JP5340440B2 - Video encoding apparatus and video encoding method - Google Patents

Description

  The present invention relates to a video encoding device, a video encoding method, a program, and a recording medium. The present invention relates to a technique suitable for use in H.264 intra prediction.

  One of the encoding methods for moving image data is the MPEG-4 AVC (ISO / IEC 14496-10, also known as H.264) method, which is used in digital broadcasting and video recording media. In the MPEG-4 AVC (H.264) system, in an intra image in which motion compensation prediction is not performed, encoding efficiency is improved by intra-screen prediction of the pixel value of the image. Thus, the concept of improving the encoding efficiency by prediction using information in the screen is taken from the MPEG-4 (ISO / IEC 14496-2) system. In the MPEG-4 system, a DC coefficient (DC component) and a low-order AC coefficient (AC component) of a DCT block to be encoded are predicted from a DCT block located on the left or top in 8 × 8 pixel block units. / AC coefficient predictive coding method is adopted.

  On the other hand, in the MPEG-4 AVC system, 16 × 16 pixel blocks called macroblocks are targeted for prediction encoding in units of 16 × 16 pixel blocks, 8 × 8 pixel blocks, or 4 × 4 pixel blocks. The size can be selected. Further, in the MPEG-4 AVC system, prediction is performed in a plurality of prediction modes using encoded pixel values in pixels adjacent to the upper, upper right, left, and upper left of each pixel block to be encoded. Then, by selecting a prediction mode that is most similar to the pixel value of each pixel block to be encoded, all pixel values of the macroblock are predicted to further increase the encoding efficiency. Also, it is possible to perform intra-frame prediction and encoding for non-intra-pictures in units of macroblocks.

  The intra prediction is performed by a process of selecting any one of a plurality of intra prediction modes. The details of the intra prediction mode are disclosed in Patent Document 1, for example.

JP 2006-5659 A

  As described above, in-picture prediction in the MPEG-4 AVC format requires encoded pixels adjacent to the pixel block to be encoded, and pixel data necessary for actual in-picture prediction processing is the pixel to be encoded. It is a pixel value of a part of pixels adjacent to the block. However, it is necessary to retain all the encoded images in preparation for inter-screen prediction, and furthermore, the de-blocking filter processing is performed between pixel blocks on the encoded image that is referred to in inter-screen prediction by motion compensation. There is a need to do.

  On the other hand, an encoded image that is referred to in intra prediction requires an encoded pixel value before the deblocking filter process. Therefore, in order to perform intra prediction, it is necessary to consider the timing for performing deblocking filter processing on the encoded image, and there is a problem that it takes extra time to perform intra prediction. there were.

  In general, encoded pixel blocks are held in a memory in units of pixel blocks. For this reason, in particular, when performing prediction processing using a pixel adjacent to the left of the pixel block to be encoded, it is necessary to redundantly read out the adjacent pixel data that is held, so that intra-screen prediction processing is performed. There was a problem that it took extra time.

  In view of the above-described problems, an object of the present invention is to enable intra-screen prediction processing to be performed at high speed.

  A video encoding apparatus according to the present invention is a video encoding apparatus that performs encoding by dividing an encoding target image into pixel block units each having a predetermined number of pixels, and stores the encoded image in a memory in units of the pixel blocks. Generating an intra-screen prediction processing unit that performs intra-screen prediction processing using the adjacent pixel data, and generating a prediction residual indicating a difference between the encoding target image and a predicted image obtained by the intra-screen prediction processing Means for performing orthogonal transform on the prediction residual and then encoding, decoding means for performing inverse transform on the data after the orthogonal transform and decoding into units of the pixel block, and the decoding The pixel data for one horizontal line at the lower end of the pixel block decoded by the conversion unit and the pixel data for one line in the vertical direction at the right end of the pixel block adjacent to the decoded pixel block Storage means for storing in the memory for use in in-plane prediction processing, the memory storing at least a first storage area for storing pixel data for one horizontal line of the encoding target image, and the pixels A second storage area for storing pixel data for one vertical line of the block, and the storage means is a rightmost pixel in one horizontal line at the lower end of the pixel block decoded by the decoding means The pixel data is stored in the first storage area after intra-screen prediction processing is performed on the pixel block on the right next with reference to the pixel data already stored in the first storage area.

  The video encoding method of the present invention is a video encoding method that performs encoding by dividing an image to be encoded into units of pixel blocks each having a predetermined number of pixels, and at least the code in units of the pixel blocks. Stored in a memory having a first storage area for storing pixel data for one line in the horizontal direction of the image to be converted and a second storage area for storing pixel data for one line in the vertical direction of the pixel block. An intra-screen prediction processing step for performing intra-screen prediction processing using adjacent pixel data, and a generation step for generating a prediction residual indicating a difference between the encoding target image and a prediction image obtained by the intra-screen prediction processing; An encoding step of performing orthogonal transform on the prediction residual and then encoding; a decoding step of performing inverse transform on the data after the orthogonal transform and decoding into units of the pixel block; and the decoding The pixel data for one horizontal line at the lower end of the decoded pixel block and the pixel data for one vertical line at the right end of the decoded pixel block are predicted in the screen of the pixel block adjacent to the decoded pixel block. A storage step of storing in the memory for use in processing, and in the storage step, the pixel data of the rightmost pixel in one horizontal line at the lower end of the pixel block decoded in the decoding step, The pixel data stored in the first storage area is stored in the first storage area after being subjected to the intra-screen prediction process for the right adjacent pixel block.

  A program of the present invention is a program for controlling a video encoding device that performs encoding by dividing an encoding target image into pixel block units each having a predetermined number of pixels, and the pixel block unit A memory having at least a first storage area for storing pixel data for one line in the horizontal direction of the encoding target image and a second storage area for storing pixel data for one line in the vertical direction of the pixel block. Generates a prediction residual indicating a difference between an intra-screen prediction processing step of performing intra-screen prediction processing using stored adjacent pixel data and the prediction target image and a predicted image obtained by the intra-screen prediction processing. A generation step, an encoding step of performing encoding after the prediction residual is orthogonally transformed, and a decoding of inversely transforming the data after the orthogonal transformation and decoding the unit of the pixel block And the pixel data for one horizontal line at the lower end of the pixel block decoded in the decoding step and the pixel data for one line in the vertical direction at the right end are adjacent to the decoded pixel block. And a storage step of storing the pixel block in the memory for use in the intra-screen prediction process, and in the storage step, one horizontal line at the lower end of the pixel block decoded in the decoding step Storing the pixel data of the rightmost pixel in the first storage area after the intra-screen prediction process of the right pixel block referring to the pixel data stored in the first storage area is performed. Features.

  The recording medium of the present invention records (stores) the program described above.

  According to the present invention, adjacent pixel data can be stored with a minimum of memory resources, and intra prediction processing can be performed at high speed.

It is a block diagram which shows the internal structural example of the video coding apparatus which concerns on embodiment of this invention. It is a figure which shows the structural example of the 1st memory of embodiment of this invention. It is a figure which shows the process target macroblock of the prediction in a screen in a picture. It is a figure which shows the position of the encoded pixel to refer in the MPEG-4 AVC system. It is a figure which shows the order of the pixel block which performs prediction within a macroblock. It is a figure which shows the prediction mode of a 4x4 pixel block unit in a luminance signal. It is a figure which shows the prediction mode of a 16 * 16 pixel block unit in a luminance signal. It is a figure which shows the prediction mode of 8x8 pixel block unit in a color difference signal. The figure which shows an example of the update procedure (upper side) of the line memory and MB memory at the time of carrying out the prediction process in a screen per 4 * 4 pixel block in order of FIG. 5 in embodiment of this invention. It is. FIG. 6 is a diagram illustrating an example of a line memory and MB memory update procedure (lower side) when intra-screen prediction processing in units of 4 × 4 pixel blocks is performed in the order shown in FIG. 5 in the embodiment of the present invention. FIG. 6 is a diagram showing an example of a procedure for updating a line memory and an MB memory when intra-screen prediction processing in units of 8 × 8 pixel blocks is performed in the order shown in FIG. 5 in the embodiment of the present invention. In an embodiment of the present invention, it is a figure showing an example of an update procedure of a line memory and an MB memory when an in-screen prediction process is performed in units of 16 × 16 pixel blocks.

(First embodiment)
A video encoding apparatus that encodes a video signal in the MPEG-4 AVC system, which is an embodiment to which the present invention can be applied, will be described with reference to the block diagram of FIG. In MPEG-4 AVC, predictive encoding is performed by dividing a plurality of screens constituting a moving image into units of macroblocks (or smaller blocks). Note that the video signal to be encoded is a 4: 2: 0 component signal.

FIG. 1 is a block diagram illustrating an internal configuration example of a video encoding device according to the present embodiment.
In FIG. 1, the video encoding apparatus 100 according to the present embodiment performs processing with a macroblock obtained by dividing a target screen to be encoded into pixel blocks of 16 × 16 pixel units. The prediction method determination unit 101 determines a prediction method for each macroblock in the encoding target screen.

  The first memory 102 stores pixel data of an encoded image used for intra-screen prediction, and the second memory 103 stores an encoded image used for inter-screen prediction. The prediction method determination unit 101 performs prediction processing in all prediction methods using pixel data of an encoded image read from the first memory 102 or the second memory 103 and an input video signal, and performs encoding. Decide on a prediction method that optimizes efficiency.

  Here, the first memory 102 stores the encoded image before the deblocking filter processing unit 111 performs the deblocking filter processing. On the other hand, the second memory 103 stores the encoded image that has been subjected to the deblocking filter processing by the deblocking filter processing unit 111.

  The prediction method determination unit 101 selects intra prediction when the macroblock to be encoded is an I slice, and, when it is a P slice or B slice, the encoding efficiency of the intra prediction and inter prediction is selected. Choose the higher one. When the intra-screen prediction is selected, the intra-screen prediction parameters such as the size of the pixel block to be subjected to the intra-screen prediction and the intra-screen prediction mode are determined. On the other hand, when the inter-screen prediction is selected, inter-screen prediction parameters such as a reference image frame, a macroblock division pattern, and a motion vector are determined.

  When the determined prediction method is intra-screen prediction, the intra-screen prediction parameter is output to the intra-screen prediction processing unit 105. When the determined prediction method is inter-screen prediction, the inter-screen prediction processing unit 104 is output. Output inter-screen prediction parameters.

  The intra-screen prediction processing unit 105 generates a prediction image from the encoded image read from the first memory 102 according to the intra-screen prediction parameter determined by the prediction method determination unit 101, and a prediction residual generation unit To 106. The inter-screen prediction processing unit 104 reads an encoded image corresponding to the motion vector in the designated reference frame from the second memory 103 according to the inter-screen prediction parameter determined by the prediction method determining unit 101. . Then, a prediction image is generated by the prediction interpolation process, and is output to the prediction residual generation unit 106.

  The prediction residual generation unit 106 holds the prediction image input from the intra-screen prediction processing unit 105 or the inter-screen prediction processing unit 104 in a memory (not shown). Further, a prediction residual signal indicating a difference between the encoding target image (pixel block) and the predicted image is generated and output to the orthogonal transform unit 107.

  The orthogonal transform unit 107 performs an orthogonal transform process by integer precision discrete cosine transform and discrete Hadamard transform on a designated pixel block unit (8 × 8 pixel or 4 × 4 pixel block unit). The discrete Hadamard transform is performed on a DC (direct current) component obtained as a result of integer-precision discrete cosine transform of each pixel block with respect to a luminance signal or color difference signal subjected to intra-screen prediction processing in units of 16 × 16 pixel blocks. Only done. Then, the orthogonally transformed transform coefficient is supplied to the quantization unit 108.

  The quantizing unit 108 performs a transform coefficient quantization process in a quantization step according to the designated quantization parameter, and outputs the result to the entropy coding unit 109. In the present embodiment, the intra prediction processing unit 105, the prediction residual generation unit 106, the orthogonal transform unit 107, and the quantization unit 108 function as intra prediction encoding means.

  The entropy encoding unit 109 performs entropy encoding processing on the input data and outputs encoded data. One encoding method is CAVLC (Context-based Adaptive Variable Length Coding) processing. The other is CABAC (Context-based Adaptive Binary Arithmetic Coding) processing. The entropy encoding unit 109 performs entropy encoding processing using any of these, and outputs encoded data.

  On the other hand, the data quantized by the quantization unit 108 is input to the local decoding unit 110 at the same time. The local decoding unit 110 functions as a decoding unit, performs inverse quantization processing and inverse orthogonal transform processing (inverse discrete Hadamard transform and inverse integer precision discrete cosine transform) on the input quantized data, and predicts residuals. The prediction image input from the generation unit 106 is added to perform decoding. The local decoding unit 110 also functions as a storage unit, and stores the decoded data in the first memory 102 and outputs it to the deblocking filter processing unit 111.

  The decoded data held in the first memory 102 is used for intra-screen prediction processing. The deblocking filter processing unit 111 performs deblocking filter processing on the input decoded data and stores it in the second memory 103. The decoded data after the deblocking filter process held in the second memory 103 is used for the inter-screen prediction process. The video encoding apparatus 100 includes the above configuration.

FIG. 2 is a diagram illustrating a configuration example of the first memory 102 according to the present embodiment.
As shown in FIG. 2, the first memory 102 has a memory A-1 area (hereinafter referred to as a line memory 201) that can hold pixel data for one line in the horizontal direction of the screen as a first storage area. Have. Further, as the second storage area, a memory A-2 area (hereinafter referred to as MB memory 202) capable of holding pixel data of 16 pixels corresponding to one line in the vertical direction in the macroblock is provided. In the present embodiment, an example is shown in which the respective regions are separated, but they may be integrated.

  Next, the procedure for reading and updating the MB memory 202 and the line memory 201 will be described by taking as an example the case where the Nth macroblock on the screen shown in FIG. In FIG. 3, pixels indicated by hatching indicate encoded pixels stored in the MB memory 202 and the line memory 201 when encoding of the (N−1) th macroblock is completed. .

First, intra-screen prediction processing in the MPEG-4 AVC format will be described. FIG. 4 is a diagram showing the positions of encoded pixels to be referred to in the MPEG-4 AVC system.
As shown in FIG. 4, in the MPEG-4 AVC system, it is possible to refer to encoded pixels adjacent to the left, upper, upper right, and upper left of the encoding target pixel block. In the case of a luminance signal, prediction can be selected in units of 4 × 4 pixel blocks, 8 × 8 pixel blocks, or 16 × 16 pixel blocks in a macro block composed of 16 × 16 pixel blocks. is there. In the case of a color difference signal, prediction can be performed in units of 8 × 8 pixel blocks in a macro block composed of 8 × 8 pixel blocks.

FIG. 5 is a diagram illustrating the order of pixel blocks for performing prediction within a macroblock.
When the luminance signal is predicted with a pixel block smaller than the 16 × 16 pixel block, in the case of a 4 × 4 pixel block unit, the prediction is performed in the order of block numbers 0 to 15 as shown in FIG. On the other hand, in the case of an 8 × 8 pixel block unit, prediction is performed in the order of block numbers 0 to 4 as shown in FIG.

  Also, an intra-screen prediction mode is defined for each pixel block size. In the case of a luminance signal, nine types of prediction modes shown in FIG. 6 are defined in units of 4 × 4 pixel blocks. Similarly, in the case of 8 × 8 pixel block units, nine types of prediction modes are defined, and in the case of 16 × 16 pixel block units having the same size as the macroblock, the four types of prediction modes shown in FIG. Is stipulated. On the other hand, in the case of a color difference signal, four types of prediction modes shown in FIG. 8 are defined. It is possible to selectively determine the size of these pixel blocks and the intra prediction mode and perform encoding by intra prediction.

  Hereinafter, a procedure for reading and updating from the MB memory 202 and the line memory 201 for each pixel block size will be described with reference to FIGS. 9 to 12. It should be noted that the intra prediction modes are all representatively DC (a prediction image is generated with an average value of reference pixels) mode.

FIGS. 9 and 10 show the update procedure of the line memory 201 and the MB memory 202 when the intra-screen prediction processing in units of 4 × 4 pixel blocks is performed in the order shown in FIG. 5A in the present embodiment. It is a figure which shows an example.
9 and FIG. 10, the line memory for storing the encoded pixel data in the pixel adjacent on the upper side with respect to the position of the macroblock to be encoded is shown separately in units of four pixels. That is, as shown in FIG. 9, the line memory 201a, the line memory 201b, the line memory 201c, and the line memory 201d are shown separately. On the other hand, the MB memory storing the encoded pixel data in the pixel adjacent on the left side with respect to the position of the macroblock to be encoded is also shown separately in units of four pixels. That is, as shown in FIG. 9, the MB memory 202a, the MB memory 202b, the MB memory 202c, and the MB memory 202d are shown separately. The numbers shown in each memory indicate the block number to which the stored encoded pixel data belongs.

  First, in FIG. 9A, the intra-screen prediction processing unit 105 generates a prediction image for the pixel block with the block number 0 using the encoded pixel data read from the line memory 201a and the MB memory 202a. To do. Then, the prediction residual generation unit 106 generates a prediction residual signal between the prediction image and the encoding target image. Next, the orthogonal transform unit 107 and the quantization unit 108 perform orthogonal transform and quantization on the prediction residual signal, respectively, and output to the entropy coding unit 109 and the local decoding unit 110.

  The local decoding unit 110 adds the data obtained by inverse quantization and inverse orthogonal transform to the predicted image and decodes the data. Of the decoded data, the pixel data for the bottom four pixels is written back to the line memory 201a and the pixel data for the rightmost four pixels is written back to the MB memory 202a for intra prediction. In the case of a pixel block located at a position other than the left end of the screen, the pixel data of the 16th pixel in the MB memory 202d is copied to the line memory 201t. Here, the line memory 201t corresponds to a portion storing data of the 16th pixel in the line memory 201 in the macroblock adjacent on the left side.

  Next, in FIG. 9B, the in-screen prediction processing unit 105 reads out the encoded pixel data from the line memory 201b and the MB memory 202a for the pixel block with the block number 1, and generates a prediction image. . At this time, the pixel data for the four rightmost encoded pixels at block number 0 updated in the MB memory 202a is used for the prediction process. Then, among the data decoded in the same manner as described above, the pixel data for the lower four pixels is written back to the line memory 201b for intra prediction, and the pixel data for the four rightmost pixels is stored in the MB memory 202a. Write back to

  Next, in FIG. 9C, the intra-screen prediction processing unit 105 reads out the encoded pixel data from the line memory 201a and the MB memory 202b for the pixel block with the block number 2, and generates a prediction image. . At this time, the pixel data for the encoded bottom four pixels of block number 0 updated in the line memory 201a is used for the prediction process. Then, among the data decoded in the same manner as described above, the pixel data for the bottom four pixels is written back to the line memory 201a for intra prediction, and the pixel data for the four pixels at the right end is stored in the MB memory 202b. Write back to

  Next, in FIG. 9D, the in-screen prediction processing unit 105 reads out the encoded pixel data from the line memory 201a and the MB memory 202b for the pixel block with the block number 3, and generates a prediction image. . At this time, the pixel data for the encoded bottom 4 pixels of block number 1 updated in the line memory 201b and the pixel data for the encoded right end 4 pixels of block number 2 updated in the MB memory 202b are predicted. Used for processing. Then, among the data decoded in the same manner as described above, the pixel data for the bottom four pixels is written back to the line memory 201a for intra prediction, and the pixel data for the four pixels at the right end is stored in the MB memory 202b. Write back to

  Next, in FIG. 9E, the intra-screen prediction processing unit 105 reads out the encoded pixel data from the line memory 201c and the MB memory 202a for the pixel block with the block number 4, and generates a prediction image. . At this time, the pixel data for the encoded right end four pixels of the block number 1 updated in the MB memory 202a is used for the prediction process. Then, among the decoded data in the same manner as described above, the pixel data for the lower four pixels is written back to the line memory 201c for intra prediction, and the pixel data for the four rightmost pixels is stored in the MB memory 202a. Write back to

  Next, in FIG. 9F, the intra-screen prediction processing unit 105 reads the encoded pixel data from the line memory 201d and the MB memory 202a for the pixel block with the block number 5, and generates a prediction image. . At this time, the pixel data for the four rightmost encoded pixels of block number 4 updated in the MB memory 202a is used for the prediction process. Then, among the data decoded in the same manner as described above, the pixel data for the lower four pixels is written back to the line memory 201d for intra prediction, and the pixel data for the four rightmost pixels is stored in the MB memory 202a. Write back to

  Next, in FIG. 9G, the intra-screen prediction processing unit 105 reads out the encoded pixel data from the line memory 201c and the MB memory 202b for the pixel block with the block number 6, and generates a prediction image. . At this time, the pixel data for the encoded bottom 4 pixels of block number 4 updated in the line memory 201c and the pixel data for the encoded right end 4 pixels of block number 3 updated in the MB memory 202b are predicted. Used for processing. Then, among the decoded data in the same manner as described above, the pixel data for the lower four pixels is written back to the line memory 201c for intra prediction, and the pixel data for the four rightmost pixels is stored in the MB memory 202b. Write back to

  Next, in FIG. 9H, the intra-screen prediction processing unit 105 reads out the encoded pixel data from the line memory 201d and the MB memory 202b for the pixel block with the block number 7, and generates a prediction image. . At this time, the pixel data for the encoded bottom 4 pixels of block number 5 updated in the line memory 201d and the pixel data for the encoded right end 4 pixels of block number 6 updated in the MB memory 202b are predicted. Used for processing. Then, among the decoded data in the same manner as described above, the pixel data for the lower four pixels is written back to the line memory 201d for intra prediction, and the pixel data for the four rightmost pixels is stored in the MB memory 202b. Write back to

  Next, in FIG. 10A, the intra-screen prediction processing unit 105 reads out the encoded pixel data from the line memory 201a and the MB memory 202c for the pixel block with the block number 8, and generates a predicted image. . At this time, the pixel data for the encoded bottom four pixels of block number 2 updated in the line memory 201a is used for the prediction process. Then, among the data decoded in the same manner as described above, the pixel data for the bottom four pixels is written back to the line memory 201a for intra prediction, and the pixel data for the four pixels at the right end is stored in the MB memory 202c. Write back to

  Next, in FIG. 10B, the intra-screen prediction processing unit 105 reads out the encoded pixel data from the line memory 201b and the MB memory 202c for the pixel block with the block number 9, and generates a prediction image. . At this time, the pixel data for the encoded bottom 4 pixels of block number 3 updated in the line memory 201b and the pixel data for the encoded right end 4 pixels of block number 8 updated in the MB memory 202c are predicted. Has been used. Then, among the data decoded in the same manner as described above, the pixel data for the bottom four pixels is written back to the line memory 201b for intra prediction, and the pixel data for the four pixels at the right end is stored in the MB memory 202c. Write back to

  Next, in FIG.10 (c), the intra prediction process part 105 reads the encoded pixel data from the line memory 201a and MB memory 202d with respect to the pixel block of the block number 10, and produces | generates a prediction image. . At this time, the pixel data for the encoded bottom 4 pixels of block number 8 updated in the line memory 201a is used for the prediction process. Then, among the data decoded in the same manner as described above, the pixel data for the lower four pixels is written back to the line memory 201a for intra prediction, and the pixel data for the rightmost four pixels is stored in the MB memory 202d. Write back to

  Next, in FIG. 10D, the intra-screen prediction processing unit 105 reads out encoded pixel data from the line memory 201b and the MB memory 202d for the pixel block with the block number 11, and generates a prediction image. . At this time, the pixel data for the encoded bottom 4 pixels of block number 9 updated in the line memory 201b and the pixel data for the encoded right end 4 pixels of block number 10 updated in the MB memory 202d are predicted. Used for processing. Then, among the data decoded in the same manner as described above, the pixel data for the lower four pixels is written back to the line memory 201b for intra prediction, and the pixel data for the right four pixels is stored in the MB memory 202d. Write back to

  Next, in FIG.10 (e), the intra prediction process part 105 reads the encoded pixel data from the line memory 201c and MB memory 202c with respect to the pixel block of the block number 12, and produces | generates a prediction image. . At this time, the pixel data for the encoded bottom 4 pixels of block number 6 updated in the line memory 201c and the pixel data for the encoded right end 4 pixels of block number 9 updated in the MB memory 202c are predicted. Used for processing. Then, among the data decoded in the same manner as described above, the pixel data for the lower four pixels is written back to the line memory 201c for intra prediction, and the pixel data for the four rightmost pixels is stored in the MB memory 202c. Write back to

  Next, in FIG. 10F, the intra-screen prediction processing unit 105 reads the encoded pixel data from the line memory 201d and the MB memory 202c for the pixel block with the block number 13, and generates a prediction image. . At this time, the pixel data for the encoded bottom four pixels of block number 7 updated in the line memory 201d and the pixel data for the encoded right end four pixels of block number 12 updated in the MB memory 202c are predicted. Used for processing. Then, among the data decoded in the same manner as described above, the pixel data for the lower four pixels is written back to the line memory 201d for intra prediction, and the pixel data for the four rightmost pixels is stored in the MB memory 202c. Write back to

  Next, in FIG. 10G, the intra-screen prediction processing unit 105 reads out the encoded pixel data from the line memory 201c and the MB memory 202d for the pixel block with the block number 14, and generates a prediction image. . At this time, the pixel data for the encoded bottom 4 pixels of block number 12 updated in the line memory 201c and the pixel data for the encoded right end 4 pixels of block number 11 updated in the MB memory 202d are predicted. Used for processing. Then, among the decoded data in the same manner as described above, the pixel data for the lower four pixels is written back to the line memory 201c for intra prediction, and the pixel data for the four rightmost pixels is stored in the MB memory 202d. Write back to

  Finally, in FIG. 10H, the intra prediction processing unit 105 reads out the encoded pixel data from the line memory 201d and the MB memory 202d for the pixel block with the block number 15, and generates a prediction image. . At this time, the pixel data for the encoded bottom 4 pixels of block number 13 updated in the line memory 201d and the pixel data for the encoded right end 4 pixels of block number 14 updated in the MB memory 202d are predicted. Used for processing. Then, among the decoded data in the same manner as described above, the pixel data for the bottom three pixels excluding the right end is written back to the line memory 201d for intra prediction, and the pixel data for the four pixels at the right end is written. Write back to the MB memory 202d. When the pixel block is at the right end of the screen, the pixel data for the four pixels at the lower end is written back to the line memory 201d.

  Here, when the pixel block is not the right end of the screen, as described above, the pixel block of the first block number 0 of the next macroblock is processed and then updated from the MB memory 202 to the line memory 201. This is because the upper left adjacent pixel is referred to in the first pixel block of the next macroblock.

  As described above, the in-screen prediction process in units of 4 × 4 pixel blocks is performed by sequentially updating the line memory 201 and the MB memory 202 in accordance with the position of the pixel block for each prediction process in units of 4 × 4 pixel blocks. Can be done without problems. Also, the line memory 201 is updated with encoded pixel data necessary for intra prediction of the macro block located one level below, and the MB memory 202 is necessary for intra prediction of the right macro block. Encoded pixel data is updated.

FIG. 11 shows an example of the update procedure of the line memory 201 and the MB memory 202 when the intra-screen prediction processing in units of 8 × 8 pixel blocks is performed in the order shown in FIG. 5B in the present embodiment. FIG.
In FIG. 11, the line memory for storing the encoded pixel data in the pixel adjacent on the upper side with respect to the position of the macroblock to be encoded is shown in units of 8 pixels. That is, as shown in FIG. 11, the line memory 201e and the line memory 201f are shown separately. On the other hand, the MB memory for storing the encoded pixels in the pixels adjacent to the left side with respect to the position of the macroblock to be encoded is shown separately in units of 8 pixels. In other words, as shown in FIG. 11, the MB memory 202e and the MB memory 202f are shown separately. The numbers shown in each memory indicate the block numbers to which the stored encoded pixels belong.

  First, in FIG. 11A, the intra-screen prediction processing unit 105 generates a predicted image using the encoded pixel data read from the line memory 201e and the MB memory 202e for the pixel block having the block number 0. To do. Then, the prediction residual generation unit 106 generates a prediction residual signal between the prediction image and the encoding target image. Next, the orthogonal transform unit 107 and the quantization unit 108 perform orthogonal transform and quantization on the prediction residual signal, respectively, and output to the entropy coding unit 109 and the local decoding unit 110.

  The local decoding unit 110 adds the data obtained by inverse quantization and inverse orthogonal transform to the predicted image and decodes the data. Of the decoded data, the pixel data for the bottom 8 pixels is written back to the line memory 201e and the pixel data for the right 8 pixels is written back to the MB memory 202e for intra prediction. In the case of a pixel block located at a position other than the left end of the screen, the pixel data of the 16th pixel in the MB memory 202f is copied to the line memory 201t. Here, the line memory 201t corresponds to a portion storing pixel data of the 16th pixel in the line memory 201f in the macroblock adjacent on the left side.

  Next, in FIG. 11B, the intra-screen prediction processing unit 105 reads out the encoded pixel data from the line memory 201f and the MB memory 202e for the pixel block with the block number 1, and generates a prediction image. . At this time, the pixel data of the encoded right end 8 pixels of the block number 0 updated in the MB memory 202e is used for the prediction process. Then, among the data decoded in the same manner as described above, the pixel data for the bottom 8 pixels is written back to the line memory 201f for intra prediction, and the pixel data for the rightmost 8 pixels is written to the MB memory 202e. Write back.

  Next, in FIG. 11C, the intra-screen prediction processing unit 105 reads the encoded pixel data from the line memory 201e and the MB memory 202f for the pixel block with the block number 2, and generates a predicted image. . At this time, the pixel data for the encoded bottom 8 pixels of block number 0 updated in the line memory 201e is used for the prediction process. Then, among the data decoded in the same manner as described above, the pixel data for the bottom 8 pixels is written back to the line memory 201e for intra prediction, and the pixel data for the right 8 pixels is stored in the MB memory 202f. Write back to

  Finally, in FIG. 11D, the intra-screen prediction processing unit 105 reads the encoded pixel data from the line memory 201f and the MB memory 202f for the pixel block with the block number 3, and generates a prediction image. . At this time, the pixel data for the encoded bottom 8 pixels of block number 1 updated in the line memory 201f and the pixel data for the encoded right end 8 pixels of block number 2 updated in the MB memory 202f are predicted. Used for processing. Then, among the decoded data in the same manner as described above, the pixel data for the bottom 7 pixels excluding the right end is written back to the line memory 201f for intra prediction, and the pixel data for the right end 8 pixels is written. Write back to the MB memory 202f. If the pixel block is the right end of the screen, the pixel data for the bottom 8 pixels is written back to the line memory 201f.

  Here, when the pixel block is not the right end of the screen, as described above, the pixel block of the first block number 0 of the next macroblock is processed and then updated from the MB memory 202 to the line memory 201. This is because the upper left adjacent pixel is referred to in the first pixel block of the next macro block, as in the case of the 4 × 4 pixel block unit.

  As described above, the line memory 201 and the MB memory 202 are sequentially updated according to the position of the pixel block for each prediction process in units of 8 × 8 pixel blocks, so that the in-screen prediction process in units of 8 × 8 pixel blocks is performed. Can be done without problems. Also, the line memory 201 is updated with encoded pixel data necessary for intra prediction of the macro block located one level below, and the MB memory 202 is necessary for intra prediction of the right macro block. Encoded pixel data is updated.

FIG. 12 is a diagram illustrating an example of a procedure for updating the line memory and the MB memory when the intra prediction process is performed in units of 16 × 16 pixel blocks in the present embodiment.
In FIG. 12, the line memory 201g storing the encoded pixel data in the pixel adjacent on the upper side with respect to the position of the macroblock to be encoded is indicated by a line memory 201g in units of 16 pixels. On the other hand, the MB memory 202 storing the encoded pixel data in the pixel adjacent to the left side with respect to the position of the macroblock to be encoded is indicated by an MB memory 202g of 16 pixel units.

  In FIG. 12, the in-screen prediction processing unit 105 generates a predicted image using the encoded pixel data read from the line memory 201g and the MB memory 202g for the pixel block (that is, the macro block) of block number 0. To do. Then, the prediction residual generation unit 106 generates a prediction residual signal between the prediction image and the encoding target image. Next, the orthogonal transform unit 107 and the quantization unit 108 perform orthogonal transform and quantization on the prediction residual signal, respectively, and output to the entropy coding unit 109 and the local decoding unit 110.

  The local decoding unit 110 adds the data obtained by inverse quantization and inverse orthogonal transform to the predicted image and decodes the data. Of the decoded data, pixel data for the lowermost 15 pixels is written back to the line memory 201g and pixel data for the rightmost 16 pixels is written back to the MB memory 202g for intra prediction. If the macroblock is at the right end of the screen, the pixel data for the bottom 16 pixels is written back to the line memory 201g.

  Here, when the macro block is not the right end of the screen, the MB memory 202 is updated to the line memory 201 after the processing of the next macro block as described above. In the case of a macro block located at a position other than the left end of the screen, the pixel data of the 16th pixel in the MB memory 202g is copied to the line memory 201t.

  As described above, even in the case of a 16 × 16 pixel block, the line memory 201 and the MB memory 202 are updated. The line memory 201 updates the encoded pixel data necessary for the intra-frame prediction of the macroblock located immediately below, and the MB memory 202 stores the code necessary for the intra-frame prediction of the right macroblock. The converted pixel data is updated. Further, in the case of the 8 × 8 pixel block in the color difference signal, only the number of pixels is different from that of the 16 × 16 pixel block in the luminance signal, and can be realized by the same processing procedure.

  As described above, in the present embodiment, the first memory 102 includes the line memory 201 that holds pixel data for one line in the horizontal direction of the screen, and the pixels for 16 pixels corresponding to one line in the vertical direction in the macroblock. It has an MB memory 202 for holding data. In the intra prediction, limited memory such as the memory for one line of the screen and the vertical 16 pixels of the macro block is updated in the above-described procedure regardless of the pixel block size. As a result, in-screen prediction processing is performed at high speed in any case of a 4 × 4 pixel block, an 8 × 8 pixel block, a 16 × 16 pixel block in a luminance signal, or an 8 × 8 pixel block in a color difference signal. Can do.

  In this embodiment, since block number 0 refers to the upper left adjacent pixel, after performing intra prediction of the pixel block of block number 0, the pixel data of the 16th pixel of MB memory 202 is stored in line memory 201t. Copied to Although omitted in the description of FIGS. 9 to 11, in the 4 × 4 pixel block and the 8 × 8 pixel block in the luminance signal, there is a possibility of referring to the upper left adjacent pixel even after block number 1.

  For example, in FIG. 9D, when performing intra prediction of the pixel block of block number 3, pixel data corresponding to the upper left adjacent pixel (lower right pixel in the pixel block of block number 0) is stored in the line memories 201a and MB. It is not stored in any of the memories 202a. Therefore, a storage area is separately provided in the first memory 102 for the pixel data corresponding to the encoded upper left adjacent pixel. In this case, erasing is performed at a stage where the pixel is used as a reference pixel, and a memory capacity is secured to a minimum necessary.

  Depending on the block number, after performing in-screen prediction of the pixel block with block number 0, the upper left adjacent pixel is secured in the same manner as the procedure for copying the pixel data of the 16th pixel of the MB memory 202 to the line memory 201t. Also good. For example, in FIG. 9A, after performing intra prediction of block number 0, pixel data for the lower three pixels excluding the right end is written back to the line memory 201a for intra prediction, and the right end four pixels. Minute pixel data is written back to the MB memory 202a. Then, after intra prediction of the pixel block of block number 1 is performed, the pixel data of the fourth pixel in the MB memory 202a may be copied to the line memory 201a.

(Other embodiments according to the present invention)
Each means constituting the video encoding apparatus and each step of the video encoding method in the embodiment of the present invention described above can be realized by operating a program stored in a RAM or ROM of a computer. This program and a computer-readable recording medium recording the program are included in the present invention.

  Further, the present invention can be implemented as, for example, a system, apparatus, method, program, or recording medium. Specifically, the present invention may be applied to a system including a plurality of devices. The present invention may be applied to an apparatus composed of a single device.

  Note that the present invention includes a case where a software program that realizes the functions of the above-described embodiments is supplied directly or remotely to a system or apparatus. This includes the case where the system or the computer of the apparatus is also achieved by reading and executing the supplied program code.

  Accordingly, since the functions of the present invention are implemented by computer, the program code installed in the computer also implements the present invention. In other words, the present invention includes a computer program itself for realizing the functional processing of the present invention.

  In that case, as long as it has the function of a program, it may be in the form of object code, a program executed by an interpreter, script data supplied to the OS, and the like.

  Examples of the recording medium for supplying the program include a flexible disk, a hard disk, an optical disk, and a magneto-optical disk. Further, there are MO, CD-ROM, CD-R, CD-RW, magnetic tape, nonvolatile memory card, ROM, DVD (DVD-ROM, DVD-R) and the like.

  As another program supply method, there is a method of connecting to a homepage on the Internet using a browser of a client computer. The computer program itself of the present invention or a compressed file including an automatic installation function can be downloaded from the homepage by downloading it to a recording medium such as a hard disk.

  It can also be realized by dividing the program code constituting the program of the present invention into a plurality of files and downloading each file from a different homepage. That is, a WWW server that allows a plurality of users to download a program file for realizing the functional processing of the present invention on a computer is also included in the present invention.

  As another method, the program of the present invention is encrypted, stored in a recording medium such as a CD-ROM, distributed to users, and encrypted from a homepage via the Internet to users who have cleared predetermined conditions. Download the key information to be solved. It is also possible to execute the encrypted program by using the key information and install the program on a computer.

  Further, the functions of the above-described embodiments are realized by the computer executing the read program. Furthermore, based on the instructions of the program, an OS or the like running on the computer performs part or all of the actual processing, and the functions of the above-described embodiments can be realized by the processing.

  As another method, the program read from the recording medium is first written in a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer. Then, based on the instructions of the program, the CPU or the like provided in the function expansion board or function expansion unit performs part or all of the actual processing, and the functions of the above-described embodiments are also realized by the processing.

DESCRIPTION OF SYMBOLS 100 Video coding apparatus 101 Prediction method determination part 102 1st memory 103 2nd memory 104 Inter prediction process part 105 Intra-screen prediction process part 106 Prediction residual production | generation part 107 Orthogonal transformation part 108 Quantization part 109 Entropy encoding Unit 110 local decoding unit 111 deblocking filter processing unit 201 line memory 202 MB memory

Claims (6)

A video encoding device that performs encoding by dividing an encoding target image into units of pixel blocks each having a predetermined number of pixels,
In-screen prediction processing means for performing in-screen prediction processing using adjacent pixel data stored in a memory in units of the pixel block;
Generating means for generating a prediction residual indicating a difference between the encoding target image and the prediction image obtained by the intra-screen prediction processing;
Encoding means for performing encoding after orthogonal transformation of the prediction residual;
Decoding means for inversely transforming the data after the orthogonal transformation and decoding the unit of the pixel block;
The pixel data for one horizontal line at the lower end of the pixel block decoded by the decoding means and the pixel data for one line in the vertical direction at the right end of the pixel block adjacent to the decoded pixel block. Storage means for storing in the memory for use in in-screen prediction processing,
The memory includes at least a first storage area for storing pixel data for one line in the horizontal direction of the encoding target image, and a second storage area for storing pixel data for one line in the vertical direction of the pixel block. Have
The storage means refers to the pixel data of the rightmost pixel in one horizontal line at the lower end of the pixel block decoded by the decoding means, with reference to the pixel data stored in the first storage area on the right side. A video encoding apparatus, wherein an intra-screen prediction process of a pixel block is performed and then stored in the first storage area.   The storage means copies the pixel data corresponding to the position of the lower end among the pixel data for one line in the vertical direction at the right end of the decoded pixel block stored in the second storage area, The video encoding apparatus according to claim 1, wherein pixel data of a rightmost pixel is stored in the first storage area. A video encoding method for performing encoding by dividing an encoding target image into units of pixel blocks each having a predetermined number of pixels,
In a unit of the pixel block, a first storage area that stores at least one line of pixel data in the horizontal direction of the encoding target image, and a second storage area that stores pixel data of one line in the vertical direction of the pixel block. An intra-screen prediction processing step for performing intra-screen prediction processing using adjacent pixel data stored in a memory having a storage area; and
Generating a prediction residual indicating a difference between the encoding target image and the prediction image obtained by the intra-screen prediction process;
An encoding step of encoding the prediction residual after orthogonal transformation;
A decoding step of inversely transforming the data after the orthogonal transformation and decoding the unit of the pixel block;
The pixel data for one horizontal line at the bottom of the pixel block decoded in the decoding step and the pixel data for one line in the vertical direction at the right end of the pixel block adjacent to the decoded pixel block A storage step of storing in the memory for use in the in-screen prediction process,
In the storing step, the pixel data of the rightmost pixel in one horizontal line at the lower end of the pixel block decoded in the decoding step is adjacent to the right side referring to the pixel data stored in the first storage area. A video encoding method comprising: storing in the first storage area after intra-screen prediction processing of the pixel block is performed.   In the storing step, by copying pixel data corresponding to the position of the lower end among the pixel data for one line in the vertical direction at the right end of the decoded pixel block stored in the second storage area, 4. The video encoding method according to claim 3, wherein pixel data of the rightmost pixel is stored in the first storage area. A program for controlling a video encoding device that performs encoding by dividing an encoding target image into units of pixel blocks each having a predetermined number of pixels,
In a unit of the pixel block, a first storage area that stores at least one line of pixel data in the horizontal direction of the encoding target image, and a second storage area that stores pixel data of one line in the vertical direction of the pixel block. An intra-screen prediction processing step for performing intra-screen prediction processing using adjacent pixel data stored in a memory having a storage area; and
Generating a prediction residual indicating a difference between the encoding target image and the prediction image obtained by the intra-screen prediction process;
An encoding step of encoding the prediction residual after orthogonal transformation;
A decoding step of inversely transforming the data after the orthogonal transformation and decoding the unit of the pixel block;
The pixel data for one horizontal line at the bottom of the pixel block decoded in the decoding step and the pixel data for one line in the vertical direction at the right end of the pixel block adjacent to the decoded pixel block A storage step of storing in the memory for use in the in-screen prediction process;
In the storing step, the pixel data of the rightmost pixel in one horizontal line at the lower end of the pixel block decoded in the decoding step is adjacent to the right side referring to the pixel data stored in the first storage area. A program that is stored in the first storage area after intra-screen prediction processing of the pixel block is performed.   6. A computer-readable recording medium on which the program according to claim 5 is recorded.

Images