JP3851317B2 - Image predictive coding method and apparatus - Google Patents

Description

  The present invention relates to an image predictive coding apparatus and method, an image predictive decoding apparatus and method, and a recording medium. In particular, an image predictive encoding apparatus and method, and an image predictive decoding apparatus and method for storing digital image data of an image that is a still image or a moving image in a recording medium such as an optical disc or transmitting a communication line About. The present invention also relates to a recording medium on which a program including steps of the image predictive encoding method is recorded, and a recording medium on which a program including steps of the image predictive decoding method is recorded.

  In order to efficiently store or transmit digital images, it is necessary to perform compression encoding. In addition to discrete cosine transform (hereinafter referred to as DCT transform) represented by JPEG (Joint Photographic Experts Group) and MPEG (Motion Picture Experts Group) as methods for compressing and coding digital images, subband coding is also possible. And waveform encoding methods such as wavelet encoding and fractal encoding. Also, in order to remove redundant signals between images, inter-image prediction using motion compensation is performed, and the difference signal is waveform-encoded.

  In the MPEG system, an input image is divided into a plurality of 16 × 16 macroblocks for processing. One macroblock is further divided into 8 × 8 blocks, subjected to 8 × 8 DCT transform processing, and then quantized. This is called intraframe coding.

  On the other hand, a motion detection method such as block matching detects the predicted macroblock with the smallest error in the target macroblock from other frames adjacent in time, and subtracts the detected predicted macroblock from the target macroblock. Then, a differential macroblock is generated, subjected to 8 × 8 DCT transform, and then quantized. This is called interframe coding, and the prediction macroblock is called a time domain prediction signal. Thus, MPEG does not predict an image from the same frame.

  A normal image has many spatially similar regions, and this property can be used to approximate an image to a spatial region. Similar to the time domain prediction signal, it is also possible to obtain the prediction signal from the same frame. This is called a spatial domain prediction signal.

  Since two pixel values that are spatially close to each other are close, the prediction signal in the spatial region is generally located near the target signal. On the other hand, since there is no original image on the reception side or reproduction side, it is necessary to use a signal that has been encoded and reproduced in the past as the prediction signal. From these two elements, it is necessary to generate a prediction signal in the spatial domain at high speed. This is because the pixel value is used for generating a prediction signal immediately after being decoded and reproduced.

  Therefore, it is necessary to generate a prediction signal in the spatial domain easily and with high accuracy. In addition, a configuration capable of high-speed computation is required in the encoding device and the decoding device.

  Incidentally, encoding of image data has been widely used in many international standards such as JPEG, MPEG1, H.261, MPEG2 and H.263. Each of the latter standards further improves coding efficiency. That is, efforts have been made to further reduce the number of bits compared to conventional standards to represent the same image quality.

The encoding of image data for a moving image consists of intra frame encoding and prediction frame encoding. Here, intra-frame coding refers to intra-frame coding within a screen of one frame. In a typical hybrid coding system such as the MPEG1 standard, consecutive frames can be classified into three different types:
(A) Intra frame (hereinafter referred to as I frame),
(B) Prediction frame (hereinafter referred to as P frame) and (c) Bidirectional prediction frame (hereinafter referred to as B frame).

  I frames are encoded independently of other frames, ie, I frames are compressed without using other frames. A P frame is encoded through motion detection and compensation by using the previous frame to predict the content of the encoded frame (which is a P frame). B frames are encoded by using motion detection and compensation using information from the previous frame and information from subsequent frames that predict the content data of the B frame. The previous frame and the subsequent frame are either I frames or P frames. The I frame belongs to the intra code mode. The P frame and the B frame belong to the prediction code mode.

  The compression methods are different as the encoding characteristics of the I frame, the P frame, and the B frame are different. I-frames require a larger number of bits compared to P-frames and B-frames because they do not use temporal prediction to reduce redundancy.

  Here, MPEG2 will be described as an example. Assume that the bit rate is 4 Mbit / sec and the image is an image of 30 frames / sec. In general, the ratio of the number of bits used for I, P and B frames is 6: 3: 1. Thus, an I frame uses about 420 Kbits / s and a B frame uses about 70 Kbits / s. This is because the B frame is sufficiently predicted from both directions.

  FIG. 14 is a block diagram showing a configuration of a conventional image predictive coding apparatus. Since the DCT transformation is performed on a block basis, all modern image coding methods are based on dividing the image into smaller blocks. In intra-frame coding, block sampling processing 1001 is first performed on an input digital image signal as shown in FIG. Next, after these blocks after the block sampling process 1001 are performed, a DCT transform process 1004 is performed, and then a quantization process 1005 and a run length Hoffman variable length coding (VLC: Entropy coding) process 1006. Is executed. On the other hand, in predictive frame coding, a motion compensation process 1003 is performed on an input digital image, and a DCT transform process 1004 is performed on a motion compensated block (ie, a predicted block). The Next, a quantization process 1005 and a run length Hoffman VLC encoding (entropy encoding) process 1006 are executed.

  The block-based DCT transform process 1004 removes or reduces the spatial redundancy in the block being processed, and the motion detection and compensation processes 1002, 1003 are temporary redundancy between adjacent frames. It is known from conventional image coding techniques to remove or reduce. In addition, a run length Hoffman VLC encoding or other entropy encoding process 1006, which is performed after the DCT transform process 1004 and the quantization process 1005, removes the statistical redundancy between the quantized DCT transform coefficients. . However, that process is only for blocks in the screen.

  Digital images inherently have a large spatial redundancy. This redundancy exists not only between the blocks in the frame of the image but also between the blocks beyond the blocks. However, it is clear from the above that the actual method does not use a method for removing redundancy between image blocks.

  In current image coding techniques, DCT transform processing 1004 or other transform processing is performed on a block basis for hardware formation and computational constraints.

  Spatial redundancy is reduced by a block-based transformation process, but only within one block. Redundancy between two adjacent blocks is not well considered, but can be further reduced by using intraframe coding, which always consumes a large number of bits.

  In addition, block-based DCT transform processing removes or reduces spatial redundancy in the block being processed, and motion prediction and compensation processing is performed temporarily between two adjacent frames. It is known from current image coding techniques to remove or reduce redundancy. A zigzag scan and run-length Hoffman VLC encoding process or other entropy encoding process, which is performed after the DCT transform process and the quantization process, removes statistical redundancy in the quantized DCT transform coefficients. Note that it is limited to one block.

  Digital images inherently contain high spatial redundancy. This redundancy exists not only inside the blocks, but also across the blocks of the image and between the blocks. Therefore, as is clear from the above, the existing method does not use any method for removing redundancy between blocks of one image except for prediction of DC coefficients of JPEG, MPEG1 and MPEG2.

  In MPEG1 and MPEG2, DC coefficient prediction is performed by subtracting the DC value of the previous encoded block from the current encoded block. This is a simple prediction method that does not have adaptability or mode switching when the prediction is not appropriate. Furthermore it only includes DC coefficients.

  In the current state of the art, zigzag scanning is used for all blocks prior to run-length encoding. No attempt has been made to adaptively scan based on block content data.

  FIG. 22 is a block diagram showing a configuration of a conventional image predictive coding apparatus. In FIG. 22, the conventional image predictive encoding apparatus includes a block sampling unit 2001, a DCT transform unit 2003, a quantization unit 2004, a zigzag scan unit 2005, and an entropy encoding unit 2006. In this specification, the term “unit” means a circuit device.

  In intra-frame coding (that is, intra-frame coding), after a block sampling process 2001 is performed on an input image signal, a DCT conversion process 2003 is directly performed, and a quantization process is performed. 2004, a zigzag scan process 2005 and an entropy encoding process 2006 are sequentially executed. On the other hand, in inter-frame coding (ie, inter-frame coding, ie, predictive frame coding), after block sampling processing 2001, motion detection and compensation processing is performed in unit 2011, and then from block sampling 2001, A prediction error is obtained by the adder 2002 by subtracting the detection value from the unit 2011 from the image data. Further, a DCT transform process 2003 is performed on the prediction error, and then a quantization process 2004, a zigzag scan process 2005, and an entropy coding process 2006 are performed similarly to the intra frame coding.

  In the local decoder provided in the image predictive coding apparatus in FIG. 22, the inverse quantization process and the inverse DCT transform process are executed in units 2007 and 2008. In intra-frame coding, the motion detection and compensated prediction value is added by the adder 2009 to the prediction error reconstructed by the units 2007 and 2008, and the added value is the locally decoded image data. The decoded image data is stored in the frame memory 2010 of the local decoder. Finally, the bit stream is output from the entropy encoding unit 2010 and transmitted to the counterpart image predictive decoding apparatus.

  FIG. 23 is a block diagram showing a configuration of a conventional image predictive decoding apparatus. The bit stream is decoded by a variable length decoder (VLD: Variable Length Decoding) unit (or entropy decoding unit) 2021, and then the decoded image data is subjected to an inverse quantization process and an inverse DCT transform process. It is executed in units 2023 and 2024. In inter-frame coding, the motion detection and compensated prediction value formed in unit 2027 is added to the prediction error reconstructed by the adder 2025 to form local decoded image data. The locally decoded image data is stored in the frame memory 1026 of the local decoder.

Japanese Patent Laid-Open No. 4-306095.

  In current image coding techniques, DCT transform processing or other transform processing is performed on a block basis due to hardware formation and computational constraints. Spatial redundancy will be reduced by block-based transformations. However, it is only within the block. Redundancy between adjacent blocks is not considered enough. In particular, no special consideration is given to intra-frame coding that always consumes a large amount of bits.

  A first object of the present invention is to provide an image predictive coding apparatus and method, and an image predictive decoding apparatus and method capable of generating predicted image data in a spatial domain simply, at high speed and with high accuracy. There is.

  In addition, the second object of the present invention is to eliminate the redundancy in the block and to encode the image data more efficiently compared to the conventional image predictive coding apparatus and image predictive decoding apparatus. It is an object to provide an image predictive coding apparatus and method, and an image predictive decoding apparatus and method that can be encoded or decoded.

  Furthermore, a third object of the present invention is to solve the problem that important transform coefficients are concentrated in different areas of the block depending on the internal properties of the image data, and to provide a correct scanning method for the block. An object of the present invention is to provide an image predictive encoding apparatus and method, and an image predictive decoding apparatus and method that can improve the efficiency of entropy encoding processing by determining.

  A fourth object of the present invention is to provide a recording medium on which the steps of the image predictive encoding method or the image predictive decoding method are recorded.

An image predictive coding method according to the present invention includes sampling an image signal into a plurality of blocks,
Converting the image signal of the block into a two-dimensional sequence of DCT coefficients having DC and AC coefficients;
Predicting the DC coefficient of the current block (C) from the DC coefficient of the adjacent block adaptively selected from either the left block (B) or the upper block (A).

  The image predictive coding method preferably further includes quantizing the DCT coefficient before predicting the DC coefficient.

  Further, in the image predictive coding method, preferably, the prediction of the DC coefficient of the current block (C) is repeated independently over all the blocks of the macroblock.

An image predictive coding method according to the present invention includes sampling an image signal into a plurality of blocks,
Converting the image signal of the block into a two-dimensional sequence of DCT coefficients having DC and AC coefficients;
Predicting the DC and AC coefficients of the current block (C) from the DCT coefficients of adjacent blocks adaptively selected from either the left block (B) or the upper block (A). Features.

The image predictive encoding method according to the present invention includes sampling an image signal into a plurality of blocks,
Converting the image signal of the block into a two-dimensional sequence of DCT coefficients having DC and AC coefficients;
Predicting the DC coefficient of the current block (C) from the DC coefficients of adjacent blocks adaptively selected from either the left block (B) or the upper block (A);
The AC coefficient of the current block (C) is
(A) When the DC coefficient of the current block is selected from the left block (B), the first column of the left block (B);
(B) when the DC coefficient of the current block is selected from the upper block (A), predicting from any set of coefficients with the first row of the upper block (A). And

  In the image predictive coding method, preferably, the DC coefficient of the current block (C) is calculated from the DC coefficient of the adjacent block selected from either the left block (B) or the upper block (A). Is predicted according to the least bit usage rule.

  Furthermore, the image predictive coding method preferably further includes supplying an instruction bit indicating whether or not the AC coefficient of the current block is predicted from the adjacent block.

An image predictive coding method according to the present invention includes sampling an image signal into a plurality of blocks,
Converting the image signal of the block into a two-dimensional sequence of DCT coefficients having DC and AC coefficients;
Quantizing the DCT coefficients into quantized DCT values;
Predicting the quantized DC coefficients of the current block from the DC coefficients of neighboring blocks adaptively selected from either the left block or the upper block;
The DC coefficient of the block selected for use in the prediction of the quantized DC coefficient of the current block is measured by the ratio of the quantization step size of the current block to the quantization step size of the selected block. It is characterized by being.

An image predictive coding method according to the present invention includes sampling an image signal into a plurality of blocks,
Converting the image signal of the block into a two-dimensional sequence of DCT coefficients having DC and AC coefficients;
Quantizing the DCT coefficients into quantized DCT values;
Predicting the quantized DC coefficients and quantized AC coefficients of the current block from the DCT coefficients of adjacent blocks adaptively selected from either the left block or the upper block;
The DCT coefficient of the block selected to be used for the prediction of the quantized DC coefficient and the quantized AC coefficient of the current block (C) is the quantization step size of the selected block. Measured by the ratio of the quantization step size of the current block to.

An image predictive coding method according to the present invention converts a block of a sampled image signal into a two-dimensional sequence of DCT coefficients;
Obtaining a predicted DCT transform coefficient by predicting a DCT coefficient of a current block from a DCT coefficient of an adjacent block adaptively selected from either the left block or the upper block;
Obtaining a two-dimensional sequence of a plurality of differential DCT coefficients by subtracting the predicted DCT transform coefficients from the DCT coefficients of the current block;
Scanning a two-dimensional sequence of the plurality of differential DCT coefficients and converting to a one-dimensional sequence of DCT coefficients;
The above scan is
(1) In the prediction, zigzag scanning when the AC coefficient of the current block is not predicted;
(2) vertical scanning when the AC coefficient of the current block is predicted and the DC prediction refers to the left block;
(3) The AC coefficient of the current block is predicted, and the DC prediction is executed by one of horizontal scanning when referring to the upper block.

  The image predictive coding method preferably further includes performing variable length coding of the one-dimensional sequence of the DCT coefficients.

  Further, in the image predictive encoding method, preferably, the conversion is characterized in that a block of a sampled image signal is converted into a plurality of 8 × 8 blocks of DCT coefficients.

Still further, in the image predictive encoding method, preferably, the scanning is performed as follows.
(1) When the AC coefficient of the current block (C) is not predicted in the prediction, zigzag scanning of 8 × 8 blocks of a plurality of DCT transform coefficients in the following order:
0, 1, 5, 6, 14, 15, 27, 28;
2, 4, 7, 13, 16, 26, 29, 42;
3, 8, 12, 17, 25, 30, 41, 43;
9, 11, 18, 24, 31, 40, 44, 53;
10, 19, 23, 32, 39, 45, 52, 54;
20, 22, 33, 38, 46, 51, 55, 60;
21, 34, 37, 47, 50, 56, 59, 61;
35, 36, 48, 49, 57, 58, 62, 63;
(2) When the AC coefficient of the current block (C) is predicted in the prediction, and the DC prediction refers to the left block, vertical scanning is performed in the following order:
0, 4, 6, 20, 22, 36, 38, 52;
1, 5, 7, 21, 23, 37, 39, 53;
2, 8, 19, 24, 34, 40, 50, 54;
3, 9, 18, 25, 35, 41, 51, 55;
10, 17, 26, 30, 42, 46, 56, 60;
11, 16, 27, 31, 43, 47, 57, 61;
12, 15, 28, 32, 44, 48, 58, 62;
13, 14, 29, 33, 45, 49, 59, 63;
(3) When the AC coefficient of the current block (C) is predicted in the prediction and the DC prediction refers to the upper block, horizontal scanning is performed in the following order:
0, 1, 2, 3, 10, 11, 12, 13;
4, 5, 8, 9, 17, 16, 15, 14;
6, 7, 19, 18, 26, 27, 28, 29;
20, 21, 24, 25, 30, 31, 32, 33;
22, 23, 34, 35, 42, 43, 44, 45;
36, 37, 40, 41, 46, 47, 48, 49;
38, 39, 50, 51, 56, 57, 58, 59;
52, 53, 54, 55, 60, 61, 62, 63;
It is characterized in that the scanning is executed.

  In the image predictive coding method, preferably, the DC coefficient of the current block (C) is calculated from the DC coefficient of the adjacent block selected from either the left block (B) or the upper block (A). Predicting according to the least bit usage rule.

  Furthermore, the image predictive coding method preferably further includes supplying an instruction bit indicating whether or not the AC coefficient of the current block is predicted from the adjacent block.

An image predictive encoding device according to a first aspect of the present invention includes a dividing unit that divides input encoded image data into image data of a plurality of small areas adjacent to each other;
When the image data of the small area to be processed is encoded among the image data of the plurality of small areas adjacent to each other divided by the dividing unit, the image data reproduced adjacent to the image data of the small area to be processed is encoded. The image data of the reproduction subregion is the image data of the intra prediction small region of the processing target small region, the image data of the intra prediction small region is the image data of the optimal prediction subregion, and First generation means for generating difference subregion image data that is a difference between the image data and the optimal prediction subregion image data;
Encoding means for encoding the image data of the small difference area generated by the generating means;
Decoding means for decoding the image data of the small difference area encoded by the encoding means;
Second generation means for generating reproduced subregion image data reproduced by adding the difference subregion image data decoded by the decoding unit to the optimum predicted subregion image data.

Further, the image predictive coding apparatus according to the second invention includes a dividing unit that divides input coded image data into image data of a plurality of small regions adjacent to each other;
Image data of a reproduced reproduction small area adjacent to the image data of the small area to be processed when encoding the small area to be processed among a plurality of small areas adjacent to each other divided by the dividing means Only the significant image data indicated by the input significant signal indicating whether or not the encoded image data is significant is used as the image data of the predicted small region in the screen of the small region to be processed, and The image data of the predicted small area in the screen is used as the image data of the optimum predicted small area, and image data of the difference small area that is the difference between the image data of the small area to be processed and the image data of the optimum predicted small area is generated. First generation means;
Encoding means for encoding the image data of the small difference area generated by the first generation means;
Decoding means for decoding the image data of the small difference area encoded by the encoding means;
Second generation means for generating reproduced subregion image data reproduced by adding the difference subregion image data decoded by the decoding unit to the optimum predicted subregion image data.

Further, the image predictive decoding apparatus according to the third invention comprises an analyzing means for analyzing the input encoded image data sequence and outputting an image difference signal;
Decoding means for decoding image data of a reproduction difference small area from the difference image signal output from the analysis means;
A line memory for storing image data for generating image data of a predetermined intra prediction small area;
By executing the prediction signal generation process on the image data from the line memory, the reproduced image data adjacent to the image data of the reproduction difference small region is set as the image data of the prediction small region in the screen, and Generating means for outputting the image data of the predicted small area as the image data of the optimum predicted small area;
Adds the image data of the reproduction difference small area from the decoding means and the image data of the optimal prediction small area from the generation means, and outputs the image data for generating the intra prediction small area of the addition result And adding means for storing in the line memory.

Still further, the image predictive decoding device according to the fourth invention comprises an analysis means for analyzing the input encoded image data sequence and outputting an image difference signal, a motion vector signal, and a control signal. ,
Decoding means for decoding the difference image signal output from the analysis means into image data of a reproduction difference small area;
Control means for outputting a switching signal for controlling the motion compensation means and the generating means to selectively operate based on the control signal output from the analysis means,
A frame memory for storing predetermined reproduction image data;
A line memory for storing image data for generating image data of a predetermined intra prediction small area;
In response to the switching signal from the control means, by performing motion compensation processing on the input motion vector signal, image data of the temporal prediction small region is generated from the frame memory, and the optimal prediction small region Motion compensation means for outputting as image data,
In response to a switching signal from the control means, a predicted signal generation process is performed on the image data from the line memory, thereby reproducing reproduced image data adjacent to the image data of the reproduction difference small area on the screen. Generating means for outputting image data of the intra prediction small area, and outputting the image data of the intra prediction small area as image data of the optimal prediction small area;
By adding the image data of the reproduction difference small area from the decoding means and the optimum prediction small area from the generating means, the reproduction image data as the addition result is output, and the reproduction image data is stored in the frame memory. And adding means for storing only the image data for generating the image data of the intra prediction small area in the line memory.

An image predictive decoding device according to a fifth aspect of the invention is an analysis unit that analyzes an input encoded image data sequence and outputs a compressed shape signal and an image difference signal;
First decoding means for decoding the compressed shape signal output from the analyzing means into a reproduced shape signal;
Second decoding means for decoding the difference image signal output from the analysis means into image data of a reproduction difference small area;
A line memory for storing image data for generating image data of a predetermined intra prediction small area;
Significant image data indicated by the reproduction shape signal from the reproduced image data adjacent to the image data of the reproduction difference small area by executing prediction signal processing on the image data from the line memory. Generating means for outputting only the image data of the predicted small area in the screen, and outputting the image data of the predicted small area in the screen as the image data of the optimal predicted small area;
By adding the image data of the reproduction difference small area from the second decoding means and the optimum prediction small area from the generating means, the addition result image data is output, and the intra-screen prediction small area Adding means for storing only the image data for generating the image data in the line memory.

Further, the image predictive decoding apparatus according to the sixth invention analyzes the input encoded image data sequence and outputs a compressed shape signal, an image difference signal, a motion vector signal, and a control signal. Analysis means to
First decoding means for decoding the compressed shape signal output from the analyzing means into a reproduced shape signal;
Second decoding means for decoding the difference image signal output from the analysis means into a reproduction difference small area;
Control means for outputting a switching signal for controlling the motion compensation means and the generation means to selectively operate based on the control signal output from the analysis means,
A frame memory for storing predetermined reproduction image data;
A line memory for storing image data for generating image data of a predetermined intra prediction small area;
In response to the switching signal output from the control means, a time prediction is performed by executing a motion compensation process on the reproduced image data from the frame memory based on the motion vector signal output from the analysis means. Motion compensation means for generating image data of a small area and outputting it as image data of an optimally predicted small area;
In response to the switching signal output from the control means, predictive signal processing is performed on the image data from the line memory, so that the reproduced image data adjacent to the image data in the reproduction difference small area Generating means for outputting only significant image data indicated by the reproduction shape signal as image data of an intra prediction small area, and outputting image data of the intra prediction small area as image data of an optimal prediction sub area,
By adding the image data of the reproduction difference small area from the second decoding means and the optimal prediction small area from the generating means, the reproduction image data as the addition result is output, and the reproduction image data is Adding means for storing in the line memory only the image data stored in the frame memory and for generating the intra prediction small area.

An image predictive encoding device according to a seventh aspect of the present invention is a sampling means for sampling an input image signal into image data of a plurality of blocks each including a two-dimensional array of pixel values;
Conversion means for converting the image data of the block sampled by the sampling means into coefficient data of a predetermined conversion area;
A block memory for storing the coefficient data of the restored block;
Prediction means for forming coefficient data of a plurality of prediction blocks based on the coefficient data of the block converted by the conversion means, based on the coefficient data of the block reconstructed before stored in the block memory;
Among the coefficient data of the plurality of prediction blocks formed by the prediction means, the coefficient data of the prediction block with the highest efficiency is determined, selected and output, and the indicator representing the selected prediction block is in the format of the instruction bit Determining means for transmitting to the image predictive decoding device,
First addition means for outputting coefficient data of prediction error of the subtraction result by subtracting coefficient data of the prediction block selected by the determination means from coefficient data of the current block at the present time;
Quantization means for quantizing prediction error coefficient data output from the first addition means;
Encoding means for entropy encoding the prediction error coefficient data from the quantization means, and transmitting the encoded prediction error coefficient data to the image predictive decoding device;
Dequantizing the coefficient data of the prediction error from the quantizing means, and outputting the restored block coefficient data; and
By adding the coefficient data of the prediction block output from the determination unit to the coefficient data of the prediction error output from the inverse quantization unit, the coefficient data of the restored block is output and also to the block memory A second adding means for storing;
And inverse transform means for generating image data of the restored block by inversely transforming the coefficient data of the block output from the second adding means.

According to an eighth aspect of the present invention, there is provided an image predictive encoding device, wherein sampling means for sampling an input image signal into a plurality of blocks of image data including pixel values of a two-dimensional array;
Conversion means for converting image data of a plurality of blocks sampled by the sampling means into coefficient data of a predetermined conversion area;
Quantization means for quantizing coefficient data in the transform region from the transform means;
A block memory for storing the coefficient data of the restored block;
Prediction means for forming coefficient data of a plurality of prediction blocks based on the coefficient data of the block converted by the conversion means, based on the coefficient data of the block reconstructed before stored in the block memory;
Among the coefficient data of the plurality of prediction blocks formed by the prediction means, the coefficient data of the prediction block with the highest efficiency is determined, selected and output, and the indicator representing the selected prediction block is in the format of the instruction bit Determining means for transmitting to the image predictive decoding device,
First addition means for outputting coefficient data of prediction error of the subtraction result by subtracting coefficient data of the prediction block selected by the determination means from coefficient data of the current block at the present time;
Encoding means for entropy encoding the prediction error coefficient data from the first addition means, and transmitting the encoded prediction error coefficient data to the image predictive decoding apparatus;
By adding the coefficient data of the prediction block output from the determination means to the coefficient data of the prediction error from the first addition means, the coefficient data of the quantized current block is restored and output, Second addition means for storing in the block memory;
Dequantizing means for dequantizing and outputting the coefficient data of the current block output from the second adding means;
And inverse transform means for generating image data of the restored block by inverse transforming the coefficient data of the current block from the inverse quantization means.

Furthermore, the image predictive coding apparatus according to the ninth invention comprises sampling means for sampling the input image signal into image data of a plurality of blocks each including a two-dimensional array of pixel values;
Compensation means for generating and outputting image data of a prediction error of a block subjected to motion compensation by executing motion compensation processing on the image data of the input block;
First addition means for subtracting the image data of the prediction error of the block output from the compensation means from the image data of the block output from the sampling means, and outputting the image data of the block as a subtraction result;
Conversion means for converting the image data of the block output from the first addition means into coefficient data of a predetermined conversion area;
A block memory for storing the coefficient data of the restored block;
Prediction means for forming coefficient data of a plurality of prediction blocks based on the coefficient data of the block converted by the conversion means, based on the coefficient data of the block reconstructed before stored in the block memory;
Among the coefficient data of the plurality of prediction blocks formed by the prediction means, the coefficient data of the prediction block with the highest efficiency is determined, selected and output, and the indicator representing the selected prediction block is in the format of the instruction bit Determining means for transmitting to the image predictive decoding device,
Subtracting the coefficient data of the prediction block selected by the determination means from the coefficient data of the current block at the current time, thereby outputting the coefficient data of the prediction error of the subtraction result;
Quantization means for quantizing prediction error coefficient data output from the second addition means;
Encoding means for entropy encoding the prediction error coefficient data from the quantization means, and transmitting the encoded prediction error coefficient data to the image predictive decoding device;
Dequantizing the coefficient data of the prediction error from the quantizing means, and outputting the restored block coefficient data; and
By adding the coefficient data of the prediction block output from the determination unit to the coefficient data of the prediction error output from the inverse quantization unit, the coefficient data of the restored block is output, and the block memory is also output. A third adding means for storing;
Inverse transform means for generating image data of the restored block by inversely transforming the coefficient data of the block output from the third adding means;
The image data of the restored block output from the compensation means is added to the image data of the restored block from the inverse transform means, so that the image data of the restored block is added to the compensation means. 4th addition means to output to.

Still further, the image predictive coding apparatus according to the tenth invention comprises a sampling means for sampling an input image signal into a plurality of blocks of image data including a two-dimensional array of pixel values;
Compensation means for generating and outputting image data of a prediction error of a block subjected to motion compensation by executing motion compensation processing on the image data of the input block;
First addition means for subtracting the image data of the prediction error of the block output from the compensation means from the image data of the block output from the sampling means, and outputting the image data of the block as a subtraction result;
Conversion means for converting the image data of the block output from the first addition means into coefficient data of a predetermined conversion area;
Quantization means for quantizing coefficient data in the transform region from the transform means;
A block memory for storing the coefficient data of the restored block;
Prediction means for forming coefficient data of a plurality of prediction blocks based on the coefficient data of the block converted by the conversion means, based on the coefficient data of the block reconstructed before stored in the block memory;
Among the coefficient data of the plurality of prediction blocks formed by the prediction means, the coefficient data of the prediction block with the highest efficiency is determined, selected and output, and the indicator representing the selected prediction block is in the format of the instruction bit Determining means for transmitting to the image predictive decoding device,
Subtracting the coefficient data of the prediction block selected by the determination means from the coefficient data of the current block at the current time, thereby outputting the coefficient data of the prediction error of the subtraction result;
Encoding means for entropy encoding the prediction error coefficient data from the second addition means and transmitting the encoded prediction error coefficient data to the image predictive decoding apparatus;
The coefficient data of the prediction error output from the second adding means is added to the coefficient data of the prediction block output from the determining means, thereby restoring and outputting the quantized current block coefficient data. Third addition means for storing in the block memory;
Dequantizing means for dequantizing and outputting coefficient data of the current block output from the third adding means;
Inverse transform means for generating image data of the restored block by inverse transforming the coefficient data of the current block from the inverse quantization means;
The image data of the restored block output from the compensation means is added to the image data of the restored block from the inverse transform means, so that the image data of the restored block is added to the compensation means. 4th addition means to output to.

An image predictive decoding device according to an eleventh invention is an image predictive decoding device provided corresponding to the image predictive encoding device according to the seventh invention,
Extracting means for extracting instruction bits from the received data received from the image predictive encoding device;
A block memory for storing the coefficient data of the restored block;
Based on the prediction block indicated by the instruction bit extracted by the extraction means, the coefficient data of the current block included in the received data using the coefficient data of the previously restored block stored in the block memory Another prediction means for generating and outputting coefficient data of a prediction block for
Decoding means for entropy decoding the received data and outputting decoded prediction error coefficient data;
Dequantizing means for dequantizing and outputting the coefficient data of the prediction error output from the decoding means;
By adding the coefficient data of the prediction block output from the other prediction means to the coefficient data of the prediction error output from the inverse quantization means, the coefficient data of the current current block is restored and output. Third addition means for storing in the block memory;
Another inverse transformation means for inversely transforming the coefficient data of the current block output from the third addition means and outputting the restored image data of the current block.

An image predictive decoding apparatus according to a twelfth aspect of the present invention is an image predictive decoding apparatus provided corresponding to the image predictive encoding apparatus of the eighth aspect of the invention,
Extracting means for extracting instruction bits from the received data received from the image predictive encoding device;
A block memory for storing the coefficient data of the restored block;
Based on the prediction block indicated by the instruction bit extracted by the extraction means, the coefficient data of the current block included in the received data using the coefficient data of the previously restored block stored in the block memory Another prediction means for generating and outputting coefficient data of a prediction block for
Decoding means for entropy decoding the received data and outputting decoded prediction error coefficient data;
By adding the coefficient data of the prediction block output from the prediction unit to the coefficient data of the prediction error output from the decoding unit, the coefficient data of the current block is restored and output, and the block A third adding means for storing in the memory;
Dequantization means for dequantizing and outputting prediction error coefficient data output from the third addition means;
Another inverse transform unit that inversely transforms the coefficient data of the current block output from the inverse quantization unit and outputs the restored image data of the current block.

Further, an image predictive decoding apparatus according to a thirteenth aspect of the present invention is an image predictive decoding apparatus provided corresponding to the image predictive encoding apparatus according to the ninth aspect of the present invention,
Extracting means for extracting instruction bits from the received data received from the image predictive encoding device;
A block memory for storing the coefficient data of the restored block;
Based on the prediction block indicated by the instruction bit extracted by the extraction means, the coefficient data of the current block included in the received data using the coefficient data of the previously restored block stored in the block memory Another prediction means for generating and outputting coefficient data of a prediction block for
Decoding means for entropy decoding the received data and outputting decoded prediction error coefficient data;
Dequantizing means for dequantizing and outputting the coefficient data of the prediction error output from the decoding means;
By adding the coefficient data of the prediction block output from the other prediction means to the coefficient data of the prediction error output from the inverse quantization means, the coefficient data of the current current block is restored and output. Third addition means for storing in the block memory;
Another inverse transformation means for inversely transforming the coefficient data of the current block output from the third addition means and outputting the restored image data of the current block;
Another compensation means for outputting motion compensation prediction error data by performing motion compensation processing on the image data of the current block output from the other inverse transform means;
The motion compensation prediction error data output from the other compensation unit is subtracted from the image data of the current block output from the other inverse transform unit, and the image data of the block whose subtraction result is restored is output. And a fifth adding means.

Still further, an image predictive decoding apparatus according to a fourteenth aspect of the present invention is an image predictive decoding apparatus provided corresponding to the image predictive encoding apparatus according to the tenth aspect of the present invention,
Extracting means for extracting instruction bits from the received data received from the image predictive encoding device;
A block memory for storing the coefficient data of the restored block;
Based on the prediction block indicated by the instruction bit extracted by the extraction means, the coefficient data of the current block included in the received data using the coefficient data of the previously restored block stored in the block memory Another prediction means for generating and outputting coefficient data of a prediction block for
Decoding means for entropy decoding the received data and outputting decoded prediction error coefficient data;
By adding the coefficient data of the prediction block output from the prediction unit to the coefficient data of the prediction error output from the decoding unit, the coefficient data of the current block is restored and output, and the block A third adding means for storing in the memory;
Dequantization means for dequantizing and outputting prediction error coefficient data output from the third addition means;
Another inverse transformation means for inversely transforming the coefficient data of the current block output from the inverse quantization means and outputting the restored image data of the current block;
Another compensation means for outputting motion compensation prediction error data by performing motion compensation processing on the image data of the current block output from the other inverse transform means;
The motion compensation prediction error data output from the other compensation unit is subtracted from the image data of the current block output from the other inverse transform unit, and the image data of the block whose subtraction result is restored is output. And a fifth adding means.

An image predictive encoding device according to a fifteenth aspect of the present invention is a sampling means for sampling an input image signal into image data of a plurality of blocks each including a two-dimensional array of pixel values;
Conversion means for converting the image data of the block sampled by the sampling means into coefficient data of a predetermined conversion area;
A block memory for storing the coefficient data of the restored block;
Prediction means for forming coefficient data of a plurality of prediction blocks based on the coefficient data of the block converted by the conversion means, based on the coefficient data of the block reconstructed before stored in the block memory;
An instruction for determining and selecting the coefficient data and scanning method of the most efficient prediction block among the coefficient data of the plurality of prediction blocks formed by the prediction means, selecting and outputting the data, and indicating the selected prediction block and scanning method Determining means for transmitting the child to the image predictive decoding apparatus in the form of instruction bits;
First addition means for outputting coefficient data of prediction error of the subtraction result by subtracting coefficient data of the prediction block selected by the determination means from coefficient data of the current block at the present time;
Quantization means for quantizing prediction error coefficient data output from the first addition means;
A scanning unit that performs a scanning process on the coefficient data of the prediction error from the quantization unit by the scanning method determined by the determination unit, and outputs coefficient data of the prediction error after the scanning process;
Encoding means for entropy encoding the prediction error coefficient data output from the scanning means, and transmitting the encoded prediction error coefficient data to the image predictive decoding device;
Dequantizing the coefficient data of the prediction error from the quantizing means, and outputting the restored block coefficient data; and
By adding the coefficient data of the prediction block output from the determination unit to the coefficient data of the prediction error output from the inverse quantization unit, the coefficient data of the restored block is output and also to the block memory A second adding means for storing;
And inverse transform means for generating image data of the restored block by inversely transforming the coefficient data of the block output from the second adding means.

According to a sixteenth aspect of the present invention, there is provided an image predictive encoding device, wherein the input image signal is sampled into a plurality of blocks of image data including a two-dimensional array of pixel values, and sampling means;
Conversion means for converting image data of a plurality of blocks sampled by the sampling means into coefficient data of a predetermined conversion area;
Quantization means for quantizing coefficient data in the transform region from the transform means;
A block memory for storing the coefficient data of the restored block;
Prediction means for forming coefficient data of a plurality of prediction blocks based on the coefficient data of the block converted by the conversion means, based on the coefficient data of the block reconstructed before stored in the block memory;
An instruction for determining and selecting the coefficient data and scanning method of the most efficient prediction block among the coefficient data of the plurality of prediction blocks formed by the prediction means, selecting and outputting the data, and indicating the selected prediction block and scanning method Determining means for transmitting the child to the image predictive decoding apparatus in the form of instruction bits;
First addition means for outputting coefficient data of prediction error of the subtraction result by subtracting coefficient data of the prediction block selected by the determination means from coefficient data of the current block at the present time;
A scan unit that performs a scan process on the coefficient data of the prediction error from the first adding unit by the scan method determined by the determination unit, and outputs the coefficient data of the prediction error after the scan process;
Encoding means for entropy encoding the prediction error coefficient data output from the scanning means, and transmitting the encoded prediction error coefficient data to the image predictive decoding device;
By adding the coefficient data of the prediction block output from the determination means to the coefficient data of the prediction error from the first addition means, the coefficient data of the quantized current block is restored and output, Second addition means for storing in the block memory;
Dequantizing means for dequantizing and outputting the coefficient data of the current block output from the second adding means;
And inverse transform means for generating image data of the restored block by inverse transforming the coefficient data of the current block from the inverse quantization means.
Furthermore, the image predictive coding apparatus according to the seventeenth aspect of the present invention is a sampling means for sampling an input image signal into image data of a plurality of blocks each including a two-dimensional array of pixel values;
Compensation means for generating and outputting image data of a prediction error of a block subjected to motion compensation by executing motion compensation processing on the image data of the input block;
First addition means for subtracting the image data of the prediction error of the block output from the compensation means from the image data of the block output from the sampling means, and outputting the image data of the block as a subtraction result;
Conversion means for converting the image data of the block output from the first addition means into coefficient data of a predetermined conversion area;
A block memory for storing the coefficient data of the restored block;
Prediction means for forming coefficient data of a plurality of prediction blocks based on the coefficient data of the block converted by the conversion means, based on the coefficient data of the block reconstructed before stored in the block memory;
An instruction for determining and selecting the coefficient data and scanning method of the most efficient prediction block among the coefficient data of the plurality of prediction blocks formed by the prediction means, selecting and outputting the data, and indicating the selected prediction block and scanning method Determining means for transmitting the child to the image predictive decoding apparatus in the form of instruction bits;
Subtracting the coefficient data of the prediction block selected by the determination means from the coefficient data of the current block at the current time, thereby outputting the coefficient data of the prediction error of the subtraction result;
Quantization means for quantizing prediction error coefficient data output from the second addition means;
A scanning unit that performs a scanning process on the coefficient data of the prediction error from the quantization unit by the scanning method determined by the determination unit, and outputs coefficient data of the prediction error after the scanning process;
Encoding means for entropy encoding the prediction error coefficient data output from the scanning means, and transmitting the encoded prediction error coefficient data to the image predictive decoding device;
Dequantizing the coefficient data of the prediction error from the quantizing means, and outputting the restored block coefficient data; and
By adding the coefficient data of the prediction block output from the determination unit to the coefficient data of the prediction error output from the inverse quantization unit, the coefficient data of the restored block is output, and the block memory is also output. A third adding means for storing;
Inverse transform means for generating image data of the restored block by inversely transforming the coefficient data of the block output from the third adding means;
The image data of the restored block output from the compensation means is added to the image data of the restored block from the inverse transform means, so that the image data of the restored block is added to the compensation means. 4th addition means to output to.

Still further, an image predictive encoding device according to an eighteenth aspect of the present invention is a sampling means for sampling an input image signal into a plurality of blocks of image data including a two-dimensional array of pixel values;
Compensation means for generating and outputting image data of a prediction error of a block subjected to motion compensation by executing motion compensation processing on the image data of the input block;
First addition means for subtracting the image data of the prediction error of the block output from the compensation means from the image data of the block output from the sampling means, and outputting the image data of the block as a subtraction result;
Conversion means for converting the image data of the block output from the first addition means into coefficient data of a predetermined conversion area;
Quantization means for quantizing coefficient data in the transform region from the transform means;
A block memory for storing the coefficient data of the restored block;
Prediction means for forming coefficient data of a plurality of prediction blocks based on the coefficient data of the block converted by the conversion means, based on the coefficient data of the block reconstructed before stored in the block memory;
An instruction for determining and selecting the coefficient data and scanning method of the most efficient prediction block among the coefficient data of the plurality of prediction blocks formed by the prediction means, selecting and outputting the data, and indicating the selected prediction block and scanning method Determining means for transmitting the child to the image predictive decoding apparatus in the form of instruction bits;
Subtracting the coefficient data of the prediction block selected by the determination means from the coefficient data of the current block at the current time, thereby outputting the coefficient data of the prediction error of the subtraction result;
A scan unit that performs a scan process on the prediction error coefficient data from the second addition unit by the scan method determined by the determination unit, and outputs the prediction error coefficient data after the scan process;
Encoding means for entropy encoding the prediction error coefficient data output from the scanning means, and transmitting the encoded prediction error coefficient data to the image predictive decoding device;
The coefficient data of the prediction error output from the second adding means is added to the coefficient data of the prediction block output from the determining means, thereby restoring and outputting the quantized current block coefficient data. Third addition means for storing in the block memory;
Dequantizing means for dequantizing and outputting coefficient data of the current block output from the third adding means;
Inverse transform means for generating image data of the restored block by inverse transforming the coefficient data of the current block from the inverse quantization means;
The image data of the restored block output from the compensation means is added to the image data of the restored block from the inverse transform means, so that the image data of the restored block is added to the compensation means. 4th addition means to output to.

An image predictive decoding device according to a nineteenth invention is an image predictive decoding device provided corresponding to the image predictive encoding device according to the fifteenth invention,
Extracting means for extracting instruction bits from the received data received from the image predictive encoding device;
A block memory for storing the coefficient data of the restored block;
Based on the prediction block indicated by the instruction bit extracted by the extraction means, the coefficient data of the current block included in the received data using the coefficient data of the previously restored block stored in the block memory Another prediction means for generating and outputting coefficient data of a prediction block for
Decoding means for entropy decoding the received data and outputting decoded prediction error coefficient data;
The coefficient data of the prediction error output from the decoding unit is subjected to reverse scanning processing based on the scanning method indicated by the instruction bit extracted by the extracting unit, and the prediction error after the reverse scanning processing is calculated. Reverse scanning means for outputting coefficient data;
Inverse quantization means for inversely quantizing and outputting prediction error coefficient data output from the inverse scan means after the inverse scan processing;
By adding the coefficient data of the prediction block output from the other prediction means to the coefficient data of the prediction error output from the inverse quantization means, the coefficient data of the current current block is restored and output. Third addition means for storing in the block memory;
Another inverse transformation means for inversely transforming the coefficient data of the current block output from the third addition means and outputting the restored image data of the current block.

The image predictive decoding apparatus according to the twentieth invention is an image predictive decoding apparatus provided corresponding to the image predictive encoding apparatus according to the sixteenth invention,
Extracting means for extracting instruction bits from the received data received from the image predictive encoding device;
A block memory for storing the coefficient data of the restored block;
Based on the prediction block indicated by the instruction bit extracted by the extraction means, the coefficient data of the current block included in the received data using the coefficient data of the previously restored block stored in the block memory Another prediction means for generating and outputting coefficient data of a prediction block for
Decoding means for entropy decoding the received data and outputting decoded prediction error coefficient data;
The coefficient data of the prediction error output from the decoding unit is subjected to reverse scanning processing based on the scanning method indicated by the instruction bit extracted by the extracting unit, and the prediction error after the reverse scanning processing is calculated. Reverse scanning means for outputting coefficient data;
The coefficient data of the prediction block output from the prediction unit is added to the coefficient data of the prediction error output from the inverse scanning unit, thereby restoring and outputting the coefficient data of the current current block, and the block A third adding means for storing in the memory;
Dequantization means for dequantizing and outputting prediction error coefficient data output from the third addition means;
Another inverse transform unit that inversely transforms the coefficient data of the current block output from the inverse quantization unit and outputs the restored image data of the current block.

The image predictive decoding apparatus according to the twenty-first invention is an image predictive decoding apparatus provided corresponding to the image predictive encoding apparatus according to the seventeenth invention,
Extracting means for extracting instruction bits from the received data received from the image predictive encoding device;
A block memory for storing the coefficient data of the restored block;
Based on the prediction block indicated by the instruction bit extracted by the extraction means, the coefficient data of the current block included in the received data using the coefficient data of the previously restored block stored in the block memory Another prediction means for generating and outputting coefficient data of a prediction block for
Decoding means for entropy decoding the received data and outputting decoded prediction error coefficient data;
The coefficient data of the prediction error output from the decoding unit is subjected to reverse scanning processing based on the scanning method indicated by the instruction bit extracted by the extracting unit, and the prediction error after the reverse scanning processing is calculated. Reverse scanning means for outputting coefficient data;
Inverse quantization means for inversely quantizing and outputting prediction error coefficient data output from the inverse scan means after the inverse scan processing;
By adding the coefficient data of the prediction block output from the other prediction means to the coefficient data of the prediction error output from the inverse quantization means, the coefficient data of the current current block is restored and output. Third addition means for storing in the block memory;
Another inverse transformation means for inversely transforming the coefficient data of the current block output from the third addition means and outputting the restored image data of the current block;
Another compensation means for outputting motion compensation prediction error data by performing motion compensation processing on the image data of the current block output from the other inverse transform means;
The motion compensation prediction error data output from the other compensation unit is subtracted from the image data of the current block output from the other inverse transform unit, and the image data of the block whose subtraction result is restored is output. And a fifth adding means.
Still further, an image predictive decoding apparatus according to a twenty-second aspect of the present invention is an image predictive decoding apparatus provided corresponding to the image predictive encoding apparatus according to the eighteenth aspect of the invention,
Extracting means for extracting instruction bits from the received data received from the image predictive encoding device;
A block memory for storing the coefficient data of the restored block;
Based on the prediction block indicated by the instruction bit extracted by the extraction means, the coefficient data of the current block included in the received data using the coefficient data of the previously restored block stored in the block memory Another prediction means for generating and outputting coefficient data of a prediction block for
Decoding means for entropy decoding the received data and outputting decoded prediction error coefficient data;
The coefficient data of the prediction error output from the decoding unit is subjected to reverse scanning processing based on the scanning method indicated by the instruction bit extracted by the extracting unit, and the prediction error after the reverse scanning processing is calculated. Reverse scanning means for outputting coefficient data;
The coefficient data of the prediction block output from the prediction unit is added to the coefficient data of the prediction error output from the inverse scanning unit, thereby restoring and outputting the coefficient data of the current current block, and the block A third adding means for storing in the memory;
Dequantization means for dequantizing and outputting prediction error coefficient data output from the third addition means;
Another inverse transformation means for inversely transforming the coefficient data of the current block output from the inverse quantization means and outputting the restored image data of the current block;
Another compensation means for outputting motion compensation prediction error data by performing motion compensation processing on the image data of the current block output from the other inverse transform means;
The motion compensation prediction error data output from the other compensation unit is subtracted from the image data of the current block output from the other inverse transform unit, and the image data of the block whose subtraction result is restored is output. And a fifth adding means.

  An image predictive coding method according to a twenty-third aspect includes a step in which each means in the image predictive coding apparatus is replaced with each step.

  Furthermore, an image predictive decoding method according to a twenty-fourth aspect includes a step in which each means in the image predictive decoding apparatus is replaced with each step.

  A recording medium according to a twenty-fifth aspect of the invention is a recording medium on which a program including each step in the image predictive encoding method is recorded.

  Furthermore, a recording medium according to a twenty-sixth aspect of the invention is a recording medium on which a program including each step in the image predictive decoding method is recorded.

As described above in detail, according to the image predictive encoding device according to the present invention, the dividing means for dividing the input encoded image data into image data of a plurality of small regions adjacent to each other;
When the image data of the small area to be processed is encoded among the image data of the plurality of small areas adjacent to each other divided by the dividing unit, the image data reproduced adjacent to the image data of the small area to be processed is encoded. The image data of the reproduction subregion is the image data of the intra prediction small region of the processing target small region, the image data of the intra prediction small region is the image data of the optimal prediction subregion, and First generation means for generating difference subregion image data that is a difference between the image data and the optimal prediction subregion image data;
Encoding means for encoding the image data of the small difference area generated by the generating means;
Decoding means for decoding the image data of the small difference area encoded by the encoding means;
Second generation means for generating reproduced subregion image data reproduced by adding the difference subregion image data decoded by the decoding unit to the optimum predicted subregion image data.
Therefore, simply by using the reconstructed pixel value adjacent to the image data of the small area to be processed as the pixel value of the in-screen prediction signal, a highly accurate prediction signal can be easily generated with a small amount of computation compared to the conventional technology. And a unique effect that the number of bits of intra-frame coding can be reduced.
Further, according to the image predictive coding apparatus according to the present invention, sampling means for sampling the input image signal into image data of a plurality of blocks each including a two-dimensional array of pixel values;
Conversion means for converting the image data of the block sampled by the sampling means into coefficient data of a predetermined conversion area;
A block memory for storing the coefficient data of the restored block;
Prediction means for forming coefficient data of a plurality of prediction blocks based on the coefficient data of the block converted by the conversion means, based on the coefficient data of the block reconstructed before stored in the block memory;
Among the coefficient data of the plurality of prediction blocks formed by the prediction means, the coefficient data of the prediction block with the highest efficiency is determined, selected and output, and the indicator representing the selected prediction block is in the format of the instruction bit Determining means for transmitting to the image predictive decoding device,
First addition means for outputting coefficient data of prediction error of the subtraction result by subtracting coefficient data of the prediction block selected by the determination means from coefficient data of the current block at the present time;
Quantization means for quantizing prediction error coefficient data output from the first addition means;
Encoding means for entropy encoding the prediction error coefficient data from the quantization means, and transmitting the encoded prediction error coefficient data to the image predictive decoding device;
Dequantizing the coefficient data of the prediction error from the quantizing means, and outputting the restored block coefficient data; and
By adding the coefficient data of the prediction block output from the determination unit to the coefficient data of the prediction error output from the inverse quantization unit, the coefficient data of the restored block is output and also to the block memory A second adding means for storing;
And inverse transform means for generating image data of the restored block by inversely transforming the coefficient data of the block output from the second adding means.
Accordingly, it is possible to provide a new image predictive coding apparatus and image predictive decoding apparatus that increase the current coding efficiency. In this apparatus, no complicated means is required to increase the coding efficiency, and the circuit configuration can be formed very simply and easily.
Further, according to the image predictive coding apparatus according to the present invention, sampling means for sampling the input image signal into a plurality of blocks of image data each including a two-dimensional array of pixel values;
Conversion means for converting the image data of the block sampled by the sampling means into coefficient data of a predetermined conversion area;
A block memory for storing the coefficient data of the restored block;
Prediction means for forming coefficient data of a plurality of prediction blocks based on the coefficient data of the block converted by the conversion means, based on the coefficient data of the block reconstructed before stored in the block memory;
An instruction for determining and selecting the coefficient data and scanning method of the most efficient prediction block among the coefficient data of the plurality of prediction blocks formed by the prediction means, selecting and outputting the data, and indicating the selected prediction block and scanning method Determining means for transmitting the child to the image predictive decoding apparatus in the form of instruction bits;
First addition means for outputting coefficient data of prediction error of the subtraction result by subtracting coefficient data of the prediction block selected by the determination means from coefficient data of the current block at the present time;
Quantization means for quantizing prediction error coefficient data output from the first addition means;
A scanning unit that performs a scanning process on the coefficient data of the prediction error from the quantization unit by the scanning method determined by the determination unit, and outputs coefficient data of the prediction error after the scanning process;
Encoding means for entropy encoding the prediction error coefficient data output from the scanning means, and transmitting the encoded prediction error coefficient data to the image predictive decoding device;
Dequantizing the coefficient data of the prediction error from the quantizing means, and outputting the restored block coefficient data; and
By adding the coefficient data of the prediction block output from the determination unit to the coefficient data of the prediction error output from the inverse quantization unit, the coefficient data of the restored block is output and also to the block memory A second adding means for storing;
And inverse transform means for generating image data of the restored block by inversely transforming the coefficient data of the block output from the second adding means.
Therefore, it is very effective in reducing or eliminating redundancy in the transform domain beyond adjacent blocks, reducing the number of bits used and, as a result, greatly improving the coding efficiency. be able to. This is also useful as a tool in new video compression algorithms.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.

<First embodiment group>
The first embodiment group includes first to fourth embodiments.

<First Embodiment>
FIG. 1 is a block diagram showing the configuration of an image predictive coding apparatus according to the first embodiment of the present invention.

  In FIG. 1, 101 is an input terminal, 102 is a first adder, 103 is an encoder, 106 is an output terminal, 107 is a decoder, 110 is a second adder, 111 is a line memory, and 112 is a prediction. It is a signal generator.

  Hereinafter, the configuration and operation of the image predictive coding apparatus will be described. Image data to be encoded is input to the input terminal 101. Here, the input image data is divided into a plurality of adjacent small areas.

  2 shows an image of input image data when divided into 8 × 8 small areas, and FIG. 3 shows an image of input image data when divided into triangular small areas. The image data of a plurality of small areas are sequentially encoded. The image data of the small area to be processed is input to the adder 102 via the input terminal 101 and the line 113. On the other hand, the prediction signal generator 112 generates image data of the intra prediction small region, and outputs the generated image data to the adder 102 via the line 121 as image data of the optimal prediction small region.

  The adder 102 subtracts the corresponding pixel value of the optimal prediction small region from the prediction signal generator 112 from the pixel value of the input image data in the small region to be processed, and the difference small region image data of the subtraction result Is output to the encoder 103 and the compression encoding process is executed. In the present embodiment, the encoder 103 includes a DCT converter 104 and a quantizer (Q) 105, and the image data in the small difference area is converted into an image signal in the frequency domain by the DCT converter 104, and a DCT conversion coefficient is obtained. Get. Next, the DCT transform coefficient is quantized by the quantizer 105. The quantized image data of the small area is output to the output terminal 106 via the line 116, and further converted into a variable length or fixed length code, and then stored in a recording medium such as an optical disk or via a communication line. (Not shown).

  At the same time, the quantized image data of the small region is input to the decoder 107, where the decoder 107 includes an inverse quantizer 108 and an inverse DCT converter 109, and the input small region Is restored to the image data of the expansion difference small area. In the present embodiment, the input small-area image data is inversely quantized by the inverse quantizer 108, and the inversely quantized image data is referred to as an inverse discrete cosine transformer (hereinafter referred to as an inverse DCT transformer). .) 109 is converted into an image signal in the spatial domain. The image data of the expanded difference small area obtained in this way is output to the adder 110, and the adder 110 outputs the image data of the expanded difference small area from the prediction signal generator 112 via the line 121 and the line 122. The optimal prediction image signal to be added is added to generate image data of the reproduction small area, and the reproduction pixel value for generating the prediction image signal in the screen is stored in the line memory 111 from the image data of the reproduction small area To do. The prediction signal generator 112 generates image data of the intra prediction small area as follows. That is, the prediction signal generator 112 generates the pixel value of the reproduced image data adjacent to the image data of the small region to be processed as the pixel value of the image data of the intra prediction small region.

In FIG. 2, if the block 200 is a small area to be processed, the pixel values of the adjacent reproduced image data are a ₀ , a ₁ , a ₂ ,..., A ₆ , a ₇ , b ₀ , b ₁ , b _2, ..., it is a b _6, b _7. 3, when the triangle 301 and the objective small region to be processed, the pixel values of the adjacent reconstructed image data _{g 0, g 1, ...,} g 4, f 0, f 1, f 2, ..., f 7 , it is a f _8. Also, if the objective small region to be processed triangle 300 of FIG. 3, the pixel values of the adjacent reconstructed image data _{e 0, h 0, h 1} , ..., a h _4. These pixel values are stored in the line memory 111. The prediction signal generator 112 accesses the line memory 111 and reads the pixel value of the adjacent image data as the pixel value of the image data of the intra prediction small area.

  4 and 5 are block diagrams showing the configurations of the first and second embodiments of the prediction signal generator used in the image predictive coding apparatus of FIG. 1, respectively.

In FIG. 4, pixel values a ₀ , a ₁ , a ₂ ,..., A ₆ , a _{7 that} are adjacent to the small area to be processed in the horizontal direction are input from the line memory 111 to the prediction signal generator 112 to be predicted. The generator 401 in the signal generator 112 generates the image data 403 of the predicted small area in the screen by repeatedly outputting the same pixel in the horizontal direction, for example, eight times. Here, the image data 403 of the intra prediction small region is used when there is no pixel adjacent in the vertical direction to the small region to be processed.

In FIG. 5, pixel values b ₀ , b ₁ , b ₂ ,..., B ₆ , b _{7 that} are adjacent in the vertical direction with respect to the small region to be processed are input from the line memory 111 to the prediction signal generator 112 to be predicted. The generator 402 in the signal generator 112 generates image data 404 of the predicted small area in the screen by repeatedly outputting the pixels in the vertical direction, for example, eight times. Here, the image data 404 of the intra prediction small area is used when there is no pixel adjacent in the horizontal direction to the small area to be processed. When there are adjacent pixel values in both the horizontal direction and the vertical direction, the image data of the intra prediction small area is generated as in the third embodiment shown in FIG.

  FIG. 6 is a block diagram showing the configuration of a third embodiment of the prediction signal generator used in the image predictive coding apparatus of FIG.

  In FIG. 6, the image data 403 (see FIG. 5) of the intra prediction small area generated by the generator 401 and the image data 404 of the intra prediction small area generated by the generator 402 are sent to the adder 500. The inputted adder 500 divides the sum of the inputted two image data by 2, and averages these two image data. As described above, since the regenerated pixels adjacent to each other are repeatedly output by the generators 401 and 402 and only the averaging operation is performed by the adder 500, the image data of the intra prediction small area can be generated at high speed. . Note that the image data of the intra prediction small area may be generated by linearly interpolating the pixel values of two adjacent image data.

  FIG. 7 is a block diagram showing the configuration of the fourth embodiment of the prediction signal generator used in the image predictive coding apparatus of FIG.

In FIG. 7, pixel values a ₀ , a ₁ , a ₂ ,..., A ₆ , a ₇ of image data adjacent to the small area to be processed in the horizontal direction are input from the line memory 111 to the generator 401. The generator 401 generates image data of the first intra prediction small area by repeatedly outputting pixels in the horizontal direction. On the other hand, pixel values b ₀ , b ₁ , b ₂ ,..., B ₆ , b _{7 that} are adjacent to the small area to be processed in the vertical direction are input from the line memory 111 to the generator 402, and the generator 402 Then, the image data of the second intra prediction small area is generated by repeatedly outputting the pixels in the vertical direction. The image data of the first intra-screen prediction small region and the image data of the second intra-screen prediction small region are input to the adder 500, and by averaging these two image data, the third intra-screen prediction small region is obtained. Generate image data for the region.

  On the other hand, the image data of the small area to be processed is input to the error calculators 601, 602, and 603 via the line 616. Here, the image data of the first predicted small area in the screen and the image data of the small area to be processed are input to the error calculator 601, and the error calculator 601 calculates the absolute value of the error between the two image data. The first absolute error is calculated and output to the comparator 604. Further, the image data of the second intra-screen prediction small region and the image data of the small region to be processed are input to the error calculator 602, and the error calculator 602 is the absolute value of the error between these two image data. The second absolute error is calculated and output to the comparator 604. Further, the image data of the third predicted small area in the screen and the image data of the small area to be processed are input to the error calculator 603, which calculates the absolute value of the error between these two image data. A third absolute error is calculated and output to the comparator 604.

  The comparator 604 compares the input three absolute errors with each other, determines the one with the smallest absolute error, and switches the switch 605 so as to output the image data of the predicted intra-screen small region to the line 121 corresponding thereto. Control. At the same time, the comparator 604 outputs an identifier for identifying the image data of the first, second, and third intra-screen prediction small regions to the reception side or reproduction side device via the line 615. By this identifier, the image data of the intra prediction small area is uniquely determined on the reception side or the reproduction side. By using the image data of the intra prediction small area with the smallest error in this way, the differential signal at the time of encoding can be suppressed and the number of generated bits can be reduced.

<Second Embodiment>
FIG. 8 is a block diagram showing a configuration of an image predictive coding apparatus according to the second embodiment of the present invention, and the same components as those in FIG. 1 are denoted by the same reference numerals. The image predictive coding apparatus in FIG. 8 includes a motion detector 700, a motion compensator 701, an optimum mode selector 703, and a frame memory 702, as compared with the image predictive coding apparatus in FIG. Features.

  Hereinafter, the configuration and operation of the image predictive coding apparatus in FIG. 8 will be described.

  As in the first embodiment, the input image data of the small area to be processed is input to the adder 102 via the input terminal 101, and the adder 102 converts the image data of the small area to be processed to After subtraction from the image data of the optimal prediction small region input from the optimal mode selector 703 via the line 121, the image data of the subtraction result is output to the encoder 103. The encoder 103 compresses and encodes the input subtraction image data and outputs it via the output terminal 106. At the same time, the encoder 103 outputs the compression-encoded image data of the small area to the decoder 107 for decompression decoding. Then, the image data is output to the adder 110, and the decompressed and decoded image data is added to the image data of the optimal prediction small area.

  Next, as in the first embodiment, only the pixel values of the image data used for generating the image data of the intra prediction small area are stored in the line memory 111, while all the pixel values of the reproduced image are stored. Store in the frame memory 702.

  When the image data of the next image is input via the input terminal 101, the motion detector 700 is input with the image data of the small area to be processed and the reproduced image data stored in the frame memory 702. The motion detector 700 detects the motion of the image by a method such as block matching and outputs a motion vector via the line 705. The output motion vector is, for example, variable-length encoded and stored or transmitted (not shown) and sent to the motion compensator 701 at the same time. The motion compensator 701 generates temporal prediction small region image data from the reproduced image in the frame memory 702 based on the motion vector, and outputs the image data to the optimum mode selector 703. The motion detection process and the motion compensation process include forward prediction, backward prediction, and bidirectional prediction, and these methods are disclosed in, for example, US Pat. No. 5,193,004.

  On the other hand, as in the first embodiment, the prediction signal generator 112 generates image data of an intra-screen prediction small region and outputs it to the optimum mode selector 703, and at the same time, optimizes the image data of the small region to be processed. Output to the mode selector 703. The optimum mode selector 703 obtains the most error (for example, the sum of absolute values of differences for each pixel) from the image data of the intra prediction small area and the image data of the temporal prediction small area to the image data of the small area to be processed. ) Is selected, and the selected image data is output to the adder 102 as image data of the optimal prediction small region. Further, an identifier indicating which predicted small region of image data has been selected is output and transmitted to the reception side or the reproduction side via the line 709.

  In this way, it is not necessary to transmit an inter-frame motion vector by introducing intra-picture prediction into inter-frame motion compensation encoded image data, and therefore the number of bits can be further reduced.

  In the first and second embodiments, there is a case where significant pixels exist in the entire screen. There may be pixels that are not so significant in the screen. For example, in an image photographed with a chroma key, pixels representing a subject are significant and pixels representing a region such as blue as a background are insignificant pixels. By encoding and transmitting a significant object texture and its shape, it is possible to reproduce and display object units. For such an input image, when the prediction signal generator 112 generates image data of an intra prediction small area, an insignificant pixel value cannot be used.

  9 to 11 are schematic diagrams of input images having significant pixels and non-significant pixels. In this embodiment, a shape signal is used to indicate whether a pixel is significant. The shape signal is compression-coded by a predetermined method and transmitted to the receiving side or the reproducing side. As a method of coding the shape, there is a method such as chain coding. The compressed shape signal is expanded and reproduced again, and an intra-screen prediction signal is generated using the reproduced shape signal as described below.

In FIG. 9, the shape curve 800 is a boundary line, the direction indicated by the arrow is inside the object, and the image data inside the object is composed of significant pixels. Among the reproduced pixels adjacent to the processing target small area 802, b ₄ , b ₅ , b ₆ , and b ₇ are significant pixels, and only these pixel values are repeated to display the screen of the processing target small area 802. The pixel value of the inner prediction small area is used.

In FIG. 10, the shape curve 804 is a boundary, the direction indicated by the arrow is inside the object, and the image data inside the object is composed of significant pixels. Among the reconstructed pixels adjacent to the small area 805 to be processed, a ₄ , a ₅ , a ₆ and a ₇ are significant pixels, and only these pixel values are repeated to repeat the small area 805 to be processed. The pixel value of the predicted small area in the screen.

Further, in FIG. 11, a curve 808 is a boundary line, the direction indicated by the arrow is inside the object, and image data inside the object is composed of significant pixels. Among the reproduced pixels adjacent to the small area 810 to be processed, a ₅ , a ₆ , a ₇ , b ₄ , b ₅ , b ₆ , b ₇ are significant pixels, and only these pixel values Are repeatedly output, and where two pixel values overlap, a value obtained by averaging the pixel values is set as a pixel value of the intra prediction small area of the small area 810 to be processed.

In FIG. 11, for example, the value of the pixel z ₇₇ of the small area 810 to be processed is an average value of a ₇ and b ₇ . Where there are no pixel values, an average value of two pixel values adjacent in the horizontal direction and the vertical direction is taken. For example, the value of the pixel z ₁₄ is an average value of a ₅ and b ₄ . In this way, image data of a screen prediction small area of an image having an arbitrary shape is generated.

  In the above embodiment, the small area divided into squares has been described. However, the present invention is not limited to this, and the screen may be divided into triangular small areas as in FIG. Even in this case, the image processing is executed in the same manner.

As another embodiment, an average value may be obtained using only significant pixel values, and the average value may be used as the pixel value of the intra prediction small area. Specifically, in FIG. 9, the average value of the pixels b ₄ , b ₅ , b ₆ , and b ₇ is calculated, and the calculated average value is set as the pixel value of the intra prediction small area. In FIG. 10, the average value of the pixels a ₄ , a ₅ , a ₆ , and a ₇ is calculated, and the calculated average value is set as the pixel value of the intra prediction small area. In FIG. 11, the average value of the pixels a ₅ , a ₆ , a ₇ , b ₄ , b ₅ , b ₆ , b ₇ is calculated and set as the pixel value of the intra prediction small area.

<Third Embodiment>
FIG. 12 is a block diagram showing a configuration of an image predictive decoding apparatus according to the third embodiment of the present invention.

  In FIG. 12, 901 is an input terminal, 902 is a data analyzer, 903 is a decoder, 906 is an adder, 907 is an output terminal, 908 is a controller, 909 is a motion compensator, 910 is a prediction signal generator, and 911 is A line memory 912 is a frame memory.

  Hereinafter, the configuration and operation of the image predictive decoding apparatus in FIG. 12 will be described. In FIG. 12, the compression-encoded image data is input to the data analyzer 902, which analyzes the input image data and decodes the image data of the small area of the compression difference via the line 915. The control signal is output to the controller 908 via the line 926, and the above-described motion vector (only when present) is output to the motion compensator 909. The decoder 903 includes an inverse quantizer 904 and an inverse DCT converter 905. The decoder 903 decompresses the compressed image data of the small difference region and restores the image data of the decompressed difference small region.

  In the present embodiment, the compressed image data of the small difference area is inversely quantized by the inverse quantizer 904, and the frequency domain image data after inverse quantization is converted to the spatial domain image by the inverse DCT transformer 905. Converted to image data. The converted image data of the expanded difference small region is input to the adder 906, and the adder 906 converts the input image data of the expanded difference small region from the motion compensator 923 or the prediction signal generator 922 to the switch 913 and the line. It adds to the image data of the optimal prediction small area sent via 924, and generates the image data of the reproduction small area as the addition result. The adder 906 outputs the reproduced image data to the output terminal 907 via the line 917 and simultaneously stores it in the frame memory 912. Further, the pixel value of the image data used for generating the image of the intra prediction small area is stored in the line memory 911.

  The image data of the optimal prediction small region is determined by the controller 908 based on the control signal from the data analyzer 902, and the switching of the switch 913 is controlled. When the image data of the predicted small area in the screen is selected by the controller 908, the switch 913 connects the line 924 to the line 922, and in response to the control signal from the controller 908, the prediction signal generator 910 sets the line memory 911. Access is performed, and the adjacent reproduction pixel value is output as the pixel value of the intra prediction small area. Details of the operation of the prediction signal generator 910 have been described in detail above with reference to FIGS. In addition, when image data of the time prediction small region is selected by the controller 908, the switch 913 connects the line 924 to the line 923, and the motion compensator 909 is responsive to the control signal from the controller 908 in response to the data analyzer 902. Based on the motion vector sent from the line 925 to the image data from the frame memory 912, the motion compensation processing is executed on the image data from the frame memory 912 to generate the image data of the temporal prediction small region, and the switch 913 and the line 924 Is output to the adder 906.

<Fourth Embodiment>
FIG. 13 is a block diagram showing a configuration of an image predictive decoding apparatus according to the fourth embodiment of the present invention. In FIG. 13, the same components as those in FIG. 12 are denoted by the same reference numerals. The image predictive decoding apparatus of FIG. 13 is characterized in that a shape decoder 990 is additionally provided in addition to the basic configuration of the image predictive decoding apparatus of FIG. Since the basic operation of the image predictive decoding apparatus in FIG. 13 is also the same as that in FIG. 12, only different operations will be described in detail below.

  In the present embodiment, the compression-encoded image data includes compression-encoded shape data. The data analyzer 902 extracts the shape data and outputs it to the shape decoder 990. In response to this, the shape decoder 990 decompresses and reproduces the shape signal. The reproduced shape signal is transmitted to the receiving side or the reproducing side and simultaneously input to the prediction signal generator 910. Based on the reproduced shape signal, the prediction signal generator 910 generates image data of an intra-screen prediction small region as described with reference to FIGS. 9 to 11. In this manner, it is possible to generate image data of an intra-screen prediction small region of an image having an arbitrary shape, and decode and reproduce the image data on the reception side or the reproduction side.

  A feature of the third and fourth embodiments is that a line memory 911 is provided. Without the line memory 911, the pixel for generating the image data of the intra prediction small area must be accessed from the frame memory 912. In order to generate a prediction signal with pixels in adjacent small regions, it is necessary to write and read the frame memory at high speed. By providing a dedicated line memory and buffer, it is possible to generate image data of the predicted small area in the screen at high speed without using a high-speed frame memory.

  In the above embodiment, the average value of the plurality of pixel values may be a predetermined weighted average value.

  As described above, according to the first exemplary embodiment group of the present invention, the reproduced pixel value adjacent to the image data of the small area to be processed is simply used as the pixel value of the in-screen prediction signal. Compared to the above, it is possible to easily generate a highly accurate prediction signal with a small amount of computation, and to obtain a specific effect that the number of bits of intra-frame coding can be reduced. Further, since the line memory 911 is provided to store the regenerated pixel value used for generating the intra prediction signal, the pixel value can be accessed at high speed, and the intra prediction signal can be processed at high speed. Can be generated.

<Second Embodiment Group>
The second embodiment group includes fifth to seventh embodiments.

  In view of the problems of the prior art, the present inventor considered that the image coding efficiency is not only between two images or between two blocks in one image, but also between two blocks in one image. It has been found that the image coding efficiency is further improved by removing the redundancy.

  The inventor has found that the DCT transform coefficients at the same position of adjacent blocks are often very close. In particular, it has been found that the approximation is high if the texture of the original image for the two blocks is very similar, or if it contains the same image pattern, eg straight lines, corners, etc. The same information means redundancy by information theory.

  This type of redundancy that exists beyond the block and in the DCT transform domain can be removed or greatly reduced by adaptive intra prediction (intraframe prediction) from the previous block. Then, the next VLC entropy encoding process can achieve a higher encoding efficiency due to a small prediction entropy. As a result of this DCT transform domain prediction, the redundant data input to the VLC entropy coding circuit can be greatly reduced. Therefore, a lot of bit savings can be expected. Therefore, the image quality of the encoded drawing data is clearly improved.

  The present embodiment according to the present invention provides a method for appropriately predicting DCT transform coefficients from other blocks. This scheme eliminates redundancy that exists beyond adjacent blocks and reduces the entropy of the quantized DCT transform coefficients, thereby reducing the number of bits required to encode the DCT transform coefficients. be able to.

  The DCT transform coefficient of the current block to be processed (hereinafter referred to as the current block) can be predicted from the DCT transform coefficient at the same position in the previous adjacent block. Adjacent blocks have already been decoded during processing. That is, the first DC coefficient is predicted by the first DC coefficient in one of the neighboring blocks previously decoded. Also, the second coefficient AC1 is predicted from the second coefficient AC1 in the same decoded block. Thereafter, the same processing is performed. By using this method, several predicted blocks are adjacent to the currently encoded DCT transform block on the left side upward, diagonally left side, upward diagonally right side, and upward. Can be obtained from the decoded block. These predicted blocks are checked by performing the actual entropy coding. Then, after a prediction block having a smaller number of bits is selected, it is entropy-coded and transmitted to an image predictive decoding apparatus on the reception side or reproduction side together with additional instruction bits. The image predictive decoding apparatus reports from which adjacent block the current block is predicted.

  The method of the present embodiment according to the present invention can predict the DCT transform coefficient of the current block. The DCT transform coefficient generally has good correlation with the DCT transform coefficients of other adjacent blocks. The reason is that the DCT transform tends to give the same value or the same distribution of DCT transform coefficients to the same block image.

  In general, first, a block-based DCT conversion process is performed on input image data that is an intra frame or a temporarily predicted frame. After the DCT transform coefficients of the current block are obtained, the DCT transform region prediction process can be executed before quantization and after quantization.

  As shown in FIG. 15, the DCT transform coefficients of the current block are blocks that have already been decoded and are adjacent blocks, that is, the upper left block B1, the upper block B2, the upper right block B3, and the left block. It can be predicted from B4. The predicted block is obtained by subtracting all of the DCT transform coefficients of the current block from all of the DCT transform coefficients of the previous neighboring block at the same position. Moreover, it can obtain by subtracting a DCT conversion coefficient partially instead of all the DCT conversion coefficients.

  Predicted DCT transform coefficients in different predicted blocks are quantized if prediction is performed prior to quantization. Next, an entropy encoding process is performed on the DCT transform coefficients. The entropy coding process is the same as that of the image predictive coding apparatus, and it is checked which predicted block is used as the lower bit.

  A prediction block using lower bits is selected, and the selected prediction block is entropy encoded with an instruction bit that informs the image predictive decoding device about the prediction decision.

  In the image predictive decoding apparatus, a block predicted using the instruction bits is decoded. That is, after inverse entropy decoding of a predicted DCT transform coefficient for one block, the DCT transform coefficient for that block is the reference DCT transform coefficient of the adjacent block decoded before being represented by the indicator bit Is added to the decoded DCT transform coefficient. Finally, an inverse DCT transform process is applied to the restored DCT transform coefficients for each block to obtain decoded image data.

  The present embodiment according to the present invention is among spatial redundancy usually removed by transformation such as DCT transformation, redundancy removed between frames by motion detection and compensation, and quantized transform coefficients in a block. In addition to the statistical redundancy removed by entropy coding, an image coding apparatus capable of reducing other types of redundancy existing in a DCT transform region beyond adjacent blocks is provided.

  As can be seen from FIG. 14 showing a prior art image predictive coding apparatus, an image predictive coding apparatus generally used for conventional image coding (for example, in MPEG) includes a block sampling unit 1001, a DCT transform. A unit 1004, a quantizer 1005 and an entropy encoder 1006 are provided.

  In intra-frame coding (intra-frame coding), block sampling processing is first performed on an input image signal. Next, the DCT conversion process is directly executed. Subsequently, a quantization process and an entropy encoding process are performed. On the other hand, in the inter-frame coding (predictive frame coding), after the block sampling process, the process of the motion detection unit 1002 and the motion compensation unit 1003 is performed on the image data of the current frame to be processed, Further, a DCT conversion process is executed. Further, a quantization process and an entropy encoding process are executed.

  Here, in the entropy coding unit 1006, the quantized value is entropy-coded and code data is output. Entropy coding is performed by assigning short codewords to frequently occurring values and long codewords to values that do not occur frequently, so that encoding approaches the entropy that is the average information amount, and the total code amount This is a method that greatly reduces This is lossless encoding. Various schemes have been proposed for entropy coding, but the baseline system uses Huffman coding. The quantized DC coefficient value and the AC coefficient value have different Huffman coding methods, that is, the DC coefficient indicates an average value of an 8 × 8 pixel block, but in an ordinary image, the average value with an adjacent block is Often they have similar values. Therefore, entropy coding is performed after taking the difference from the previous block. In this way, since the values are concentrated near 0, entropy coding is effective. For the AC coefficient, for example, zigzag scanning is performed to convert two-dimensional data into one-dimensional data. Further, since many AC coefficients including high-frequency components are generated in many cases, entropy coding is performed with a set of AC coefficient values having a value other than 0 and the number of 0s before that (run length) as a set.

  The rate controller 1007 feeds back the bits used in the previously encoded block, controls the processing of the quantization unit 1005 and adjusts the code bit rate. Here, the rate controller 1007 allocates different bit amounts to each encoded object data, each frame, and each encoded block based on the characteristics of the encoded unit and the available bits. Control the code bit rate. Further, the inverse quantization process and the inverse DCT transform process are executed in units 1008 and 1009 as part of the local decoder. Image data decoded by the local decoder is stored in the local decoded frame memory 1010 and used for motion detection processing. Reference numeral 1011 denotes a reference frame memory for storing a previous original frame for motion detection. Finally, the bit stream is output from the entropy encoding unit 1006 and sent to the image predictive decoding apparatus on the reception side or the reproduction side.

  FIG. 15 is a schematic diagram of an image for explaining an adaptive DCT transform region for intra prediction.

  FIG. 15 shows that in the DCT transform region, four 8 × 8 DCT transform blocks constitute a macro block. Here, B0 indicates the current block having a DCT transform coefficient of 8 × 8. B2 indicates the upper adjacent block that has already been decoded. B1 and B3 represent two adjacent blocks on the diagonal that have already been decoded. B4 indicates the immediately preceding block adjacent to the left side. It can be seen from FIG. 15 that a block having DCT transform coefficients can be predicted from a plurality of adjacent decoded blocks having 8 × 8 DCT transform coefficients.

  Note that it is always different from which block the current block was predicted. Accordingly, a determination based on the minimum bit usage rule is performed, and the determination is adaptively given to different blocks on the image predictive decoding apparatus side. The determination is notified to the image predictive decoding apparatus by the instruction bit. Here, the minimum bit usage rule is used to determine a prediction method among different prediction methods, and after each prediction method is applied, the amount of bits used to encode a block is counted. The As a result, the method that results in the smallest amount of bits used is selected as the prediction method to use.

  Note that DCT transform domain prediction can be performed after and also before quantization.

<Fifth Embodiment>
FIG. 16 is a block diagram showing a configuration of an image predictive coding apparatus according to the fifth embodiment of the present invention. The image predictive coding apparatus in FIG. 16 is characterized in that the DCT transform region prediction process is executed after the quantization process.

  In FIG. 16, first, block sampling is executed by the block sampling unit 1012 on the input image signal. In intra-frame coding, the sampled block image data is input to the DCT transform unit 1014 through the adder 1013 without being processed by the adder 1013. On the other hand, in the prediction frame coding, the adder 1013 subtracts the motion detection image data output from the motion detection and compensation unit 1025 from the sampled block image data, and the subtraction result image data is converted into the DCT conversion unit 1014. Output to. Then, after the DCT transform process is executed in the unit 1014, the quantization process is executed in the unit 1015.

  DCT transform domain prediction processing is performed in unit 1017, where 1018 is a block memory for storing previously decoded blocks for prediction. The adder 1016 subtracts the decoded adjacent block output from the DCT transform domain prediction unit 1017 from the current DCT transform block output from the quantization unit 1015. The determination of the encoded adjacent block is performed in the DCT transform domain prediction unit 1017. Finally, an entropy VLC encoding process is performed by the unit 1020 on the predicted DCT transform block, and the encoded bits are written into the bitstream.

  The adder 1019 restores the current DCT transform block by adding the previous adjacent block used for prediction to the prediction block. Next, the inverse quantization process and the inverse DCT transform process are performed on the restored DCT transform block in units 1021 and 1022, respectively. The block image data that is locally decoded and output from the inverse DCT transform unit 1022 is input to the adder 1023. The adder 1023 obtains reconstructed image data by adding the image data of the previous frame to the image data of the restored block, and stores it in the frame memory 1024. Motion detection and compensation processing is performed in unit 1025. A frame memory 1024 is used to store previous frames for motion detection and compensation processing.

<Sixth Embodiment>
FIG. 17 is a block diagram showing a configuration of an image predictive coding apparatus according to the sixth embodiment of the present invention. The image predictive coding apparatus in FIG. 17 is characterized in that the DCT transform region prediction process is executed before the quantization process. Block sampling processing is performed in the unit 1026 on the input image signal. Next, the adder 1027 performs subtraction for predictive frame encoding, and the image data as the subtraction result is passed through the DCT transform unit 1028, the adder 1029, and the quantization unit 1030, and the entropy VLC encoding unit 1034 and the inverse quantum. Is output to the conversion unit 1033.

  The block memory 1032 stores the image data of the previous block for the DCT transform region prediction process of the unit 1031. The image data of the current DCT transform block output from the DCT transform unit 1028 is subtracted by the adder 1029 from the previous DCT transform block selected by the DCT transform domain prediction unit 1031 according to the minimum bit usage rule. The image data of the DCT transform block as a subtraction result is quantized by the quantization unit 1030 and then output to the inverse quantization unit 1033 and the entropy VLC encoding unit 1034. The inverse quantization unit 1033 restores the input quantized DCT transform block image data by inverse quantization and outputs the restored image data to the adder 1035. The adder 1035 adds the image data of the previous DCT transform block from the DCT transform region prediction unit 1031 to the restored image data of the DCT transform block, and the image data of the block before the addition result is added to the block memory 1032. And output to the inverse DCT transform unit 1036.

  The inverse DCT conversion unit 1036 performs an inverse DCT conversion process on the image data of the block before being input from the adder 1035, and outputs the restored image data after the conversion process to the adder 1037. The adder 1037 adds the image data of the previous frame output from the motion detection and compensation unit 1025 to the restored image data output from the inverse DCT transform unit 1036, and the image data of the addition result is added to the frame memory 1038. And temporarily output to the motion detection and compensation unit 1025.

<B1. General description of mode decision>
FIG. 18 is a block diagram illustrating the configuration of the DCT transform region prediction circuits 1017 and 1031 in FIGS. 16 and 17.

In FIG. 18, reference numeral 1040 denotes a block memory for storing image data of a previous adjacent block for prediction. The current block to be processed is input to the unit 1041, and the unit 1041 subtracts the image data of the current block input from the previous adjacent DCT transform block stored in the block memory 1040 to obtain the next 4 blocks. Image data of a type of predicted DCT transform block is obtained.
(A) No-Pred block indicated by 1042;
(B) Up-Pred block indicated by 1043;
(C) a Left-Pred block indicated by 1044;
(D) An Other-Pred block indicated by 1045.

  Here, the above four types of blocks are shown using 2 bits. That is, for example, “00” indicates a No-Pred block, “01” indicates an Up-Pred block, “10” indicates a Left-Pred, and “11” indicates an Other-Pred block.

  The No-Pred block is the image data itself of the current DCT transform block when there is no prediction. The Up-Pred block indicates image data of a prediction block obtained when the block used for prediction is the DCT transform block B2 adjacent above. The Left-Pred block indicates image data of a prediction block obtained when the block used for prediction is a DCT transform block B4 adjacent on the left side. The Other-Pred block indicates the image data of the prediction block when the prediction is performed only on the DC coefficient. In the case of Other-Pred, there are two types of prediction methods. That is, Up-DC-Pred (1046) indicates the image data of the prediction block obtained when the prediction is performed only on the DC coefficient based on the DCT transform block B2 adjacent above. Left-DC-Pred (1047) indicates image data of a prediction block obtained when prediction is performed on only the DC coefficient based on the DCT transform block B4 adjacent on the left side. For these two cases, one more bit is needed for indication. For example, “0” indicates Up-DC-Pred (1046), and “1” indicates Left-DC-Pred (1047).

  Although prediction based on diagonally adjacent blocks B1 and B3 is possible, the prediction result is not as good as the result of prediction in the upper and left blocks, so it is not used in this embodiment.

  All predicted blocks are examined and checked by the actual entropy encoding process performed by unit 1048. The bits used for the different predicted blocks are compared in unit 1049. Finally, unit 1050 determines the predicted DCT transform block based on the minimum bit usage rule, and outputs the predicted DCT transform block along with the indicator bit. That is, the predicted DCT transform block having the smallest number of bits is selected.

<B2. Implementation of mode determination>
FIG. 19 is a schematic diagram of an image showing an example of an encoding method of DC / AC prediction in the DCT transform domain prediction circuit of FIG.

  In FIG. 19, a subset of the previously defined DC / AC predicted image data is illustrated for actual use. The current block 1101 is an 8 × 8 block on the upper left side of the current macro block, and the current block 1102 is an 8 × 8 block on the upper right side of the current macro block. A and B are 8 × 8 blocks adjacent to the current block 1101. The highlighted upper row and left column of the current block 1101 are predicted from the same location of adjacent blocks A and B, respectively. That is, the top row of the current block 1101 is predicted from the top row of the block A above it, and the left column of the current block 1101 is predicted from the left column of the left block B. In the same procedure, the current block 1102 is predicted from the block D above and the current block 1 on the left.

C (u, v) is a block to be encoded, E _i (u, v) is a prediction error in mode i, and each block of A (u, v) and / or B (u, v) It is assumed that it is obtained by subtracting the predicted value from. In actual implementation, only the next three most frequent modes described above in Section B1 are used.

(A) Mode 0: DC prediction only

[Equation 1]
E ₀ (0,0) = C (0,0) − (A (0,0) + B (0,0)) / 2
E ₀ (u, v) = C (u, v),
u ≠ 0; v ≠ 0; u = 0,..., 7; v = 0,.

(B) Mode 1: DC / AC prediction from the upper block

[Equation 2]
E ₁ (0, v) = C (0, v) −A (0, v), v = 0,.
E ₁ (u, v) = C (u, v),
u = 1,..., 7; v = 0,.

(C) Mode 2: DC / AC prediction from the left block [Equation 3]
E ₂ (u, 0) = C (u, 0) −B (u, 0), u = 0,.
E ₂ (u, v) = C (u, v),
u = 0,..., 7; v = 1,.

The mode is selected by calculating the sum of the absolute values of errors predicted for the four luminance signal blocks in the macroblock, SAD _modei , and selecting the mode having the minimum value.

[Equation 4]
SAD _modei
= Σ [E _i (0,0) + 32 · ΣE _i (u, 0) + 32 · ΣE _i (0, v)],
b u v
i = 0, ..., 2; b = 0, ..., 3; u, v = 1, ..., 7

  Mode determination can be performed both on a block basis and on a macroblock basis, depending on the difference in applications targeting different coding bit rates. The mode is encoded using the variable length code in Table 1 below.

  When performing DC / AC prediction after quantization, the preceding horizontal adjacent block or vertical adjacent block usually uses different quantization steps, so in order to perform DC / AC prediction accurately, Several types of weighting factors are required to scale the quantized DCT transform coefficients.

  Let QacA be the quantized DCT transform coefficient of block A (see FIG. 19) and QacB be the quantized DCT transform coefficient of block B (see FIG. 19). If QstepA is a quantization step used for quantization of block A, QstepB is a quantization step used for quantization of block A, and QstepC is a quantization step to be used for quantization of current block C. Thus, the scaling equation is:

[Equation 5]
Q'acA = (QacA × QstepA) / QstepC
[Equation 6]
Q′acB = (QacB × QstepB) / QstepC

  Here, Q′acA is a DCT transform coefficient from block A, and is used for prediction of the top row of current block C. Q'acB is a DCT transform coefficient from the block B and is used for prediction of the left column of the current block C.

<Seventh Embodiment>
FIG. 20 is a block diagram showing a configuration of an image predictive decoding apparatus according to the seventh embodiment of the present invention.

  In FIG. 20, the bitstream from the image predictive coding apparatus is input to the entropy VLD decoding unit 1051 and subjected to variable length decoding. The decoded image data is added to the image data of the previous adjacent DCT transform block from the DCT transform region prediction unit 1053 by the adder 1052, so that the image data of the DCT transform block is restored. Which block is a previously adjacent DCT transform block is informed by an indication bit taken from the bitstream and used in unit 1053 for prediction. Reference numeral 1054 denotes a block memory for storing adjacent DCT transform blocks used for prediction. The restored DCT transform block obtained from the adder 1052 is output to the inverse DCT transform unit 1055. The inverse DCT transform unit 1055 generates image data of the DCT transform coefficient restored by executing the inverse DCT transform process on the input DCT transform block, and outputs the image data to the adder 1056. The adder 1056 adds the restored image data from the inverse DCT transform unit 1055 with the image data of the previous frame from the motion detection and compensation unit 1057, thereby detecting the motion detected and compensated and decoded image. Generate and output data. The decoded image data is temporarily stored in a frame memory that stores image data of the previous frame for motion detection and compensation, and then output to the motion detection and compensation unit 1057. The motion detection and compensation unit 1057 performs motion detection and compensation processing on input image data.

  Further, the decoded image data output from the adder 1056 is restored to the original image data by an inverse restoration process corresponding to the process of the block sampling units 1012 and 1026 of FIGS.

  Further, reference numeral 1059 denotes an inverse quantization unit. When the DCT transform region prediction process is performed before the quantization process as shown in FIG. 17, the inverse quantization unit 1059 is inserted at the position 1059a in FIG. On the other hand, when the DCT transform region prediction process is performed after the quantization process as shown in FIG. 16, the inverse quantization unit 1059 is inserted at the position 1059b in FIG.

  FIG. 21 is a flowchart showing a DC / AC prediction decoding method in the image predictive decoding apparatus of FIG. That is, FIG. 21 illustrates details of decoding of a bitstream for acquiring a DC / AC prediction mode and reconstructing a DCT transform coefficient from adjacent DC / AC prediction values.

  First, in step 1059, the instruction bit is decoded from the input bit stream. In step 1060, the flag of the instruction bit is checked. If it is “0”, the image data of the upper block and the left block in step 1061. Then, the DC value is calculated from the average value and the process proceeds to step 1063. If “NO” in the step 1060, the process proceeds to a step 1062, and if the instruction flag checked in the step 1062 is “10”, the image data of the left column of the left block is extracted in a step 1063, and the process proceeds to the step 1065. If “NO” in the step 1062, the process proceeds to a step 1064, and if the display flag checked in the step 1064 is “11”, the image data of the uppermost row of the upper block is extracted in a step 1065, and the process proceeds to the step 1066. Finally, in step 1066, the DCT transform coefficient acquired or extracted in step 1061, 1063, or 1065 is added to the corresponding DCT transform coefficient of the current block.

  Further, modifications of the present embodiment group will be described below.

(A) In the block sampling units 1012 and 1026, the pixels of the two-dimensional array in the group of four blocks are composed of odd-numbered pixels in odd-numbered rows in the first block, and in the second block Consisting of even-numbered pixels in odd-numbered rows, consisting of odd-numbered fields in even-numbered rows in the third block, and even-numbered pixels in even-numbered rows in the fourth block, An interleaving process in which pixels are alternately inserted may be included.
(B) the prediction block is a block that has been stored in the block memory and has been previously restored, and is selected from blocks positioned adjacent to the encoded current block, and all the transforms in the block; A coefficient may be selected.
(C) The prediction block is a block that is stored in the block memory and has been previously restored, and is selected from blocks positioned adjacent to the encoded current block. It may be selected as a transform coefficient for the block.

(D) The prediction block is a block that is stored in the block memory and has been previously restored, and is selected from blocks positioned adjacent to the upper and left sides of the encoded current block. Only the transform coefficients in the top row and the leftmost column of the block may be used, and the remaining transform coefficients may be set to zero.
(E) The prediction block is a block that is stored in the block memory and has been previously restored, and is selected from blocks that are located in the vicinity of the encoded current block. It may be weighted with a function.
(F) The prediction block is a block that is stored in the block memory and has been previously restored, and is selected from blocks that are located in the vicinity of the encoded current block. A conversion operation may be performed.

(G) The prediction block may be a weighted average of a plurality of blocks that are stored in the block memory and previously restored and positioned near the encoded current block.
(H) When the original image data is restored by forming a two-dimensional array of pixels from a plurality of groups of four interleaved blocks based on the decoded image data, All odd-numbered pixels are determined from the first block, even-numbered pixels in the odd-numbered rows are determined from the second block, odd-numbered pixels in the even-numbered rows are determined from the third block, and even numbers The deinterleave process may be performed on the decoded image data so that the even-numbered pixels in the th row are obtained from the fourth block.

  As described above, according to this embodiment group according to the present invention, there is a great effect in removing or reducing the redundancy of the DCT transform region between adjacent blocks, and as a result, the number of used bits is reduced. Finally, the coding efficiency can be greatly improved. Referring to FIG. 18 as an example of a detailed image predictive encoder, the prediction process is preferably performed only by using the upper or left adjacent block.

  For sequences containing QCIF, 6.4% bits can be saved for higher bit rate coding and 20% bits can be saved for lower bit rate coding. Also, for example, about 10% of bits can be saved over other QCIF sequences such as test sequences such as Akiyo, Mother, and Daughter. In addition, more bit savings are possible for CIF and CCIR sequences.

  As described above, according to the second embodiment group of the present invention, it is possible to provide a new image predictive coding apparatus and image predictive decoding apparatus that increase the current coding efficiency. In this apparatus, no complicated means is required to increase the coding efficiency, and the circuit configuration can be formed very simply and easily.

<Third embodiment group>
The third embodiment group includes the eighth embodiment.

  In view of the problems of the prior art, the present inventor reduced image coding efficiency not only between two images or within blocks within one image, but also between blocks within an image. At the same time, it was considered to further improve the redundancy by making the scan pattern of the block appropriate.

  It has been found that the DCT transform coefficients in adjacent blocks even in the same location are often very similar. The above is true if the characteristics of the original image for the two blocks are very similar, or if the horizontal or vertical lines, diagonals and other image patterns contain the same. From the information theory perspective, the same information implies redundancy.

  Redundancy that exists across the block and in the DCT transform domain can be removed or reduced by the adaptive prediction of the previous block. This results in VLC entropy coding that can achieve higher coding efficiency for smaller entropy of the prediction error signal.

  At the same time, horizontal and vertical structures are known to result in significant DCT transform coefficients concentrating on the leftmost column and top row transform block. Therefore, the embodiment according to the present invention can solve the above-described problems in the coefficient scan by adapting the scan based on the prediction mode.

  Embodiments according to the present invention provide a method for adaptively predicting the DCT transform coefficient of a current block from other blocks, thereby removing redundancy between adjacent blocks. Prediction error information is further reduced by adapting the scan method to a prediction mode that reduces the entropy of the quantized DCT transform coefficients. As a result, the number of bits for encoding the DCT transform coefficient can be reduced.

  In order to solve this problem, a method for determining the prediction mode is obtained based on the actual bit rate generated by each prediction and scanning method.

  The embodiment according to the present invention provides a method for predicting the DCT transform coefficient of the current current block. Since the DCT transform tends to give the same value or the same DCT transform coefficient distribution to the image data of the same block, the current block usually retains a good correlation with the DCT transform coefficients in other adjacent blocks. Yes.

  The input image data is either an intra frame or a temporarily predicted frame. First, a DCT conversion process based on a normal block is performed on the input image data. Executed. After the DCT transform coefficients of the current block are obtained, the DCT transform domain prediction can be performed before or after quantization.

  The DCT transform coefficient in the current block can be predicted from the previous adjacent block located diagonally (diagonally) on the upper left side. They have already been decoded at that time, as shown in FIG. The predicted block generates a predicted error signal by subtracting one or more DCT coefficients of the previous adjacent block from the DCT coefficients at the same position in the current block.

  Prediction error signals from different prediction modes are quantized if the prediction is made before the quantization process. The quantized prediction error signal is scanned against a (sequence) of image data before entropy coding is performed. The block predicted based on the minimum bit usage rule, i.e., the prediction block with the least bit, is selected. The encoded data of this block is sent to the image predictive decoding apparatus together with the prediction mode to be used.

  The image predictive decoding apparatus decodes the predicted block using the used prediction mode and the encoded data of the block. After inverse entropy decoding for the encoded data for the block, the quantized prediction error is scanned inversely according to the scan mode used. If the quantization process is performed after the prediction process, the block will be dequantized. The reconstructed block can be obtained by adding the DCT transform coefficient in the previously decoded neighboring block indicated by the prediction mode to the current DCT transform coefficient. On the other hand, if the quantization process is performed before the prediction process, the reconstructed coefficient is inversely quantized. Finally, an inverse DCT transform process is applied to the reconstructed DCT transform coefficients for each block to obtain a decoded image.

  Embodiments according to the present invention provide an image predictive coding apparatus and an image predictive decoding apparatus that reduce redundancy existing in a DCT transform region beyond adjacent blocks.

<Eighth Embodiment>
FIG. 24 is a block diagram showing a configuration of an image predictive encoding device according to the eighth embodiment of the present invention. The image predictive coding apparatus of FIG. 24 is compared with the image predictive coding apparatus of the prior art of FIG.
(A) Adder 2035,
(B) H / V / Z scan unit 2036,
(C) Adder 2038,
It is characterized by comprising (d) a block memory 2039 and (e) a DCT transform domain prediction unit 2040 having quantization scaling.

  In intra-frame coding (intra-frame coding), a block sampling process is performed on the input image signal by the unit 2031 and then a DCT conversion process is directly performed by the unit 2033. Next, a quantization process is performed on the DCT transform coefficient output from the DCT transform unit 2033 in the unit 2034. On the other hand, in inter-frame coding or inter-frame coding (predictive frame coding), after the block sampling process of the unit 2031, the adder 2032 starts from the image data after the block sampling process, from the motion detection and compensation unit 2045. The output image data is subtracted to obtain prediction error data. Next, the prediction error data is output to the adder 2035 via the DCT transform unit 2033 that performs DCT transform processing and the quantization unit 2034 that performs quantization processing. The DCT transform coefficients are predicted by the DCT transform domain processing of the unit 2040, and the predicted DCT transform coefficients are input to the adder 2035. The adder 2035 subtracts the predicted DCT transform coefficient from the DCT transform domain prediction unit 2040 from the DCT transform coefficient from the quantization unit 2034, and obtains the DCT transform coefficient of the prediction error of the subtraction result as H / V / The data is output to the Z scan unit 2036 and the adder 2038. The H / V / Z scan unit 2036 performs a horizontal scan, a vertical scan, or a zigzag scan adaptively on the input DCT transform coefficient depending on the selected prediction mode. The DCT transform coefficients are output to the entropy VLC encoding unit 2037. Next, the entropy VLC encoding unit 2037 performs entropy VLC encoding processing on the input DCT transform coefficient, and transmits the encoded bitstream to the image predictive decoding apparatus on the reception side or the reproduction side.

  The adder 2038 adds the quantized DCT transform coefficient output from the adder 2035 and the predicted DCT transform coefficient from the DCT transform domain prediction unit 2040 to restore the restored quantized DCT transform coefficient. Get the data. The quantized DCT transform coefficient data is output to the block memory 2039 and the inverse quantization unit 2041.

  In the local decoder provided in the image predictive coding apparatus, the restored DCT transform coefficient data from the adder 2038 is temporarily stored in a block memory 2039 for storing data of one block for the next prediction. Is output to the DCT transform domain prediction unit 2040. The inverse quantization unit 2041 inversely quantizes the input quantized DCT transform coefficient and outputs it to the inverse DCT transform unit 2042, and then the inverse DCT transform unit 2042 converts the input restored DCT transform coefficient into the inverse DCT. A conversion process is executed to restore the image data of the current block and output it to the adder 2043.

  In interframe coding, an adder 2043 performs the motion detection and compensation by the motion detection and compensation unit 2045 and the reconstruction from the inverse DCT transform unit 2042 in order to generate locally decoded image data. The decoded image data is added to obtain locally decoded image data, which is stored in the frame memory 2044 of the local decoder. The configuration and processing of the adder 2043, the frame memory 2044, and the motion detection and compensation unit 2045 are the same as those of the prior art units 2009, 2010, and 2011 in FIG.

  Finally, the bit stream is output from the entropy coding unit 2037 and sent to the image predictive coding apparatus.

FIG. 25 is a block diagram showing a configuration of an image predictive decoding apparatus according to the eighth embodiment of the present invention. The image predictive decoding apparatus of FIG. 25 is compared with the image predictive decoding apparatus of the prior art of FIG.
(A) H / V / Z scan unit 2052,
(B) Adder 2053,
(C) DCT transform domain prediction unit 2055, and (d) block memory 2054,
It is provided with.

  In FIG. 25, the bit stream from the image predictive coding apparatus is decoded by the variable length decoder unit 2051. The decoded data is input to the H / V / Z reverse scan unit 2052 and scanned horizontally in the reverse direction, vertically in the reverse direction, or zigzag in the reverse direction depending on the scan mode. The data after the scan process is input to the adder 2053, and the adder 2053 adds the data after the inverse scan process and the prediction error data from the DCT conversion prediction unit 2055, thereby decoding the DCT transform coefficient decoded. Data is obtained and output to the inverse quantization unit 2056 and stored in the block memory 2054. Next, the inverse quantization unit 2056 performs inverse quantization on the input encoded DCT transform coefficient data to obtain inversely quantized DCT transform coefficient data, and outputs the obtained data to the inverse DCT transform unit 2057. The inverse DCT transform unit 2057 performs inverse DCT transform processing on the input DCT transform coefficient data, restores the original image data, and outputs it to the adder 2058. In interframe coding, an adder 2058 adds the image data from the inverse DCT transform unit 2057 to the prediction error data from the motion detection and compensation unit 2060 to obtain locally decoded image data. And output to an external device and stored in the frame memory 2059.

  Further, the decoded image data output from the adder 2058 is restored to the original image data by a reverse restoration process corresponding to the process of the block sampling unit 2031 in FIG.

  In the above embodiment, the quantization process is performed prior to the prediction process. In a modified example, the quantization process may be performed after the prediction process. In this case, in the local decoder and the image predictive decoding apparatus, the inverse quantization process is executed before the predicted value is added. All other details are the same as in the embodiment described above.

  FIG. 26 is a schematic diagram of an image showing a macroblock of a frame and a structure of the block obtained from block division and showing a block prediction method in the eighth embodiment. The enlarged view of FIG. 26 shows how the prediction data for the current block is encoded. Here, the block C (u, v) is obtained from the block A (u, v) adjacent on the upper side and the block B (u, v) adjacent in the left direction. Next, this embodiment will be described in more detail in the present invention.

<C1. Coefficient number used for prediction>
The number of coefficients used for prediction depends on the sequence of image data. The flag AC_Coeff is used to adaptively select the optimal number of coefficients used for each image. The flags are shown in Table 2 below, and are sent from the image predictive coding apparatus to the image predictive decoding apparatus as part of the side information. Table 2 shows the fixed length code and FLC for the flag AC_Coeff. Here, FLC (Fixed Length Coding) is a lossless encoding that assigns a fixed-length code word to represent all possible events.

  Here, ACn is A (0, n) or B (n, 0) depending on the mode used.

  In the special case of this embodiment, all AC coefficients in the top row and leftmost column are used for prediction. In this case, when both the image predictive encoding device and the image predictive decoding device agree with this default value, no flag is required.

<C2. Quantization step scaling>
When adjacent blocks are quantized using different quantization step sizes from the current block, the prediction of AC coefficients is not that efficient. Accordingly, the prediction method is modified so that the prediction data is scaled by the quantization step size ratio of the current current block and the quantization step ratio of the prediction data block. This definition is given in section C3. Is given using the equation in

<C3. Prediction mode>
The plurality of modes to be set are as follows.

(A) Mode 0: DC prediction from the upper block from the processing block (abbreviated as “upper DC mode”)

[Equation 7]
E ₀ (0,0) = C (0,0) −A (0,0),
E ₀ (u, v) = C (u, v)

(B) Mode 1: DC prediction from the left block from the processing block (abbreviated as “left DC mode”).

[Equation 8]
E ₁ (0,0) = C (0,0) −B (0,0),
E ₁ (u, v) = C (u, v)

(C) Mode 2; DC / AC prediction from the block above the processing block (abbreviated as “upper DC / AC mode”)

[Equation 9]
E ₂ (0,0) = C (0,0) −A (0,0),
E ₂ (0, v) = C (0, v) −A (0, v) · Q _A / Q _C ,
v = 1, 2,..., AC_Coeff,
E ₂ (u, v) = C (u, v)

(D) Mode 3: DC / AC prediction from the left block from the processing block (abbreviated as “left DC / AC mode”)

[Equation 10]
E ₃ (0,0) = C (0,0) −B (0,0),
E ₃ (u, 0) = C (u, 0) −B (u, 0) · Q _B / Q _C
u = 1, 2,..., AC_Coeff,
E ₃ (u, v) = C (u, v)

<C4. Adaptive Horizontal / Vertical / Zigzag Scan>
Given the four prediction modes as above, the efficiency of intraframe coding can be further improved by employing coefficient scanning.

  27, 28, and 29 are schematic diagrams of images for explaining the order of the horizontal scan, the vertical scan, and the horizontal scan used for the coefficient scan in the eighth embodiment, respectively. Here, these scans are collectively referred to as H / V / Z scans.

<C5. Determination of explicit mode>
In the determination of the explicit mode, the prediction mode is determined in the image predictive encoding device, and the determination information is obtained by using some encoded bit information in the bitstream. To the image predictive decoding apparatus.

  FIG. 30 is a flowchart showing the mode determination process used in the eighth embodiment.

  In FIG. 30, the DCT transform coefficient data of the current block is input to the unit 2062, and the unit 2062 subtracts the input DCT transform coefficient data of the current block from the DCT transform coefficient data of the adjacent block from the block memory 2061. Thus, the DCT conversion prediction process is executed. In unit 2062, section C3. The DCT transform prediction process is executed in the four modes described in the above. Next, in the H / V / Z scan unit 2063, a coefficient scan process is executed. Here, as shown in FIG. The corresponding scan processes described in the above are executed. Further, the DCT transform coefficient data after the scan processing is sent to the entropy encoding unit 2064, where variable length encoding processing is executed. Then, in unit 2065, all bits generated in the different modes are compared, and in unit 2066, the block of prediction mode DCT transform coefficients that produces the least number of bits is selected. The bits of these DCT transform coefficient data are sent as a bit stream from the unit 2066 to the image predictive decoding apparatus together with the value of the prediction mode. Note that the prediction mode is encoded using the fixed-length code shown in Table 3 below.

<C6. Implicit mode determination>
In the second embodiment of mode determination, the image predictive coding apparatus and the image predictive decoding apparatus share the same prediction mode determining function. Both the image predictive encoding device and the image predictive decoding device determine the directionality related to the determination of the prediction mode based on the DC coefficient value of the decoded block adjacent to the current block. That is, in the determination of the implicit mode, the determination of the implicit mode is performed in the image predictive coding apparatus and the image predictive decoding apparatus using some rules. Further, additional information data indicating mode determination is not sent from the image predictive coding apparatus to the image predictive decoding apparatus.

  FIG. 31 is a schematic diagram of an image showing the relationship of blocks in the implicit mode determination of the eighth embodiment. That is, FIG. 31 shows the positional relationship between each block and the current block that is the prediction target.

  In FIG. 31, block C is the current block to be processed that is currently predicted. Block A is a block above the current block C being predicted. Block C is a block located on the left side of current block C. The block C ′ is a block between the block A and the block B that are diagonal to the current block C.

  First, the direction of DC is determined. Using a separate determination method, it is determined whether the AC coefficient is also being predicted. To do this, the sum of the absolute differences of the prediction coefficients is compared with the absolute value of the non-prediction coefficients to determine which is smaller. One bit is used for this instruction to the image predictive decoding apparatus. The following formula is used to determine the directionality of DC prediction and whether or not the AC coefficient is predicted. Table 3 summarizes the four possible conclusions.

(A1) If [Equation 11]
(B (0,0) −C ′ (0,0) <C ′ (0,0) −A (0,0))
When,
[Equation 12]
E (0,0) = C (0,0) -A (0,0)
And
(A1) If

[Equation 13]
7 7
(ΣC (0, v) ≧ ΣC (0, v) −A (0, v))
v = 1 v = 1

When,

[Formula 14]
E (0, v)
= C (0, v) -A (0, v) · Q _A / Q _C , v = 1,.

(A2) If the above equation 13 is not satisfied,
[Equation 15]
E (0, v) = C (0, v)
It is.

(A2) If the above equation 11 does not hold,
[Equation 16]
E (0,0) = C (0,0) -B (0,0)
And
(B1) If

[Equation 17]
7 7
(ΣC (u, 0) ≧ ΣC (u, 0) −B (u, 0))
v = 1 v = 1

When,

[Equation 18]
E (u, 0)
= C (u, 0) −B (u, 0) · Q _B / Q _C , v = 1,.

(B2) If Equation 17 is not satisfied,
[Equation 19]
E (u, 0) = C (u, 0)
It is.

In addition, for all other coefficients,
[Equation 20]
E (u, v) = C (u, v)
It is.

  In the above eighth embodiment, the DCT transform coefficient prediction process is performed on the quantized transform coefficient data by the unit 2040, but the present invention is not limited to this, and is the same as the sixth embodiment of FIG. Alternatively, it may be performed on transform coefficient data that is not quantized. In this case, in the corresponding image predictive decoding apparatus, in FIG. 25, the inverse quantization unit 2056 is moved to the inverse scan unit 2052 and the adder 2053 and inserted.

  Hereinafter, modifications of the eighth embodiment will be described.

(A) The block sampling unit 2031 is configured such that the pixels of the two-dimensional array in the group of four blocks are composed of odd-numbered pixels in odd-numbered rows in the first block and odd-numbered pixels in the second block. The pixels are alternated to consist of even-numbered pixels in the row, consisting of odd-numbered pixels in the even-numbered rows in the third block, and even-numbered pixels in the even-numbered rows in the fourth block. Interleaving processing may be included.
(B) the prediction block is a previously restored block stored in the block memory, selected from blocks that are located adjacent to the encoded current block, and all coefficients in the block; Data may be selected.
(C) The prediction block is a block that is stored in the block memory and has been previously restored, and is selected from blocks positioned adjacent to the encoded current block. It may be selected as the coefficient data of the block.

(D) The prediction block is a block that is stored in the block memory and has been previously restored, and is selected from blocks positioned adjacent to the upper and left sides of the encoded current block. Only the coefficient data in the top row and the leftmost column of the block may be used, and the remaining coefficient data may be set to zero.
(E) the prediction block is stored in the block memory and selected according to the above criteria from previously restored blocks;
The image predictive coding apparatus and the image predictive decoding apparatus determine to use only a subset including one or more coefficient data from the top row or the leftmost column of the block by performing communication. Also good.
(F) the predicted block is selected according to the above criteria from previously restored blocks stored in the block memory;
The image predictive coding apparatus determines to use only the subset including one or more coefficient data from the top row or the leftmost column of the block, and indicates the determined subset and the number of coefficient data You may notify an image predictive decoding apparatus by inserting a flag into the data transmitted to an image predictive decoding apparatus periodically.

(G) the predicted block is selected according to the above criteria from previously restored blocks stored in the block memory;
The coefficient data of each block may be multiplied by a ratio equal to the ratio between the quantization step size of the current block to be encoded and the quantization step size of the prediction block.
(H) the predicted block is selected according to the above criteria from previously restored blocks stored in the block memory;
The coefficient data for each block may be weighted with a different weighting function.
(I) the predicted block is selected according to the above criteria from previously restored blocks stored in the block memory;
A predetermined conversion operation may be performed on the coefficient data of each block. (J) The predicted block may be obtained as a weighted average value of previously restored blocks stored in the block memory located adjacent to the current block to be encoded.

(K) The scanning method is
(I) a horizontal scan in which coefficient data is scanned from left to right, starting from the top row and ending at the bottom row, row by row;
(Ii) a vertical scan in which coefficient data is scanned from the top row to the bottom row, starting from the leftmost column and ending at the rightmost column, column by column;
(Iii) The coefficient data may include at least one scanning method of zigzag scanning that is scanned in a diagonal direction from the leftmost coefficient data in the uppermost row to the rightmost coefficient data in the lowermost row.

(L) The prediction block is selected according to the above criteria from previously restored blocks stored in the block memory,
The prediction mode of the prediction block is
(I) a first mode in which only the uppermost and leftmost coefficient data representing the average value of the block, which is called a DC coefficient, from a block located above the current block to be processed is used for prediction;
(Ii) a second mode in which only DC coefficients from a block located on the left side of the current block to be processed are used for prediction;
(Iii) a third mode in which DC coefficients and zero or more AC coefficients including high-frequency components from the top row of the block located above the current block to be processed are used for prediction;
(Iv) a fourth mode in which a DC coefficient and zero or more AC coefficients including high-frequency components from the leftmost column of the block located on the left side of the current block to be processed are used for prediction;
Including at least one prediction mode of
The prediction error coefficient data may be scanned by a zigzag scanning method.

(M) the predicted block is selected according to the above criteria from previously restored blocks stored in the block memory;
The prediction error coefficient data is scanned according to one of the scanning methods described above,
The prediction mode for predicting the coefficient data of the prediction error is
(I) a first mode in which only DC coefficients in a block located above the current block to be processed are used for prediction, and the coefficient data of the prediction error is scanned by zigzag scanning; ,
(Ii) a second mode in which only DC coefficients in a block located on the left side of the current block to be processed are used for prediction, and the coefficient data of the prediction error is scanned by zigzag scanning; ,
(Iii) The DC coefficient in the top row of the block located on the upper side of the current block to be processed and zero or more AC coefficients including high frequency components are used for prediction, and the coefficient data of the prediction error is A third mode in which scan processing is executed in horizontal scan,
(Iv) The DC coefficient in the leftmost column of the block located on the left side of the current block to be processed and zero or more AC coefficients including high frequency components are used for prediction, and the coefficient data of the prediction error is On the other hand, a fourth mode in which scanning processing is executed in vertical scanning;
May be included.

(N) When the original image data is restored by forming a two-dimensional array of pixels from a plurality of groups of four interleaved blocks based on the decoded image data, the odd-numbered rows All the odd-numbered pixels in the odd-numbered row are obtained from the first block, the even-numbered pixels in the odd-numbered row are obtained from the second block, the odd-numbered pixels in the even-numbered row are obtained from the third block, The deinterleaving process may be performed on the decoded image data so that the even-numbered pixels in the even-numbered rows are obtained from the fourth block.
(O) The image predictive encoding device and the image predictive decoding device may determine the prediction mode using the same predetermined rule.
(P) The image predictive coding apparatus and the image predictive decoding apparatus may determine the scan method using the same predetermined rule.

  As described above, according to the third embodiment group of the present invention, it is very effective to reduce or eliminate redundancy in the DCT transform region beyond adjacent blocks. As a result, the coding efficiency can be greatly improved. This is also useful as a tool in new video compression algorithms.

  In the above embodiment, the image predictive encoding device and the image predictive decoding device have been described. However, the present invention is not limited to this, and each unit, each unit, and other components in the image predictive encoding device are each An image predictive encoding method including a step replaced with a step, or an image predictive decoding method including a step in which components such as each unit and each unit in the image predictive decoding apparatus are replaced with each step, respectively. It may be. In this case, for example, each step of the image predictive encoding method and / or the image predictive decode method is stored as a program in a storage device, and a controller such as a microprocessor unit (MPU) or a central processing unit (CPU) By executing the program, the image predictive encoding process and / or the image predictive decoding process is executed.

  Further, the present invention may be a recording medium on which a program including each step in the image predictive encoding method and / or the image predictive decoding method is recorded. The recording medium has a disk shape in which the recording area is divided into sector shapes, or the recording area is spirally divided into blocks, for example, an optical disk such as a CD-ROM or DVD, or a magneto-optical disk, or A magnetic recording disk such as a floppy disk.

It is a block diagram which shows the structure of the image predictive coding apparatus which is 1st Embodiment which concerns on this invention. It is a schematic diagram at the time of dividing | segmenting the input image input into the image predictive coding apparatus of FIG. 1 into an 8 * 8 block. It is a schematic diagram at the time of dividing | segmenting the input image input into the image predictive coding apparatus of FIG. 1 into a triangular area | region. It is a block diagram which shows the structure of the 1st Example of the prediction signal generator used for the image predictive coding apparatus of FIG. It is a block diagram which shows the structure of the 2nd Example of the prediction signal generator used for the image predictive coding apparatus of FIG. It is a block diagram which shows the structure of the 3rd Example of the prediction signal generator used for the image predictive coding apparatus of FIG. It is a block diagram which shows the structure of the 4th Example of the prediction signal generator used for the image predictive coding apparatus of FIG. It is a block diagram which shows the structure of the image predictive coding apparatus which is the 2nd Embodiment which concerns on this invention. It is a schematic diagram which shows an example of the input image which is an input image input into the image predictive coding apparatus of FIG.1 and FIG.8, and has a pixel which is significant. It is a schematic diagram which shows an example of the input image which is an input image input into the image predictive coding apparatus of FIG.1 and FIG.8, and has a pixel which is significant. It is a schematic diagram which shows an example of the input image which is an input image input into the image predictive coding apparatus of FIG.1 and FIG.8, and has an insignificant pixel. It is a block diagram which shows the structure of the image predictive decoding apparatus which is the 3rd Embodiment which concerns on this invention. It is a block diagram which shows the structure of the image predictive decoding apparatus which is 4th Embodiment which concerns on this invention. It is a block diagram which shows the structure of the image predictive coding apparatus of a prior art. It is a schematic diagram of an image for explaining an adaptive DCT transform region for intra prediction. It is a block diagram which shows the structure of the image predictive coding apparatus which is the 5th Embodiment which concerns on this invention. It is a block diagram which shows the structure of the image predictive coding apparatus which is the 6th Embodiment concerning this invention. FIG. 18 is a block diagram illustrating a configuration of a DCT transform region prediction circuit in FIGS. 16 and 17. FIG. 19 is a schematic diagram of an image showing an example of a DC / AC prediction encoding method in the DCT transform region prediction circuit of FIG. 18. It is a block diagram which shows the structure of the image predictive decoding apparatus which is the 7th Embodiment which concerns on this invention. [Fig. 21] Fig. 21 is a flowchart illustrating a DC / AC prediction decoding method in the image predictive decoding device in Fig. 20. It is a block diagram which shows the structure of the image predictive coding apparatus of a prior art. It is a block diagram which shows the structure of the image prediction decoding apparatus of a prior art. It is a block diagram which shows the structure of the image predictive coding apparatus which is 8th Embodiment which concerns on this invention. It is a block diagram which shows the structure of the image predictive decoding apparatus which is 8th Embodiment which concerns on this invention. It is a schematic diagram of the image which shows the structure of the macroblock and block of a frame in 8th Embodiment, and shows the block prediction method. It is a schematic diagram of the image for demonstrating the order of the horizontal scan used for the coefficient scan in 8th Embodiment. It is a schematic diagram of the image for demonstrating the order of the vertical scan used for the coefficient scan in 8th Embodiment. It is a schematic diagram of the image for demonstrating the order of the zigzag scan used for the coefficient scan in 8th Embodiment. It is a flowchart which shows the mode determination process used for 8th Embodiment. It is a schematic diagram of the image which shows the relationship of the block in the implicit mode determination of 8th Embodiment.

Explanation of symbols

101: Input terminal,
102 ... the first adder,
103 ... Encoder,
104 ... DCT converter,
105 ... Quantizer,
106 ... output terminal,
107: Decoder,
108: Inverse quantizer,
109 ... Inverse DCT converter,
110 ... second adder,
111 ... Line memory,
112 ... Predictive signal generator,
200 ... Block,
300, 301 ... triangle,
401, 402 ... generator,
403, 404 ... image data,
500 ... adder,
601 602 603... Error calculator,
604 ... comparator,
605 ... switch,
700 ... a motion detector,
701: Motion compensator,
702: Frame memory,
703 ... Optimal mode selector,
800, 804 ... shape curve,
802, 805, 810 ... small areas to be processed,
808 ... curve,
901 ... Input terminal,
902 ... Data analyzer,
903 ... Decoder,
904 ... Inverse quantizer,
905 ... Inverse DCT converter,
906 ... adder,
907 ... output terminal,
908 ... Controller,
909 ... motion compensator,
910 ... Predictive signal generator,
911 ... line memory,
912 ... Frame memory,
913 ... Switch,
922 ... Predictive signal generator,
923 ... a motion compensator,
990 ... Shape decoder,
1001 ... Block sampling unit,
1002 ... Motion detection unit,
1003 ... Compensation unit,
1004 ... DCT converter,
1005 ... Quantization unit,
1006 ... entropy coding unit,
1007 ... Rate controller,
1008, 1009 ... unit,
1010 ... Local decoding frame memory,
1011 ... Reference frame memory,
1012: Block sampling unit,
1013: Adder,
1014 ... DCT conversion unit,
1015: Quantization unit,
1016: Adder,
1017 ... DCT transform domain prediction unit,
1018 Block memory,
1019: Adder,
1020 ... Entropy VLC encoding unit,
1021 ... Inverse quantization unit,
1022 ... Inverse DCT conversion unit,
1023 ... adder,
1024 ... frame memory,
1025 ... Motion detection and compensation unit,
1026: Block sampling unit,
1027 ... Adder,
1028 ... DCT conversion unit,
1029: Adder,
1030: Quantization unit,
1031 ... DCT transform domain prediction unit,
1032: Block memory,
1033: Inverse quantization unit,
1034 ... Entropy VLC encoding unit,
1035: Adder,
1036: Inverse DCT conversion unit,
1037 ... adder,
1038: Frame memory,
1040: Block memory,
1041 ... Unit,
1042 ... No-Pred block,
1043 ... Up-Pred block,
1044 ... Left-Pred block,
1045 ... Other-Pred block,
1048, 1049, 1050 ... unit,
1051 ... Entropy VLD decoding unit,
1052 ... Adder,
1053 ... DCT transform domain prediction unit,
1054 ... block memory,
1055 ... Inverse DCT conversion unit,
1056 ... adder,
1057 ... Motion detection and compensation unit,
1059 ... Inverse quantization unit,
1101, 1102 ... current block,
2031 ... Block sampling unit,
2032 ... adder,
2033 ... DCT conversion unit,
2034 ... Quantization unit,
2035 ... adder,
2036 ... H / V / Z scan unit,
2037 ... Entropy VLC encoding unit,
2038 ... adder,
2039: Block memory,
2040 ... DCT transform domain prediction unit,
2041 ... Inverse quantization unit,
2042 ... Inverse DCT conversion unit,
2043 ... adder,
2044 ... Frame memory,
2045 ... Motion detection and compensation unit,
2051 ... Variable length decoder unit,
2052 ... H / V / Z scan unit,
2053 ... adder,
2054: Block memory,
2055 ... DCT transform domain prediction unit,
2056 ... Inverse quantization unit,
2057 ... Inverse DCT conversion unit,
2058 ... adder,
2059 ... frame memory,
2060 ... Motion detection and compensation unit,
2061 ... Block memory,
2062 ... Subtraction unit,
2063 ... H / V / Z scan unit,
2064 ... entropy coding unit,
2065 ... comparison unit,
2066 ... selection unit,
B0: Current block,
B1 ... upper left block,
B2 ... the upper block,
B3 ... upper right block,
B4 ... Left block.

Claims (6)

A conversion step of converting the blocked image data into a conversion coefficient of a two-dimensional sequence;
A quantization step of quantizing the transformed two-dimensional sequence of transform coefficients to obtain a quantized transform coefficient;
DC coefficient of the left block and the diagonal block for the upper block located above the current block, the left block located on the left side of the current block, and the diagonal block located between the upper block and the left block difference and determines the difference between DC coefficients of the diagonal blocks and the upper block, adaptively prediction block from one of the upper block or the left block adjacent to the current block based on a comparison of these differences in A predictive block selection step of selecting
From the DC coefficient adaptively selected the prediction block from one of the upper block or the left block, the DC coefficient prediction step for predicting the DC coefficient of the current block,
And a AC coefficient prediction step of predicting the AC coefficients from the AC coefficients the current block selected the prediction block for the DC coefficient prediction,
In the AC coefficient prediction step, the AC coefficient of the prediction block is scaled by the ratio of the quantization step size of the current block and the quantization step size of the prediction block, and the AC coefficient of the scaled prediction block is calculated. An image predictive coding method characterized by predicting the AC coefficient of the current block .   2. The image predictive encoding method according to claim 1, wherein bit information indicating whether or not the AC coefficient prediction is executed at the time of encoding is output as an encoded bit stream. The AC coefficient prediction of the current block is
When the left block is selected as the prediction block in the DC coefficient prediction of the current block, AC coefficient prediction is performed using the quantized AC coefficient in the first column of the left block as a prediction value,
When the upper block is selected as the prediction block in the DC coefficient prediction of the current block, AC coefficient prediction is performed using the quantized AC coefficient in the first row of the upper block as a predicted value. The image predictive encoding method according to claim 1. An image predictive encoding device that encodes image data,
A converter for converting the blocked image data into a conversion coefficient of a two-dimensional sequence;
A quantizer that quantizes the transform coefficients of the two-dimensional sequence to obtain quantized transform coefficients;
DC coefficient of the left block and the diagonal block for the upper block located above the current block, the left block located on the left side of the current block, and the diagonal block located between the upper block and the left block difference and determines the difference between DC coefficients of the diagonal blocks and the upper block, adaptively prediction block from one of the upper block or the left block adjacent to the current block based on a comparison of these differences in prediction block selecting means for selecting,
From the DC coefficient adaptively selected the prediction block from one of the upper block or the left block, the DC coefficient prediction means for predicting a DC coefficient of the current block,
And a AC coefficient predicting means for predicting AC coefficients of the current block from the AC coefficients of the selected the prediction block for the DC coefficient prediction,
The AC coefficient prediction means performs scaling on the AC coefficient of the prediction block according to a ratio of the quantization step size of the current block and the quantization step size of the prediction block, and calculates the AC coefficient of the scaled prediction block. An image predictive coding apparatus for predicting an AC coefficient of the current block .   5. The image predictive coding apparatus according to claim 4, wherein bit information indicating whether or not the AC coefficient prediction is performed at the time of encoding is output as an encoded bit stream. The AC coefficient prediction of the current block is
When the left block is selected as the prediction block in the DC coefficient prediction of the current block, AC coefficient prediction is performed using the quantized AC coefficient in the first column of the left block as a prediction value,
When the upper block is selected as a prediction block in the DC coefficient prediction of the current block, AC coefficient prediction is performed using the quantized AC coefficient in the first row of the upper block as a predicted value. The image predictive coding apparatus according to claim 4.

Images