JP2010010768A - Image encoding apparatus and image encoding method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce deterioration in visual quality without increasing a code amount in encoding using intra-prediction. <P>SOLUTION: An image encoding apparatus 10 for encoding an input image based on an intra-prediction system includes: an intra-prediction section 102 for selecting an intra-prediction system having the highest prediction accuracy from among a plurality of intra-prediction systems for each block consisting of a plurality of pixels obtained by dividing the input image, and generating a prediction block consisting of a plurality of pixels each having a pixel value predicted in accordance with the selected intra-prediction system; a subtracting section 103 for calculating a residual block as a difference between the prediction block generated in the intra-prediction section 102 and the block corresponding to the prediction block; and an entropy encoding section 106 for encoding the residual block calculated in the subtracting section. An intra-prediction system for generating a prediction block in which the pixel value of an included pixel is a predetermined value is included in the plurality of intra-prediction systems. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image encoding device and an image encoding method, and more particularly to an image encoding device and an image encoding method that perform more efficient encoding using intra prediction.

  In recent years, digitalization of AV (Audio Visual) information has progressed, and devices capable of digitizing and handling video signals are becoming widespread. Here, since the video signal has an enormous amount of information, encoding is generally performed while reducing the amount of information in consideration of recording capacity or transmission efficiency. As a video signal encoding technique, H.264 is used. An international standard called H.264 standard is established. According to experimental results, H.C. The H.264 standard provides approximately twice the compression efficiency compared to other existing coding technologies. Therefore, H.H. The H.264 standard is an encoding method that is very suitable for recorder recording, Internet streaming, and the like.

  H. In the H.264 standard, intra prediction for performing prediction within a frame is introduced. Of the intra prediction schemes, in 4 × 4 block intra prediction, a 16 × 16 macroblock is divided into four 4 × 4 blocks in the vertical and horizontal directions, and encoding is performed for each of the divided blocks.

  FIG. 9 is a diagram illustrating the position of an intra prediction block (4 × 4 pixels) in a macroblock (16 × 16 pixels). In the figure, in the 4 × 4 block intra prediction method, the upper left 8 × 8 block (block 0) in the macroblock is encoded, and then the upper right 8 × 8 block (block 1) is encoded. Next, encoding is performed in the order of the lower left 8 × 8 block (block 2) and the lower right 8 × 8 block (block 3).

  Hereinafter, a technique for performing highly efficient encoding by determining an optimum prediction mode from among a plurality of prediction modes will be described (see Patent Document 1). First, the luminance signal will be described.

  FIG. 10 is a diagram illustrating nine intra prediction modes for luminance signals. There are nine types (mode 0 to mode 8) corresponding to the prediction directions in the intra prediction modes that are candidates for each 4 × 4 block, and one of these is selected for encoding.

  In FIG. 10, mode 0 is defined as “vertical prediction”. In mode 0, a difference is obtained by subtracting the luminance of encoded pixels A to D (peripheral pixels) adjacent to the upper part of the 4 × 4 block to be predicted from the luminances of the pixels vertically below the respective pixels. The absolute value sum of luminance is calculated by obtaining the absolute value of. That is, the luminance of the pixel A is subtracted from the luminance of the four pixels below the pixel A, and the luminance of the pixel B is subtracted from the luminance of the four pixels below the pixel B. Similarly, the luminances of the pixels C and D are subtracted from the luminances of the four pixels below the pixels C and D, respectively. The sum of absolute difference values of luminance obtained as a result of the subtraction in this way is the prediction error in prediction mode 0 of the 4 × 4 block to be predicted.

  Next, mode 1 is defined as “horizontal prediction”. In mode 1, the absolute value of the difference is obtained by subtracting the luminance of the encoded pixels I to L (peripheral pixels) adjacent to the left part of the prediction target block from the luminance of the horizontal right pixel of each pixel. Is calculated to calculate the absolute difference value sum of luminance. The sum of absolute difference values of luminance obtained as a result of this subtraction is the prediction error in prediction mode 1 of the prediction target block.

  Next, mode 2 is defined as “DC prediction”. In mode 2, the average value of the luminances of the encoded pixels A to D and the pixels I to L adjacent to the prediction target block is subtracted from the luminance of each pixel in the prediction target block, and the absolute value of the difference is calculated. By calculating, the absolute difference value sum of luminance is calculated. The sum of absolute difference values of luminance obtained as a result of this subtraction is the prediction error in prediction mode 2 of the prediction target block.

  Thereafter, the same processing is performed for the modes 3 to 8, and the prediction error in each mode is calculated. After calculating the prediction error, the mode with the smallest prediction error among the nine modes is determined as the optimum prediction mode of the prediction target block.

  Next, the color signal will be described.

  FIG. 11 is a diagram illustrating four intra prediction modes for color signals. There are four types (mode 0 to mode 8) corresponding to the prediction directions in the intra prediction modes that are candidates for each 4 × 4 block, and one of them is selected for encoding. Since the calculation of the prediction error for the color signal is the same as the calculation of the prediction error for the luminance signal described above, description thereof is omitted here.

After performing intra prediction in the optimal prediction mode determined as described above, a difference calculation is performed between the prediction target block and the prediction block formed according to the optimal prediction mode. A compression bit stream is generated by performing transformation processing, quantization processing, and entropy coding processing on the difference signal obtained by the difference calculation. Intra prediction is an effective method that can greatly reduce the amount of information of a block to be encoded and can improve compression performance.
JP 2007-189276 A

  However, the above-described conventional technique has a problem in that when the correlation between the encoding target block and the surrounding pixels used for prediction is low, the restored image obtained by restoring the encoded image is greatly degraded. Below, a subject is demonstrated in detail using FIGS.

  FIG. 12 is a diagram illustrating an example of an encoding target image and an encoding target macroblock. Hereinafter, distortion that occurs when a block including a part of an object is encoded will be described. In the following description, the color difference block is targeted. In the figure, a macro block 400 and a macro block 401 are located next to each other on the screen. In the following description, it is assumed that the macro block is composed of 8 × 8 pixels.

  FIG. 13 is a diagram illustrating pixel values of pixels included in an encoding target macroblock and peripheral pixels. It is assumed that the blocks 410, 411, 412, 413, 414, 415, 416 and 417 are flat blocks and all the pixel values in the block are 10. Blocks 410, 411, 412 and 413 belong to the macroblock 400, and blocks 414, 415, 416 and 417 belong to the macroblock 401. The macro block 400 is located next to the macro block 401 on the screen.

  The peripheral pixels 420 and 421 are pixels used for intra prediction. The peripheral pixels 420 and 421 are a part of the object (“house” in the example of FIG. 12), and the pixel values of the peripheral pixels 420 are all 100 and the pixel values of the peripheral pixels 421 are all 150.

  Here, if the horizontal prediction mode is determined as the optimum prediction mode for the block 410, the pixel values of the prediction blocks are all 100. Therefore, the pixel values of the residual blocks obtained by performing the difference calculation between the block 410 and the prediction block are all −90. The other blocks 411, 412 and 413 belonging to the macro block 400 also have the same result as the block 410.

  In intra prediction encoding, an orthogonal transform process is performed on a residual signal indicating a residual block, and transform coefficients obtained by the orthogonal transform process are quantized. When restoring the encoding target image, the quantized transform coefficient is inversely quantized, and the restored transform coefficient (hereinafter referred to as the restored transform coefficient) is subjected to inverse orthogonal transform to restore the residual signal. be able to. Then, the restored residual block (reconstructed residual block) indicated by the restored residual signal (hereinafter referred to as the restored residual signal) is added to the prediction block, thereby restoring the restored block (restored block). Can be obtained.

  However, when the quantization coefficient is greater than 90 with respect to the transform coefficient obtained by performing orthogonal transform processing on the residual signal, the quantized transform coefficient becomes zero. In this case, the restored transform coefficient after inverse quantization is also 0, and the restored residual signal obtained by performing inverse orthogonal transform on the restored transform coefficient is also 0. That is, all the pixel values in the restored residual block are 0. Subsequently, a restored block is obtained by performing an addition operation on the restored residual block and the prediction block.

  FIG. 14 is a diagram illustrating pixel values of a restored macroblock obtained by restoring a macroblock encoded by the conventional technique. As shown in FIG. 14, the pixel values in the restoration macroblock 430 are all 100. This is because the signal of the prediction block appears as it is because all the pixel values of the restored residual blocks of the respective blocks 410, 411, 412 and 413 are 0.

  FIG. 15 is a diagram illustrating a restored image obtained by restoring an image encoded by the conventional technique. As shown in FIG. 15, a signal component that did not exist in the original macroblock 400 newly appears in the restored macroblock 430 by performing the encoding process. In this case, a part of the object included in the peripheral pixels extends to the entire encoding target block, resulting in a large visual deterioration.

  Next, a similar verification is performed for block 414. As shown in FIG. 14, it is assumed that the block 414 is a flat block and all the pixel values in the block are 10. The peripheral pixels 422 and 423 are pixels used for intra prediction when the block 414 is encoded. The peripheral pixels also include the pixels of the restored macroblock 430 (peripheral pixels 422), and as described above, a part of the object is diffused in the macroblock 400 by intra prediction.

  Here, assuming that the horizontal prediction mode is selected as the optimum prediction mode for the block 414, the pixel values of the prediction blocks are all 100. Therefore, the pixel values of the residual blocks obtained by calculating the difference between the block 414 and the prediction block are all −90. The other blocks 415, 416 and 417 belonging to the macro block 401 have the same result as the block 414.

  Similar to the processing to block 410 described above, when the quantization coefficient is larger than 90 with respect to the transform coefficient obtained by performing orthogonal transform processing on the residual signal, the transform coefficient after quantization becomes zero. In this case, the restored transform coefficient after inverse quantization and the restored residual signal are also zero. That is, all the pixel values in the restored residual block are 0. Subsequently, a restored block is obtained by performing an addition operation on the restored residual block and the prediction block.

  FIG. 16 is a diagram illustrating pixel values of a restored macroblock obtained by restoring a macroblock encoded by the conventional technique. As shown in FIG. 16, the pixel values in the restoration macroblock 431 are all 100. This is because the signal of the prediction block appears as it is because all the pixel values of the restored residual blocks of the respective blocks 414, 415, 416 and 417 are 0.

  FIG. 17 is a diagram illustrating a restored image obtained by restoring an image encoded by the conventional technique. As shown in FIG. 17, a signal component that did not exist in the original macroblock 401 newly appears in the restored macroblock 431 by performing the encoding process. In particular, with respect to the macro block 401, a part of the object existing at a position separated by one macro block is propagated via the macro block 400, and the visual deterioration is very large.

  As described above, in the conventional encoding method using intra prediction, when the correlation between the encoding target block and the surrounding pixels used for prediction is low, the restored image obtained by restoring the encoded image is large. There is a problem that deterioration occurs.

  As a solution to this problem, a method of performing quantization and encoding using a quantizer with a small quantization coefficient is conceivable. However, when a quantizer with a small quantization coefficient is used, there arises a problem that the code amount increases.

  Therefore, the present invention has been made in view of the above problems, and provides an image encoding device that reduces visual image quality degradation without increasing the amount of code in encoding using intra prediction. With the goal.

  In order to achieve the above object, an image encoding device of the present invention is an image encoding device that encodes an input image based on an intra prediction method, and is a block composed of a plurality of pixels into which the input image is divided. For each, an intra prediction method having the highest prediction accuracy is selected from a plurality of intra prediction methods, and a prediction block including a plurality of pixels each having a pixel value predicted according to the selected intra prediction method is generated. A prediction unit; a subtraction unit that calculates a residual block that is a difference between the prediction block generated by the intra prediction unit and the block corresponding to the prediction block; and a residual block calculated by the subtraction unit. Encoding means for encoding, and the plurality of intra prediction methods include a prediction block in which pixel values of pixels included are predetermined values. It includes intra prediction method for generating a click. For example, the predetermined value may be zero.

  Accordingly, by providing a new intra prediction method, it is possible to reduce visual image quality degradation without increasing the code amount. This is because the new intra prediction method generates a prediction block composed of pixels having a predetermined value and is not affected by the pixel values of surrounding pixels referred to in the intra prediction method. Therefore, when the correlation between the surrounding pixels and the encoding target block is bad, the degradation of image quality can be reduced when a new intra prediction method is selected.

  Further, the intra prediction means includes a prediction block generation unit that generates a prediction block corresponding to each of the plurality of intra prediction schemes, and absolute differences between a plurality of pixels included in the prediction block generated by the prediction block generation unit A prediction error calculation unit that calculates a prediction error that is a sum of values for each intra prediction method and a selection unit that selects a prediction block generated according to the intra prediction method that minimizes the prediction error may be included. .

  As a result, the intra prediction method when the prediction error is minimized can be selected as the optimal intra prediction method, and encoding with the most accurate intra prediction can be executed.

  The intra prediction unit generates the prediction block for each color block including a plurality of pixels indicating the value of the color component of the input image, and the subtraction unit generates the prediction block generated by the intra prediction unit. A residual block that is a difference from the color block corresponding to the prediction block may be calculated.

  Thereby, it can utilize only for the value of a color component. When the process is performed on the luminance component, the image quality may be deteriorated depending on the above-described predetermined value. For example, there is a case where the target block after restoration becomes completely black by setting a predetermined value to 0. By targeting only the color components, it is possible to prevent deterioration in image quality in these cases.

  Note that the present invention can be realized not only as an image encoding device but also as a method in which processing means constituting the image encoding device is used as a step. Moreover, you may implement | achieve as a program which makes a computer perform these steps. Furthermore, it may be realized as a recording medium such as a computer-readable CD-ROM (Compact Disc-Read Only Memory) in which the program is recorded, and information, data, or a signal indicating the program. These programs, information, data, and signals may be distributed via a communication network such as the Internet.

  In addition, some or all of the constituent elements included in each of the image encoding devices may be configured by a single system LSI (Large Scale Integration). The system LSI is an ultra-multifunctional LSI manufactured by integrating a plurality of components on a single chip, and specifically includes a microprocessor, a ROM, a RAM (Random Access Memory), and the like. Computer system.

  As is clear from the above description, according to the image coding apparatus of the present invention, more efficient coding is possible in image coding using the intra prediction mode, and the image quality of the restored image is degraded. Can be reduced.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the present embodiment, only the color difference signal is targeted.

  FIG. 1 is a block diagram showing the configuration of the image coding apparatus according to the present embodiment. An image encoding device 10 in FIG. 1 is an image encoding device that encodes an image or a moving image input using intra prediction. The image coding apparatus 10 includes an input unit 100, a macroblock configuration unit 101, an intra prediction unit 102, a subtraction unit 103, an orthogonal transform unit 104, a quantization unit 105, an entropy coding unit 106, and an output. Unit 107, inverse quantization unit 108, inverse transform unit 109, addition unit 110, and peripheral pixel storage unit 111.

  The input unit 100 receives an input image signal from the outside and inputs it to the macroblock configuration unit 101.

  The macroblock configuration unit 101 divides an input image signal into blocks each composed of a plurality of pixels, and collects the divided blocks to configure a macroblock. For example, the block is divided into M × M (M is a power of 2) pixels. More specifically, it is divided into blocks each composed of 4 × 4 pixels, and a plurality of blocks are collected to constitute a macro block. The configured macroblock is input to the intra prediction unit 102 and the subtraction unit 103.

  The intra prediction unit 102 performs intra prediction on the input macroblock using the peripheral pixels input from the peripheral pixel storage unit 111 for each block composed of 4 × 4 pixels. The intra prediction unit 102 includes a plurality of intra prediction modes. The intra prediction unit 102 determines an optimal prediction mode from among a plurality of intra prediction modes based on the executed intra prediction, and generates a prediction block in the optimal prediction mode. The generated prediction block is input to the subtraction unit 103. Note that a block composed of 4 × 4 pixels is referred to as an intra prediction calculation block. Intra prediction is performed in units of intra prediction calculation blocks.

  The subtraction unit 103 calculates a prediction residual signal by taking the difference between the input prediction block and the intra prediction calculation block corresponding to the prediction block. The calculated prediction residual signal is input to the orthogonal transform unit 104.

  The orthogonal transform unit 104 performs orthogonal transform on the input prediction residual signal and outputs an orthogonal transform coefficient. The orthogonal transform coefficient is input to the quantization unit 105.

  The quantization unit 105 performs a quantization process on the input transform coefficient, and outputs the quantized transform coefficient. The quantized transform coefficient is input to the entropy encoding unit 106 and the inverse quantization unit 108.

  The entropy encoding unit 106 performs entropy encoding on the input quantized transform coefficient, and outputs a bit stream. The bit stream is input to the output unit 107.

  The output unit 107 outputs the input bit stream to an arbitrary medium.

  The inverse quantization unit 108 performs inverse quantization on the input quantized transform coefficient, and outputs the inversely quantized transform coefficient. The inversely quantized transform coefficient is input to the inverse transform unit 109.

  The inverse transform unit 109 performs inverse orthogonal transform on the input inverse quantized transform coefficient and outputs a restored residual signal. The restored residual signal is input to the adding unit 110.

  The adder 110 generates a restored image signal by adding the input prediction block and the restored residual signal. The restored image signal is input to the peripheral pixel storage unit 111.

  The peripheral pixel storage unit 111 stores peripheral pixels used for intra prediction.

  With the above configuration, the image coding apparatus according to the present embodiment can prevent an increase in the amount of codes by performing intra prediction using peripheral pixels of the coding target block.

  Next, the configuration of the intra prediction unit 102 of the image encoding device 10 according to the present embodiment will be described in detail.

  FIG. 2 is a block diagram illustrating a configuration of the intra prediction unit 102. The intra prediction unit 102 in FIG. 1 includes a neighboring pixel input unit 200, an intra prediction calculation block input unit 201, a prediction block generation unit 202, a prediction error calculation unit 203, a comparison unit 204, a selection unit 205, and a prediction And a block output unit 206.

  The peripheral pixel input unit 200 receives peripheral pixels used for intra prediction of the intra prediction calculation block from the peripheral pixel storage unit 111. The peripheral pixel input unit 200 outputs the input peripheral pixels to the prediction block generation unit 202.

  The prediction block generation unit 202 generates a prediction block corresponding to each of the intra prediction modes. In the present embodiment, there are five intra prediction modes, and a prediction block is generated based on each intra prediction mode. The five intra prediction modes are “horizontal prediction”, “vertical prediction”, “DC prediction”, “Plane prediction”, and “0 prediction” modes. These are expressed as H mode, V mode, DC mode, Plane mode, and 0 mode, respectively.

  The prediction block generation unit 202 includes a prediction block generation unit corresponding to each prediction mode. That is, the prediction block generation unit 202 includes a prediction block generation unit 211 in the H mode, a prediction block generation unit 212 in the V mode, a prediction block generation unit 213 in the DC mode, and a prediction block generation unit in the Plane mode. 214 and a prediction block generation unit 215 in the 0 mode. Here, as the fifth prediction mode, the provision of the prediction block generation unit 215 in the 0 mode is the greatest feature of the present invention over the conventional method.

  The prediction block generation unit 211 in the H mode generates a prediction block in the H mode using the input neighboring pixels. The prediction block is input to the prediction error calculation unit 203.

  The prediction block generation unit 212 in the V mode generates a prediction block in the V mode using the input neighboring pixels. The prediction block is input to the prediction error calculation unit 203.

  The prediction block generation unit 213 in the DC mode generates a prediction block in the DC mode using the input neighboring pixels. The prediction block is input to the prediction error calculation unit 203.

  The prediction block generation unit 214 in the Plane mode generates a prediction block in the Plane mode using the input peripheral pixels. The prediction block is input to the prediction error calculation unit 203.

  The prediction block generation unit 215 in the 0 mode generates a prediction block in the 0 mode using the input neighboring pixels. The prediction block is input to the prediction error calculation unit 203.

  The prediction error calculation unit 203 calculates a prediction error corresponding to the prediction block generated based on each prediction mode. The prediction error is the sum of absolute difference values between the prediction block and the intra prediction calculation block. The prediction error calculation unit 203 includes a prediction error calculation unit 221 in the H mode, a prediction error calculation unit 222 in the V mode, a prediction error calculation unit 223 in the DC mode, and a prediction error calculation unit 224 in the Plane mode. , A prediction error calculation unit 225 in the 0 mode.

  The prediction error calculation unit 221 in the H mode calculates a prediction error between the input intra prediction calculation block and the prediction block in the H mode. The prediction error in the H mode is input to the comparison unit 204.

  The prediction error calculation unit 222 in the V mode calculates a prediction error between the input intra prediction calculation block and the prediction block in the V mode. The prediction error in the V mode is input to the comparison unit 204.

  The prediction error calculation unit 223 in the DC mode calculates a prediction error between the input intra prediction calculation block and the prediction block in the DC mode. The prediction error in the DC mode is input to the comparison unit 204.

  The prediction error calculation unit 224 in the Plane mode calculates a prediction error between the input intra prediction calculation block and the prediction block in the Plane mode. The prediction error in the Plane mode is input to the comparison unit 204.

  The prediction error calculation unit 225 in the 0 mode calculates a prediction error between the input intra prediction calculation block and the prediction block in the 0 mode. The prediction error in the 0 mode is input to the comparison unit 204.

  The comparison unit 204 determines an optimum prediction mode that realizes the smallest prediction error among the input prediction errors in each mode. The optimal prediction mode is input to the selection unit 205.

  The selection unit 205 outputs the prediction block corresponding to the input optimal prediction mode to the prediction block output unit 206.

  With the above configuration, the image encoding apparatus according to the present embodiment can encode an input image using a prediction block calculated according to the newly added 0 mode. By adding the 0 mode, it is possible to reduce degradation of image quality, which is a conventional problem. The reason and details of the 0 mode will be described later with reference to the drawings.

  Note that the intra prediction unit 102 illustrated in FIG. 2 executes each mode of processing in parallel processing. Thereby, the processing speed of intra prediction is improved. However, it is not always necessary to perform parallel processing, and each mode may be processed sequentially. In this case, although processing takes time, for example, only one prediction error calculation unit may be provided, and the configuration can be simplified.

  Next, the operation of the image coding apparatus according to the present embodiment will be described.

  FIG. 3 is a flowchart showing the operation of the image coding apparatus according to the present embodiment.

  First, the macroblock configuration unit 101 configures a macroblock for the input image signal input from the input unit 100 (S101). For example, the macroblock configuration unit 101 divides the block into 4 × 4 pixel blocks, and collects the divided blocks to configure a macroblock. Here, a macroblock obtained by collecting four blocks, that is, a macroblock of 8 × 8 pixels is configured.

  For each macroblock configured as described above, an encoding process using intra prediction is executed (S102). Further, the encoding process is executed for each 4 × 4 pixel block included in the macroblock (S103).

  First, the intra prediction unit 102 determines the optimal prediction mode by executing intra prediction according to each mode, and generates a prediction block based on the determined optimal prediction mode (S104). Details of the prediction block generation process (S104) will be described later with reference to FIG.

  Next, the subtraction unit 103 calculates a prediction residual signal by taking the difference between the generated prediction block and the corresponding intra prediction calculation block (S105). The orthogonal transform unit 104 performs an orthogonal transform process on the calculated prediction residual signal and outputs an orthogonal transform coefficient (S106). The quantization unit 105 performs a quantization process on the orthogonal transform coefficient, and outputs the quantized transform coefficient (S107). The entropy encoding unit 106 performs entropy encoding on the quantized transform coefficient, and outputs a bit stream to the output unit 107 (S108).

  The image encoding device 10 performs the above-described processing from the prediction block generation process (S104) to the encoding process (S108) for all the blocks included in the macroblock (S109). Then, the image encoding device 10 executes all macroblocks (S110).

  Next, details of the prediction block generation process (S104) of the flowchart shown in FIG. 3 will be described with reference to FIG.

  FIG. 4 is a flowchart showing the operation of the prediction block generation process executed by the intra prediction unit 102.

  The prediction block generation unit 202 generates a prediction block for each prediction mode (S201). In this Embodiment, the prediction block production | generation part corresponding to each prediction mode produces | generates the prediction block corresponding to each prediction mode.

  Next, the prediction error calculation unit 203 calculates a prediction error for each prediction mode (S202). In the present embodiment, a prediction error calculation unit corresponding to each prediction mode calculates a prediction error corresponding to each prediction mode.

  Next, the comparison unit 204 receives the prediction error calculated by each prediction error calculation unit, and compares the received prediction error (S203). And the comparison part 204 determines the prediction mode corresponding to the prediction error from which a value becomes the minimum as prediction prediction mode among prediction errors.

  The selection unit 205 selects a prediction block corresponding to the optimum prediction mode determined by the comparison unit 204, and outputs the prediction block to the prediction block output unit 206 (S204).

  With the above processing, the prediction mode with the smallest prediction error, that is, the prediction mode that is predicted most accurately can be selected, and the encoding process based on the accurate prediction can be executed.

  Next, advantages and effects of the 0 mode, which is a feature of the image coding apparatus according to the present embodiment, will be described with reference to the drawings.

  Hereinafter, a prediction error calculation method in the “0 prediction” mode will be described using blocks 410 and 414 included in the macroblocks 400 and 401 described in FIG. 12 as encoding target blocks, respectively.

  As described above, the block 410 and the block 414 are flat blocks, and all the pixel values in the block are 10. Prediction errors in the four intra prediction modes (H mode, V mode, DC mode, and Plane mode) that are conventionally provided for the block 410 are as follows.

  Subsequently, a prediction error in the “0 prediction” mode is calculated. In the “0 prediction” mode, the pixel values of the prediction block are all 0 regardless of the values of the surrounding pixels.

  As a result, the comparison unit 204 determines the mode with the smallest prediction error as the optimum prediction mode, and therefore the “0 prediction” mode with the smallest prediction error is selected as the optimum prediction mode of the block 410. The other blocks 411, 412 and 413 belonging to the macro block 400 also have the same result as the block 410.

  All of the pixel values of the residual block obtained by performing the difference calculation between the block 410 and the prediction block are −10. When the quantization coefficient is larger than the transform coefficient obtained by performing the orthogonal transform process on the residual block, the transform coefficient after quantization becomes 0. In this case, the restored transform coefficient after inverse quantization is also 0, and the restored residual signal obtained by performing inverse orthogonal transform on the restored transform coefficient is also 0. FIG. 5 is a diagram illustrating the pixel values of the restored macroblock obtained by restoring the macroblock encoded by the image encoding device according to the present embodiment. As shown in FIG. 5, all the pixel values in the restoration macroblock 330 are zero.

  A pixel value of 0 in the color signal indicates that there is no color component, that is, an achromatic color. FIG. 6 is a diagram illustrating an example of an image obtained by restoring an image encoded by the image encoding device according to the present embodiment. As shown in FIG. 6, the deterioration in which the color component in the restored macro block 330 disappears is a visually acceptable deterioration than the deterioration in which the color component that does not exist in the original block shown in the prior art appears.

  Next, a similar verification is performed for block 414.

  As shown in FIG. 5, it is assumed that the block 414 is a flat block and all the pixel values in the block are 10. The peripheral pixels 322 and 323 are pixels used for intra prediction. The peripheral pixels also include the pixels of the restored macroblock 330. As described above, in the restored macroblock 330, the values of all the pixels are 0 by intra prediction.

  The prediction errors in the four intra prediction modes (H mode, V mode, DC mode, and Plane mode) and the “0 prediction” mode that are conventionally provided for the block 414 are as follows.

  In this case, the prediction error between the H mode and the 0 prediction mode becomes equal, but the H mode is prioritized, and therefore the H mode is selected as the optimum prediction mode of the block 414. The other blocks 415, 416 and 417 belonging to the macro block 401 have the same result as the block 414.

  In the 0 mode, since the prediction block is forcibly set to 0, that is, intra prediction is not performed, it is not preferable to frequently use the 0 mode from the viewpoint of image quality. For this reason, in this Embodiment, when either of the prediction errors of the four intra prediction modes provided conventionally and the prediction error of 0 mode correspond, the intra prediction mode provided conventionally is given priority.

  All of the pixel values of the residual block obtained by calculating the difference between the block 414 and the prediction block are -10. When the quantization coefficient is larger than the transform coefficient obtained by performing the orthogonal transform process on the residual block, the transform coefficient after quantization becomes 0. In this case, the restored transform coefficient after inverse quantization is also 0, and the restored residual signal obtained by performing inverse orthogonal transform on the restored transform coefficient is also 0. FIG. 7 is a diagram illustrating pixel values of the restored macroblock obtained by restoring the macroblock encoded by the image encoding device according to the present embodiment. As shown in FIG. 7, the pixel values in the restoration macroblock 331 are all 0.

  Unlike the prior art, the pixel values in the restoration macroblock 331 are all 0 because a part of the object is not diffused in the restoration macroblock 330. This is because the blocks 410, 411, 412 and 413 are not diffused. This is because the prediction mode is set to the “0 prediction” mode.

  A pixel value of 0 in the color signal indicates that there is no color component, that is, an achromatic color.

  FIG. 8 is a diagram illustrating an image obtained by restoring an image encoded by the image encoding device according to the present embodiment. As shown in FIG. 8, the deterioration in which the color component in the restored macro block 331 disappears is a visually acceptable deterioration than the deterioration in which the color component that does not exist in the original block shown in the prior art appears.

  As described above, according to the image coding apparatus of the present embodiment, the “0 prediction” mode is provided as the intra prediction mode, and all the pixel values of the prediction block are set to 0 for the block for which the mode is selected. And As a result, it is possible to prevent the appearance of color components that were not present in the original block, and the image quality can be improved as compared with the prior art.

  As described above, the image coding apparatus and the image coding method of the present invention have been described based on the embodiment. However, the present invention is not limited to this embodiment. Unless it deviates from the meaning of this invention, what made the various deformation | transformation which those skilled in the art can consider to the said embodiment is also contained in the scope of the present invention.

  For example, in the present embodiment, the pixel values of the prediction block set in the “0 prediction” mode are all set to 0, but any specific value may be used. As described above, since the prediction block becomes a restoration block, a random signal such as minute noise may be used as the prediction block in order to enhance the visual texture of the image.

  Moreover, although this Embodiment demonstrated only the case where an intra prediction block was 4x4, this invention is not limited to this. For example, the present invention can be applied to 8 × 8 intra prediction, and further can be applied to 16 × 16 intra prediction.

  Further, in the present embodiment, the difference between the intra prediction calculation block and the prediction block is calculated inside the intra prediction unit 102, and the difference between the intra prediction calculation block and the prediction block is calculated again by the subtraction unit 103. It is configured to do. Since the subtraction process in two stages is performed in this way, a configuration in which a residual block obtained by calculating a difference in the intra prediction unit 102 is input to the orthogonal transform unit 104 is preferable.

  In this case, the subtraction unit 103 is omitted, and the prediction error calculation unit 203 outputs not only the prediction error that is the sum of absolute difference values but also the residual block that is the difference between the intra prediction calculation block and the prediction block. Based on the control from the comparison unit 204, the selection unit 205 selects a residual block that minimizes the prediction error and outputs the residual block to the orthogonal transform unit 104.

  With the above configuration, since the subtracting unit can be omitted, the configuration can be simplified.

  As described above, the present invention can be realized not only as an image encoding device and an image encoding method, but also as a program for causing a computer to execute the image encoding method of the present embodiment. Moreover, you may implement | achieve as recording media, such as computer-readable CD-ROM which records the said program. Furthermore, it may be realized as information, data, or a signal indicating the program. These programs, information, data, and signals may be distributed via a communication network such as the Internet.

  Further, according to the present invention, some or all of the constituent elements constituting the image encoding apparatus may be configured from one system LSI. The system LSI is an ultra-multifunctional LSI manufactured by integrating a plurality of components on a single chip. Specifically, the system LSI is a computer system including a microprocessor, a ROM, a RAM, and the like. .

  The image coding apparatus and the image coding method according to the present invention can select an intra prediction method with higher reproducibility, and are effective in improving the quality of a reproduced image. For example, the image encoding device and the image encoding method of the present invention are useful for a video encoder, a soft encoder that operates on a nonlinear editing machine, and the like.

It is a block diagram which shows the structure of the image coding apparatus of this Embodiment. It is a block diagram which shows the structure of the intra estimation part with which the image coding apparatus of this Embodiment is provided. It is a flowchart which shows operation | movement of the image coding apparatus of this Embodiment. It is a flowchart which shows the operation | movement of the production | generation process of the prediction block which an intra estimation part performs. It is a figure which shows the pixel value of the decompression | restoration macroblock obtained by decompress | restoring the macroblock encoded by the image coding apparatus of this Embodiment. It is a figure which shows the image obtained by decompress | restoring the image encoded by the image coding apparatus of this Embodiment. It is a figure which shows the pixel value of the decompression | restoration macroblock obtained by decompress | restoring the macroblock encoded by the image coding apparatus of this Embodiment. It is a figure which shows the image obtained by decompress | restoring the image encoded by the image coding apparatus of this Embodiment. It is a figure which shows the intra prediction block (4x4 pixel) in a macroblock (16x16 pixel). It is a figure which shows nine intra prediction modes with respect to a luminance signal. It is a figure which shows four intra prediction modes with respect to a color difference signal. It is a figure which shows an example of an encoding object picture and an encoding object macroblock. It is a figure which shows the pixel value of the pixel contained in an encoding object macroblock, and a surrounding pixel. It is a figure which shows the pixel value of the decompression | restoration macroblock obtained by decompress | restoring the macroblock encoded by the prior art. It is a figure which shows the image after decompression | restoration obtained by decompress | restoring the image encoded by the prior art. It is a figure which shows the pixel value of the decompression | restoration macroblock obtained by decompress | restoring the macroblock encoded by the prior art. It is a figure which shows the image after decompression | restoration obtained by decompress | restoring the image encoded by the prior art.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Image coding apparatus 100 Input part 101 Macroblock structure part 102 Intra prediction part 103 Subtraction part 104 Orthogonal transformation part 105 Quantization part 106 Entropy encoding part 107 Output part 108 Inverse quantization part 109 Inverse transformation part 110 Addition part 111 Around Pixel storage unit 200 Peripheral pixel input unit 201 Intra prediction calculation block input unit 202 Prediction block generation unit 203 Prediction error calculation unit 204 Comparison unit 205 Selection unit 206 Prediction block output unit 211 Prediction block generation unit 212 in H mode In V mode Prediction block generation unit 213 Prediction block generation unit 214 in DC mode Prediction block generation unit 215 in Plane mode Prediction block generation unit 221 in 0 mode prediction error calculation unit 222 in H mode Prediction error calculation unit 223 in V mode Prediction error in DC mode Calculation unit 224 Prediction error calculation unit 225 in Plane mode Prediction error calculation unit 322, 323, 420, 421, 422, 423 Peripheral pixels 330, 331, 430, 431 Restoration macroblock 400, 401 Macroblock 410, 411, 412, 413, 414, 415, 416, 417 blocks

Claims (7)

An image encoding device that encodes an input image based on an intra prediction method,
For each block composed of a plurality of pixels into which the input image is divided, an intra prediction method with the highest prediction accuracy is selected from among a plurality of intra prediction methods, and pixel values predicted according to the selected intra prediction method are selected. Intra prediction means for generating a prediction block comprising a plurality of pixels in each of them,
Subtracting means for calculating a residual block that is a difference between the prediction block generated by the intra prediction means and the block corresponding to the prediction block;
Encoding means for encoding the residual block calculated by the subtracting means,
The image encoding device includes an intra prediction method for generating a prediction block in which a pixel value of a pixel included in the plurality of intra prediction methods is a predetermined value. The image coding apparatus according to claim 1, wherein the predetermined value is 0. The intra prediction means includes
A prediction block generation unit that generates a prediction block corresponding to each of the plurality of intra prediction schemes;
A prediction error calculation unit that calculates a prediction error that is a sum of absolute differences of a plurality of pixels included in the prediction block generated by the prediction block generation unit, for each intra prediction method;
The image coding apparatus according to claim 1, further comprising: a selection unit that selects a prediction block generated according to an intra prediction method that minimizes the prediction error. The intra prediction unit generates the prediction block for each color block including a plurality of pixels indicating a value of a color component of the input image,
The subtraction unit calculates a residual block that is a difference between the prediction block generated by the intra prediction unit and the color block corresponding to the prediction block. Image encoding device. An image encoding method for encoding an input image based on an intra prediction method,
For each block composed of a plurality of pixels into which the input image is divided, an intra prediction method with the highest prediction accuracy is selected from among a plurality of intra prediction methods, and pixel values predicted according to the selected intra prediction method are selected. An intra prediction step for generating a prediction block composed of a plurality of pixels in each of them;
A subtraction step of calculating a residual block that is a difference between the prediction block generated in the intra prediction step and the block corresponding to the prediction block;
An encoding step for encoding the residual block calculated in the subtraction step,
The image encoding method, wherein the plurality of intra prediction methods include an intra prediction method for generating a prediction block in which a pixel value of a pixel included is a predetermined value. A program for causing a computer to execute an image encoding method for encoding an input image based on an intra prediction method,
For each block composed of a plurality of pixels into which the input image is divided, an intra prediction method with the highest prediction accuracy is selected from among a plurality of intra prediction methods, and pixel values predicted according to the selected intra prediction method are selected. An intra prediction step for generating a prediction block composed of a plurality of pixels in each of them;
A subtraction step of calculating a residual block that is a difference between the prediction block generated in the intra prediction step and the block corresponding to the prediction block;
An encoding step for encoding the residual block calculated in the subtraction step,
The plurality of intra prediction schemes include an intra prediction scheme for generating a prediction block in which pixel values of included pixels have predetermined values. An integrated circuit that encodes an input image based on an intra prediction method,
For each block composed of a plurality of pixels into which the input image is divided, an intra prediction method with the highest prediction accuracy is selected from among a plurality of intra prediction methods, and pixel values predicted according to the selected intra prediction method are selected. Intra prediction means for generating a prediction block comprising a plurality of pixels in each of them,
Subtracting means for calculating a residual block that is a difference between the prediction block generated by the intra prediction means and the block corresponding to the prediction block;
Encoding means for encoding the residual block calculated by the subtracting means,
The plurality of intra prediction schemes include an intra prediction scheme for generating a prediction block in which pixel values of pixels included are predetermined values. Integrated circuit.

Images