondence table in which a set of weighting coefficients is determined from the sum of absolute values of differences between pixel signals of a plurality of prediction adjacent regions belonging to the combination and the adjacent pixel signal, and uses the correspondence table 28. The image predictive decoding apparatus according to claim 26, wherein at least one set of weight coefficients is derived. A region dividing step for dividing the input image into a plurality of regions;
A prediction signal generating step for generating a prediction signal for a target pixel signal of a target region that is a processing target among the plurality of regions divided by the region dividing step;
A residual signal generation step for generating a residual signal between the prediction signal generated by the prediction signal generation step and the target pixel signal;
An encoding step for encoding the residual signal generated by the residual signal generation step,
The predicted signal generation step includes:
Search for a plurality of predicted adjacent areas having a high correlation with the target adjacent area consisting of the already reproduced adjacent pixel signal adjacent to the target area consisting of the target pixel signal from the search area consisting of the already reproduced image,
At least from the pixel signals of the N prediction regions based on the target region, the pixel signals of the prediction adjacent region among the N prediction adjacent regions searched for among the plurality of predicted adjacent regions searched, or the signals of both Using two signals, an evaluation value for evaluating the correlation between the N candidate prediction signals is calculated by a predetermined method,
An image predictive coding method, wherein when the evaluation value is smaller than a prescribed threshold value, a prediction signal is generated by processing the N candidate prediction signals using a predetermined synthesis method. A data decoding step for decoding encoded data of a residual signal related to a target region to be processed from the compressed data;
A residual signal restoration step for restoring a reproduction residual signal from the signal obtained by decoding in the data decoding step;
A prediction signal generation step of generating a prediction signal for the target pixel signal of the target region;
A reproduction image signal generation step of generating a reproduction image signal by adding the prediction signal generated in the prediction signal generation step and the reproduction residual signal restored in the residual signal restoration step, and
The predicted signal generation step searches a plurality of predicted adjacent areas having a high correlation with a target adjacent area composed of already reproduced adjacent pixel signals adjacent to the target area composed of the target pixel signal from a search area composed of a previously reproduced image. And
At least from the pixel signals of the N prediction regions based on the target region, the pixel signals of the prediction adjacent region among the N prediction adjacent regions searched for among the plurality of predicted adjacent regions searched, or the signals of both Using two signals, an evaluation value for evaluating the correlation between the N candidate prediction signals is calculated by a predetermined method,
An image predictive decoding characterized by generating a prediction signal by processing the N candidate prediction signals using a predetermined synthesis method when the evaluation value is smaller than a prescribed threshold value Method. A region dividing step for dividing the input image into a plurality of regions;
A prediction signal generating step for generating a prediction signal for a target pixel signal of a target region that is a processing target among the plurality of regions divided by the region dividing step;
A residual signal generation step for generating a residual signal between the prediction signal generated by the prediction signal generation step and the target pixel signal;
An encoding step for encoding the residual signal generated by the residual signal generation step,
In the prediction signal generation step, a plurality of predicted adjacent areas having the same shape as the target adjacent area adjacent to the target area are obtained from the search area including the already reproduced images, respectively.
A weighting factor for deriving a combination of arbitrary prediction adjacent areas including two or more of the obtained prediction adjacent areas from among the plurality of obtained prediction adjacent areas, and weighted averaging pixel signals of the prediction adjacent areas belonging to the combination Deriving two or more sets of
For the combination, by generating a weighted average of the pixel signals of the prediction adjacent region belonging to the combination using the set of the plurality of weighting factors, two or more comparison signals for the adjacent pixel signal are generated,
Selecting a set of weighting coefficients having a high correlation between the comparison signal and the adjacent pixel signal;
Two or more candidate prediction signals of the target pixel signal are generated from the already reproduced image based on the prediction adjacent region belonging to the combination, and the candidate prediction signals are weighted and averaged using the selected set of weighting factors. An image predictive coding method, characterized in that a prediction signal is generated by: A region dividing step for dividing the input image into a plurality of regions;
A prediction signal generating step for generating a prediction signal for a target pixel signal of a target region that is a processing target among the plurality of regions divided by the region dividing step;
A residual signal generation step for generating a residual signal between the prediction signal generated by the prediction signal generation step and the target pixel signal;
An encoding step for encoding the residual signal generated by the residual signal generation step,
In the prediction signal generation step, a plurality of predicted adjacent areas having the same shape as the target adjacent area adjacent to the target area are obtained from the search area including the already reproduced images, respectively.
Deriving two or more combinations of arbitrary predicted adjacent areas including at least one of the obtained plurality of predicted adjacent areas;
For a combination including two or more predicted adjacent areas, two or more sets of weighting factors for weighted averaging pixel signals of the predicted adjacent areas belonging to the combination are derived, and predicted adjacents are generated using the plurality of sets of weighting coefficients. Two or more comparison signals for the adjacent pixel signal are generated by weighted averaging the pixel signals in the region, and a set of weighting coefficients having a high correlation between the comparison signal and the adjacent pixel signal is selected.
For the plurality of combinations, the pixel signal of the predicted adjacent area belonging to the combination is weighted and averaged using the set of weighting factors selected above, so that the comparison signal for the adjacent pixel signal is 2 One or more, and select a combination having a high correlation between the comparison signal and the adjacent pixel signal,
Based on the predicted adjacent region belonging to the selected combination, one or more candidate prediction signals of the target pixel signal are generated from the already-reproduced image, and the candidate prediction signal is previously selected with respect to the selected combination An image predictive coding method, wherein a prediction signal is generated by weighted averaging using a set of weighting factors. A data decoding step for decoding encoded data of a residual signal related to a target region to be processed from the compressed data;
A residual signal restoration step for restoring a reproduction residual signal from the signal obtained by decoding in the data decoding step;
A prediction signal generation step of generating a prediction signal for the target pixel signal of the target region;
A reproduction image signal generation step of generating a reproduction image signal by adding the prediction signal generated in the prediction signal generation step and the reproduction residual signal restored in the residual signal restoration step, and
In the prediction signal generation step, a plurality of predicted adjacent areas having the same shape as the target adjacent area adjacent to the target area are obtained from the search area including the already reproduced images, respectively.
Deriving a combination of arbitrary predicted adjacent areas including two or more of the obtained plurality of predicted adjacent areas, and deriving two or more sets of weighting coefficients for weighted averaging pixel signals of the predicted adjacent areas belonging to the combination. And
For the combination, by generating a weighted average of the pixel signals of the prediction adjacent region belonging to the combination using the set of the plurality of weighting factors, two or more comparison signals for the adjacent pixel signal are generated,
Selecting a set of weighting coefficients having a high correlation between the comparison signal and the adjacent pixel signal;
Two or more candidate prediction signals of the target pixel signal are generated from the already reproduced image based on the prediction adjacent region belonging to the combination, and the candidate prediction signals are weighted and averaged using the selected set of weighting factors. An image predictive decoding method, characterized by generating a prediction signal by: A data decoding step for decoding encoded data of a residual signal related to a target region to be processed from the compressed data;
A residual signal restoration step for restoring a reproduction residual signal from the signal obtained by decoding in the data decoding step;
A prediction signal generation step of generating a prediction signal for the target pixel signal of the target region;
A reproduction image signal generation step of generating a reproduction image signal by adding the prediction signal generated in the prediction signal generation step and the reproduction residual signal restored in the residual signal restoration step, and
In the prediction signal generation step, a plurality of predicted adjacent areas having the same shape as the target adjacent area adjacent to the target area are obtained from the search area including the already reproduced images, respectively.
Deriving two or more combinations of arbitrary predicted adjacent areas including at least one of the obtained plurality of predicted adjacent areas;
For a combination including two or more predicted adjacent areas, two or more sets of weighting factors for weighted averaging pixel signals of the predicted adjacent areas belonging to the combination are derived, and predicted adjacents are generated using the plurality of sets of weighting coefficients. Two or more comparison signals for the adjacent pixel signal are generated by weighted averaging the pixel signals in the region, and a set of weighting coefficients having a high correlation between the comparison signal and the adjacent pixel signal is selected.
For the plurality of combinations, the pixel signal of the predicted adjacent area belonging to the combination is weighted and averaged using the set of weighting factors selected above, so that the comparison signal for the adjacent pixel signal is 2 One or more, and select a combination having a high correlation between the comparison signal and the adjacent pixel signal,
Based on the predicted adjacent region belonging to the selected combination, one or more candidate prediction signals of the target pixel signal are generated from the already-reproduced image, and the candidate prediction signal is previously selected with respect to the selected combination An image predictive decoding method, wherein a prediction signal is generated by performing weighted averaging using a set of weighting coefficients. An area division module for dividing an input image into a plurality of areas;
A prediction signal generation module that generates a prediction signal for a target pixel signal of a target region that is a processing target among the plurality of regions divided by the region division module;
A residual signal generation module that generates a residual signal between the prediction signal generated by the prediction signal generation module and the target pixel signal;
An encoding module that encodes the residual signal generated by the residual signal generation module;
The prediction signal generation module includes:
Search for a plurality of predicted adjacent areas having a high correlation with the target adjacent area consisting of the already reproduced adjacent pixel signal adjacent to the target area consisting of the target pixel signal from the search area consisting of the already reproduced image,
At least from the pixel signals of the N prediction regions based on the target region, the pixel signals of the prediction adjacent region among the N prediction adjacent regions searched for among the plurality of predicted adjacent regions searched, or the signals of both Using two signals, an evaluation value for evaluating the correlation between the N candidate prediction signals is calculated by a predetermined method,
An image prediction code characterized by generating a prediction signal by processing the N candidate prediction signals using a predetermined synthesis method when the evaluation value is smaller than a prescribed threshold value Program. A data decoding module that decodes encoded data of a residual signal related to a target region to be processed from the compressed data;
A residual signal restoration module for restoring a reproduction residual signal from a signal obtained by decoding by the data decoding module;
A prediction signal generation module that generates a prediction signal for a target pixel signal in the target region;
A reproduction image signal generation module that generates a reproduction image signal by adding the prediction signal generated by the prediction signal generation module and the reproduction residual signal restored by the residual signal restoration module;
The predicted signal generation module searches a plurality of predicted adjacent areas having a high correlation with a target adjacent area composed of already reproduced adjacent pixel signals adjacent to the target area composed of the target pixel signal from a search area composed of a previously reproduced image. And
At least from the pixel signals of the N prediction regions based on the target region, the pixel signals of the prediction adjacent region among the N prediction adjacent regions searched for among the plurality of predicted adjacent regions searched, or the signals of both Using two signals, an evaluation value for evaluating the correlation between the N candidate prediction signals is calculated by a predetermined method,
An image predictive decoding program which generates a prediction signal by processing the N candidate prediction signals using a predetermined synthesis method when the evaluation value is smaller than a predetermined threshold. An area division module for dividing an input image into a plurality of areas;
A prediction signal generation module that generates a prediction signal for a target pixel signal of a target region that is a processing target among the plurality of regions divided by the region division module;
A residual signal generation module that generates a residual signal between the prediction signal generated by the prediction signal generation module and the target pixel signal;
An encoding module that encodes the residual signal generated by the residual signal generation module;
The predicted signal generation module obtains a plurality of predicted adjacent areas having the same shape as the target adjacent area adjacent to the target area from each of the search areas including the already reproduced images,
A weighting factor for deriving a combination of arbitrary prediction adjacent areas including two or more of the obtained prediction adjacent areas from among the plurality of obtained prediction adjacent areas, and weighted averaging pixel signals of the prediction adjacent areas belonging to the combination Deriving two or more sets of
For the combination, by generating a weighted average of the pixel signals of the prediction adjacent region belonging to the combination using the set of the plurality of weighting factors, two or more comparison signals for the adjacent pixel signal are generated,
Selecting a set of weighting coefficients having a high correlation between the comparison signal and the adjacent pixel signal;
Two or more candidate prediction signals of the target pixel signal are generated from the already reproduced image based on the prediction adjacent region belonging to the combination, and the candidate prediction signals are weighted and averaged using the selected set of weighting factors. An image predictive coding program characterized by generating a prediction signal according to the above. An area division module for dividing an input image into a plurality of areas;
A prediction signal generation module that generates a prediction signal for a target pixel signal of a target region that is a processing target among the plurality of regions divided by the region division module;
A residual signal generation module that generates a residual signal between the prediction signal generated by the prediction signal generation module and the target pixel signal;
An encoding module that encodes the residual signal generated by the residual signal generation module;
The predicted signal generation module obtains a plurality of predicted adjacent areas having the same shape as the target adjacent area adjacent to the target area from each of the search areas including the already reproduced images,
Deriving two or more combinations of arbitrary predicted adjacent areas including at least one of the obtained plurality of predicted adjacent areas;
For a combination including two or more predicted adjacent areas, two or more sets of weighting factors for weighted averaging pixel signals of the predicted adjacent areas belonging to the combination are derived, and predicted adjacents are generated using the plurality of sets of weighting coefficients. Two or more comparison signals for the adjacent pixel signal are generated by weighted averaging the pixel signals in the region, and a set of weighting coefficients having a high correlation between the comparison signal and the adjacent pixel signal is selected.
For the plurality of combinations, the pixel signal of the predicted adjacent area belonging to the combination is weighted and averaged using the set of weighting factors selected above, so that the comparison signal for the adjacent pixel signal is 2 One or more, and select a combination having a high correlation between the comparison signal and the adjacent pixel signal,
Based on the predicted adjacent region belonging to the selected combination, one or more candidate prediction signals of the target pixel signal are generated from the already-reproduced image, and the candidate prediction signal is previously selected with respect to the selected combination An image predictive coding program for generating a prediction signal by performing weighted averaging using a set of weighting coefficients. A data decoding module that decodes encoded data of a residual signal related to a target region to be processed from the compressed data;
A residual signal restoration module for restoring a reproduction residual signal from a signal obtained by decoding by the data decoding module;
A prediction signal generation module that generates a prediction signal for a target pixel signal in the target region;
A reproduction image signal generation module that generates a reproduction image signal by adding the prediction signal generated by the prediction signal generation module and the reproduction residual signal restored by the residual signal restoration module;
The predicted signal generation module obtains a plurality of predicted adjacent areas having the same shape as the target adjacent area adjacent to the target area from each of the search areas including the already reproduced images,
Deriving a combination of arbitrary predicted adjacent areas including two or more of the obtained plurality of predicted adjacent areas, and deriving two or more sets of weighting coefficients for weighted averaging pixel signals of the predicted adjacent areas belonging to the combination. And
For the combination, by generating a weighted average of the pixel signals of the prediction adjacent region belonging to the combination using the set of the plurality of weighting factors, two or more comparison signals for the adjacent pixel signal are generated,
Selecting a set of weighting coefficients having a high correlation between the comparison signal and the adjacent pixel signal;
Two or more candidate prediction signals of the target pixel signal are generated from the already reproduced image based on the prediction adjacent region belonging to the combination, and the candidate prediction signals are weighted and averaged using the selected set of weighting factors. An image predictive decoding program characterized by generating a prediction signal by means of A data decoding module that decodes encoded data of a residual signal related to a target region to be processed from the compressed data;
A residual signal restoration module for restoring a reproduction residual signal from a signal obtained by decoding by the data decoding module;
A prediction signal generation module that generates a prediction signal for a target pixel signal in the target region;
A reproduction image signal generation module that generates a reproduction image signal by adding the prediction signal generated by the prediction signal generation module and the reproduction residual signal restored by the residual signal restoration module;
The predicted signal generation module obtains a plurality of predicted adjacent areas having the same shape as the target adjacent area adjacent to the target area from each of the search areas including the already reproduced images,
Deriving two or more combinations of arbitrary predicted adjacent areas including at least one of the obtained plurality of predicted adjacent areas;
For a combination including two or more predicted adjacent areas, two or more sets of weighting factors for weighted averaging pixel signals of the predicted adjacent areas belonging to the combination are derived, and predicted adjacents are generated using the plurality of sets of weighting coefficients. Two or more comparison signals for the adjacent pixel signal are generated by weighted averaging the pixel signals in the region, and a set of weighting coefficients having a high correlation between the comparison signal and the adjacent pixel signal is selected.
For the plurality of combinations, the pixel signal of the predicted adjacent area belonging to the combination is weighted and averaged using the set of weighting factors selected above, so that the comparison signal for the adjacent pixel signal is 2 One or more, and select a combination having a high correlation between the comparison signal and the adjacent pixel signal,
Based on the predicted adjacent region belonging to the selected combination, one or more candidate prediction signals of the target pixel signal are generated from the already-reproduced image, and the candidate prediction signal is previously selected with respect to the selected combination An image predictive decoding program for generating a prediction signal by performing weighted averaging using a set of weighting coefficients.

Images