JP2015146474A - Encoding device and decoding device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To employ irreversible compression in storage of prediction information in memory to reduce the required capacity of the memory, and accelerate decoding processing in a decoding device.SOLUTION: In an encoding device 1 including picture memory 17 for prediction reference in order to perform prediction encoding to moving image information including a plurality of temporally continuous frames, when an irreversible compression processing unit 16 stores pieces of moving image data for prediction in the picture memory through compression encoding, determines a quantization coefficient for each of the frames so that the pieces of moving data can be stored without fail in a storage area for moving images that is secured in advance in the memory in the decoding device, and outputs as quantization coefficients QC; and a multiplexing unit 22 multiplexes the encoded image data and the quantization coefficients QC and outputs for the decoding device.

Description

  The present invention relates to an encoding device and a decoding device, and in particular, when encoding and decoding moving image information using predictive encoding, for a memory storing image information for prediction reference, The present invention relates to an encoding device and a decoding device that store the image information after compression processing.

  In the encoding process of moving picture information composed of a plurality of temporally continuous frames, a prediction code including motion compensation is used to increase the encoding efficiency by using the characteristic of similarity of images close in time. It is very common sense to make it. In order to realize such inter-frame predictive encoding, the encoded information of the already encoded frame is returned by decoding so that the information of the already encoded frame can be used for encoding of another frame. Thus, a memory (storage means) for temporarily storing the data is required. The same applies to the decoding device side, and the decoding information of the already decoded frame is temporarily stored in the memory so that the information of the already decoded frame can be used for decoding another frame. Keep it.

  Recently, for example, “4K”, moving images have become extremely fine, and the amount of information constituting one frame has become extremely large. Accordingly, the memory capacity as described above must be increased accordingly.

  On the other hand, particularly in the display of moving images in a gaming machine such as a pachinko machine, there is a process in which a plurality of moving image streams are superimposed on one screen or a plurality of screens, or are displayed simultaneously in parallel. Often done. Even in such a case, in the encoding apparatus (encoding process) and decoding apparatus described above, it is necessary to store necessary frame information in parallel for each moving picture. Therefore, from this point of view, it is necessary to devise how to use the limited capacity of the memory for predictive coding. In particular, it is highly necessary in a typical decoding apparatus that is configured by hardware to perform high-speed real-time processing.

  From the above situation, in recent years, the information is compressed and stored in a memory for storing information of a frame necessary for predictive coding, and when used, it is read and decompressed before use. Such a method is also adopted. Although the number of necessary processes or circuits increases, this is more meaningful from the viewpoint of reducing or effectively using the memory capacity. Such a technique of compressing frame information necessary for predictive encoding and storing it in a memory is disclosed in Patent Document 1, for example. This document discloses a decoding device. For example, as shown in FIG. 1, the frame memory 102 for storing information of a frame used for predictive encoding includes an encoding circuit 101. The information compressed and encoded by is stored, and when it is used for prediction in the predicted image generation circuit 26, it is expanded and supplied by the second data expansion circuit 105b. The circuit that realizes the specific device shown in FIGS. 13 and 16 is the same as the base.

JP-A-9-261635

  By the way, in Patent Document 1, lossy compression is employed as a compression technique for storing frame information necessary for predictive coding in a memory (an orthogonal transform circuit 201 and a quantizer 202 in FIG. 2 and the like). ). The disadvantage of adopting this lossy compression is that when prediction between frames is cumulatively performed only by the decoding device, the distortion of the image also increases cumulatively accordingly. Therefore, when priority is given to a certain level of image quality, lossless compression can be adopted as the compression technique.

  However, when lossless compression is adopted as a compression technique, considering the capacity of the memory of the decoding device, confirming how much capacity each frame constituting each video stream will occupy after compression processing The design must be done. On the other hand, when irreversible compression is adopted as a compression technique, even if the memory capacity is limited, it is always possible to store it by increasing the compression rate as necessary by selecting the quantization coefficient. In addition, the selection of the quantization coefficient does not have to be performed in advance when designing each moving image, and can be determined when the memory is stored.

  Such a technique is disclosed in Cited Document 1 and proposes a contrivance to be performed according to the memory capacity when the decoding apparatus stores the prediction information in the memory for lossy compression. As one example, referring to FIG. 13 and FIG. 16 of this document, a trial calculation circuit 1201 is provided that performs trial calculation of the generated code amount with a small quantization table, and the quantizer 202 performs quantization by reflecting the trial calculation result. Thus, control is performed so that the compressed data is stored in the frame memory 102. For example, according to such a device, each moving image can be designed without worrying about the memory capacity.

  However, in the cited document 1, a special circuit is added in the decoding device for realizing a process that is sequentially executed in the course of the decoding process. Thus, although there is no concern about the memory capacity, it is necessary to repeat the compression process while changing the quantization coefficient until it fits in the memory capacity, and this process has a drawback that the entire process becomes heavy. In particular, in an application for a gaming machine that simultaneously reproduces a large amount of moving images that have been encoded in advance, it is important to be able to decode as fast as possible.

  The present invention has been made for the above-mentioned circumstances, and the object of the present invention is based on the premise that the required capacity of the memory is reduced by adopting irreversible compression for storing the prediction information in the memory. In addition, an object of the present invention is to provide an encoding device and a decoding device that can increase the speed without making the decoding processing in the decoding device unnecessarily heavy.

  In order to achieve the above object, the first encoding device of the present invention uses a prediction reference for performing inter-frame predictive encoding on moving picture information composed of a plurality of temporally continuous frames. A memory for storing image data, an irreversible compression processing unit that performs predetermined irreversible compression processing on the image data in order to store the image data in the memory, and compressed image data from the memory And a decompression processing unit that performs decompression processing corresponding to the irreversible compression processing, and inputs a single video or a plurality of videos to be processed in parallel to perform predictive coding processing. The irreversible compression processing unit corresponds to the irreversible compression processing unit so that each moving image data is surely contained in a region for each moving image secured in advance in a memory on the decoding device side corresponding to the memory. Decryption device The quantized coefficients for quantization in lossy compression processing unit and outputs the determined for each frame, its respective quantized coefficient is summarized in that output together with the image data after compression coding.

  Here, specifically, the lossy compression processing unit calculates a compression rate for each frame based on encoded image data obtained by compression encoding based on a set quantization coefficient, A compression rate calculation determination unit that determines whether or not the encoded image data fits in a moving image area including the frame, which is secured in advance in a memory on the decoding device side, based on a compression rate; When it is determined that the compression rate calculation determination unit does not fit, the quantization coefficient is sequentially updated in the direction of high information amount reduction, and compression encoding is repeated.

  Alternatively, the lossy compression processing unit calculates a compression rate for each frame based on encoded image data obtained by compression encoding based on a quantization coefficient, and based on the compression rate, the encoded image data Is determined in advance while updating the quantization coefficient to determine whether or not the image can be stored in the moving image area including the frame, which is secured in advance in the memory on the decoding device side. In doing so, the quantization coefficient is updated based on the binary search method.

  At this time, each quantization coefficient is preferably stored in an option header for each frame in an output format in which encoded image data to be output is arranged in a predetermined manner.

  In order to achieve the above object, the first decoding apparatus of the present invention inputs encoded image data output from the first encoding apparatus and supports predictive encoding processing in the encoding apparatus. A decoding device that performs the decoding processing, comprising: an irreversible compression processing unit and an expansion processing unit corresponding to the irreversible compression processing unit and the decompression processing unit in the encoding device, The gist is to perform quantization processing or inverse quantization processing for each frame based on the quantization coefficient for each frame input together with image data.

  In order to achieve the above object, the second encoding apparatus of the present invention performs prediction reference to perform interframe predictive encoding on moving image information composed of a plurality of temporally continuous frames. A memory for storing image data for use, a lossy compression processing unit that performs predetermined irreversible compression processing on the image data to store the image data in the memory, and compression from the memory A decompression processing unit that inputs image data and performs decompression processing corresponding to the lossy compression processing, and inputs predictive coding processing by inputting a single moving image or a plurality of moving images that are simultaneously processed in parallel by the encoding device. The irreversible compression processing unit corresponds to the irreversible compression processing unit so that each moving image data is surely contained in a region for each moving image secured in advance in a memory on the decoding device corresponding to the memory. Said decryption The quantization coefficient for quantization in the irreversible compression processing unit on the device side is determined and output for each block constituting the frame, and each quantization coefficient is output together with the image data after compression coding. Is the gist.

  In order to achieve the above object, the second decoding apparatus of the present invention inputs encoded image data output from the second encoding apparatus and supports predictive encoding processing in the encoding apparatus. A decoding device that performs the decoding processing, comprising: an irreversible compression processing unit and an expansion processing unit corresponding to the irreversible compression processing unit and the decompression processing unit in the encoding device, The gist is to perform quantization processing or inverse quantization processing for each block based on the quantization coefficient for each block constituting the frame input together with the image data.

  According to the encoding device and the decoding device of the present invention, decoding is performed on the assumption that lossy compression is adopted for storing the prediction information in the memory and the restriction on the capacity of the memory is not considered. It is possible to avoid unnecessarily increasing the decryption processing in the conversion apparatus. That is, when storing moving image data by irreversible compression encoding in a memory for prediction, the quantization coefficient is set for each frame so that each moving image data is surely stored in a storage area for each moving image reserved in advance in the memory. Since the determination process is performed on the encoding device side, an extra burden is not imposed on the decoding device.

It is a figure which shows the structure of one Embodiment of the encoding apparatus in this invention. It is a figure which shows the detailed structure of the lossy compression process part 16 and corresponding expansion | extension process part 18 which were shown in FIG. 4 is a flowchart illustrating a processing procedure performed by the irreversible compression processing unit 16. It is a figure which shows the structure of one Embodiment of the decoding apparatus in this invention. FIG. 5 is a diagram illustrating detailed configurations of an irreversible compression processing unit 35 and a corresponding decompression processing unit 37 illustrated in FIG. 4.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a diagram showing a configuration of an embodiment of an encoding apparatus according to the present invention. The encoding device 1 shown in the figure includes a subtracting unit 11, a compression processing unit 12, a first variable length encoding unit 13, an expansion processing unit 14, an adding unit 15, an irreversible compression processing unit 16, A picture memory 17, a corresponding decompression processing unit 18, a motion compensation prediction unit 19, a second variable length coding unit 21, and a multiplexing unit 22 are provided. Note that the encoding apparatus 1 shown in FIG. 1 inputs moving image information composed of a plurality of temporally continuous frames in units of blocks, for example, H.264. H.264 (MPEG (Moving Picture Experts Group) -4 AVC).

  The subtraction unit 11 subtracts the picture information of the macroblock sent from the motion compensation prediction unit 19 from the encoding target macroblock of the picture as the encoding target frame, and sends it to the compression processing unit 12 as a residual prediction signal. Supply. Note that the encoding target macroblock related to the I picture in the case of MPEG is directly sent to the compression processing unit 12 without passing through the subtraction unit 11.

  The compression processing unit 12 performs irreversible compression processing on the residual prediction signal of the target macroblock sent from the subtraction unit 11. Here, as the irreversible compression processing, there is an algorithm that performs orthogonal transformation processing and quantization processing like JPEG (Joint Photographic Experts Group) in a still image. Note that orthogonal transform processing includes discrete cosine transform (DCT) and Hadamard transform. Also, wavelet transform may be used as an alternative to orthogonal transform processing.

  The first variable length coding unit 13 performs lossless compression on the compressed image data output from the compression processing unit 12 by variable length coding such as Huffman coding or arithmetic coding for each block.

  Further, the decompression processing unit 14 performs irreversible decompression processing corresponding to the irreversible compression processing in the compression processing unit 12 on the compressed image data in units of macroblocks output from the compression processing unit 12, and is obtained by decompression. The macroblock is sent to the adder 15. The adding unit 15 adds the information related to the macroblock output from the decompression processing unit 14 to the information related to the preceding most highly correlated macroblock output from the motion compensation prediction unit 61, thereby performing subsequent prediction. Restore the provided macroblock. As described above, the macroblock related to the I picture in the case of MPEG does not go through the adding unit 15.

  The irreversible compression processing unit 16 performs predetermined irreversible compression processing on the image information output from the addition unit 15 or not passed through the addition unit 15, and compares the obtained compressed image data with the subsequent processing Is stored in the picture memory 17. In addition, the irreversible compression processing unit 16 determines the quantization coefficient for each frame so that the capacity of the picture memory area reserved for the processed moving image is scheduled on the decoding device side, which will be described later, and outputs it. To do. However, as a condition, it goes without saying that the lossy compression processing technique here is consistent with the technique on the decoding apparatus side, and on the encoding apparatus 1 side, the picture memory on the decoding apparatus side The amount of space reserved for each video in must be known. This capacity is preferably fixed to a predetermined ratio of capacity, such as one third of the capacity of each video before compression, in order to simplify management in the decoding apparatus.

  FIG. 2A is a diagram illustrating a detailed configuration of the lossy compression processing unit 16. The irreversible compression processing unit 16 shown in FIG. 5A is an orthogonal transform unit 161 that performs a predetermined orthogonal transform process on the image data input from the adder unit 15, and the data after the orthogonal transform, A quantization unit 162 that appropriately selects a quantization coefficient from the quantization table 164 and performs quantization processing, and a variable that performs variable-length coding processing such as Huffman coding and arithmetic coding on the quantized data. A long encoding unit 163; and a compression rate calculation determining unit 165 that calculates a compression rate for each frame of image data after variable length encoding and determines whether or not the compression rate is equal to or less than a predetermined compression rate. Yes.

  FIG. 3 is a flowchart illustrating a processing procedure performed by the lossy compression processing unit 16. With reference to FIG. 2A and FIG. 3, the operation of the lossy compression processing unit 16 will be described. First, the quantization unit 162 selects a quantization coefficient having a sufficiently small quantization effect from the quantization table 164 as an initial quantization coefficient (step S1). Then, for each block, the orthogonal transform unit 161 performs a predetermined orthogonal transform process (step S2), the quantization unit 162 performs a quantization process (step S3), and the variable length coding unit 163 performs a variable length coding process. (Step S4). Then, it is determined whether or not one frame has been completed (step S5). If it has not been completed (negative determination), steps S2 to S4 are repeated. On the other hand, when the processing is completed (affirmative determination), the compression rate calculation determination unit 165 calculates the compression rate (step S6), and a predetermined amount corresponding to the memory capacity reserved for the moving image in the decoding device. It is determined whether or not the compression rate is equal to or lower than the compression rate (step S7). In this determination, if the desired compression rate has not been reached (negative determination), the process proceeds to step S8, and the quantization unit 162 selects a new quantization coefficient (one step higher than the quantization effect) from the quantization table 164. (Step S8), and each unit repeats Steps S2 to S6. On the other hand, if the desired compression rate is reached in the determination in step S7 (positive determination), the quantization unit 162 outputs the current quantization coefficient (step S9).

  On the other hand, the motion compensated prediction unit 19 sequentially inputs each macroblock constituting the picture as the encoding target frame, and the macro having the highest correlation from the macroblocks of the picture related to the prediction frame stored in the picture memory 17. Search for a block. The corresponding decompression processing unit 18 corresponding to the processing of the irreversible compression processing unit 16 decompresses data corresponding to the macroblock retrieved by the motion compensation prediction unit 19 in correspondence with the lossless compression processing in the irreversible compression processing unit 16. Processing is performed, and the result is supplied to the motion compensation prediction unit 19.

  FIG. 2B is a diagram illustrating a detailed configuration of the corresponding decompression processing unit 18. The corresponding decompression processing unit 18 shown in the figure performs variable length decoding processing that performs variable length decoding processing corresponding to variable length coding in the variable length coding unit 233 on the compressed image data read from the picture memory 17. Based on the decoding unit 181 and the quantization table 184, an inverse quantization unit 182 that performs an inverse quantization process corresponding to the quantization in the quantization unit 162, and an inverse orthogonal transform process corresponding to the orthogonal transform in the orthogonal transform unit 161 And an inverse orthogonal transform unit 183 that performs. Note that, similarly to the output of the quantization coefficient to the decoding device 3, the inverse quantization unit 182 must be notified or recognized of the quantization coefficient related to the stored image data.

  The motion compensation prediction unit 19 supplies the information related to the macroblock with the highest correlation obtained in this way to the subtraction unit 11 and the addition unit 15 as described above. Although not shown, the motion compensation prediction unit 19 outputs the amount of movement as a motion vector when there is a part that moves integrally as a certain fixed area between two macroblocks for comparison. . The output motion vector is similarly subjected to variable length coding and supplied to the multiplexing unit 22.

  The multiplexing unit 22 encodes encoded image data from the first variable length encoding unit 13, information on the selected quantization coefficient QC for each frame from the second variable length encoding unit 21, and other motion vectors and the like. Are multiplexed and output. At this time, there are several possible methods for outputting the information of the quantization coefficient QC together with the encoded image data. It is possible to embed in an optional header of a file format such as MP4 that can store code data based on the H.264 standard. That is, in this case, it may be incorporated into SEI (Supplemental Enhancement Information) which is an option header for each frame. Of course, it can also be output as an independent file separate from the image data file.

  In the encoding apparatus as one embodiment described above, as a search method for the compression rate that fits in the memory capacity, the linear search method sequentially determines from the lower compression rate to the higher compression rate. However, the binary search method can be adopted more efficiently.

  Next, the decoding device side will be described. FIG. 4 is a diagram showing a configuration of an embodiment of a decoding apparatus according to the present invention. The decoding device 3 shown in the figure includes a separation unit 31, a first variable length decoding unit 32, an expansion processing unit 33, an addition unit 34, an irreversible compression processing unit 35, a picture memory 36, A corresponding expansion processing unit 37, a motion compensation prediction unit 38, and a second variable length decoding unit 39 are provided.

  The separation unit 31 separates the encoded data that is sent into encoded image data, encoded information related to the quantization coefficient, a motion vector, and the like. The first variable length decoding unit 32 to which the encoded image data is input performs a variable length decoding process corresponding to the variable length encoding process by the first variable length encoding unit 13 shown in FIG. Do. Thus, for example, Huffman decoding or arithmetic decoding. Similarly, the second variable length decoding unit 39 performs a variable length decoding process on the encoded information related to the quantization coefficient to reproduce the quantization coefficient QC for each frame. Also, a motion vector (not shown) is subjected to variable length decoding processing and supplied to the motion compensation prediction unit 38.

  The decompression processing unit 33 to which the image data from the first variable length decoding unit 32 is input performs an irreversible decompression process corresponding to the irreversible compression process in the compression processing unit 12 of the encoding device 1 shown in FIG. The residual prediction signal obtained by such processing is sent to the addition unit 34 and the motion compensation prediction unit 38.

  The motion compensation prediction unit 38 processes the macroblock data of the preceding image data stored in the picture memory 36 using the residual prediction signal from the decompression processing unit 33 and the motion vector, and sends the processed data to the addition unit 34. . The adding unit 34 adds the residual prediction signal from the decompression processing unit 33 and the macroblock data from the motion compensation prediction unit 38, and outputs decoded image data in macroblock units.

  The image information output from the adder 34 is output by the irreversible compression processor 35 that employs the same irreversible compression technique as the irreversible compression processor 16 shown in FIG. An irreversible compression process is performed, and the obtained compressed image data is stored in the picture memory 36 for comparison with the subsequent process.

  Here, the processing in the lossy compression processing unit 35 will be described in detail with reference to FIG. The figure shows the detailed configuration of the lossy compression processing unit 35. The irreversible compression processing unit 35 shown in the figure includes an orthogonal transformation unit 351 that performs a predetermined orthogonal transformation process on the image data input from the addition unit 15, and a quantization table for the data after the orthogonal transformation. A quantization unit 352 that selects an appropriate quantization coefficient from 354 and performs quantization processing, and a variable-length code that performs variable-length coding processing such as Huffman coding and arithmetic coding on the quantized data And a conversion unit 353. However, in the present invention, the quantization unit 352 receives each quantization coefficient applied to each frame from the second variable length decoding unit 39, and based on the corresponding quantization coefficient for each frame. To quantize. As a result, the compressed data obtained as a result is appropriately stored in the moving image area secured in the picture memory 36.

  The motion compensation prediction unit 38 sequentially inputs each macro block obtained by the expansion processing by the expansion processing unit 33, and the macro block having the highest correlation from the macro blocks of the picture related to the prediction frame stored in the picture memory 36. Search for. A corresponding decompression processing unit 37 that employs a decompression technique similar to that of the corresponding decompression processing unit 18 illustrated in FIG. 1 performs irreversible compression processing unit 35 on the data related to the macroblock retrieved by motion compensation prediction unit 38. The decompression process corresponding to the irreversible compression process is performed, and the result is supplied to the motion compensation prediction unit 19.

  Here, the processing in the corresponding decompression processing unit 37 will be described in detail with reference to FIG. The figure shows the detailed configuration of the corresponding decompression processing unit 37. The corresponding decompression processing unit 37 shown in the figure performs variable length decoding processing that performs variable length decoding processing corresponding to variable length coding in the variable length coding unit 353 on the compressed image data read from the picture memory 36. Based on the decoding unit 371 and the quantization table 374, an inverse quantization unit 372 that performs an inverse quantization process corresponding to the quantization in the quantization unit 352, and an inverse orthogonal transform process corresponding to the orthogonal transform in the orthogonal transform unit 351 And an inverse orthogonal transform unit 373 for performing. However, in the present invention, the inverse quantization unit 372 receives each quantization coefficient applied for each frame from the second variable length decoding unit 39, and sets the corresponding quantization coefficient for each frame. Based on this, inverse quantization processing is performed.

  In the above description, the corresponding decompression processing unit 37 directly receives each quantization coefficient applied for each frame from the second variable length decoding unit 39, but instead from the lossy compression processing unit 35. You may receive it.

  As described above, according to an embodiment of the encoding device and the decoding device of the present invention, when moving image data is stored in a picture memory by compression encoding for prediction, each moving image previously secured in the memory is stored. Since the encoding device 1 performs the process of determining the quantization coefficient for each frame so that each moving image data is surely contained in the storage area for use, there is no additional burden on the decoding device. That is, the decoding device 3 receives each quantized coefficient determined for each frame constituting the moving image from the encoding device 1 and performs the compression encoding based on the received quantization coefficient.

<Other embodiments>
In the embodiment shown in FIG. 1 to FIG. 5, the encoding device 1 determines and outputs a quantization coefficient in units of frames, and the lossy compression processing unit 35 of the decoding device 3 receives the quantization that has been sent. Although compression processing is performed in units of frames using coefficients, if quantization is performed in units of blocks as shown in FIG. 3, the encoding device 1 determines and outputs quantization coefficients in units of blocks and decodes them. The irreversible compression processing unit 35 of the encoding device 3 can also perform compression processing in units of blocks using the sent quantization coefficients. Also, processing can be performed in units of slices constituting a frame. As for the method of sending the quantization coefficient in these cases, a plurality of quantization coefficients relating to each block or slice may be stored in the SEI which is an option header for each frame, and independent of the image data file. It may be a file. As a result, the processing can be performed more delicately than in the embodiment described above.

  INDUSTRIAL APPLICABILITY The encoding device and decoding device of the present invention can be used for processing moving images displayed on a gaming machine such as a pachinko machine or a game machine.

DESCRIPTION OF SYMBOLS 1 Encoding apparatus 11 Subtraction part 12 Compression process part 13 1st variable length encoding part 14 Decompression process part 15 Addition part 16 Lossy compression process part 161 Orthogonal transformation part 162 Quantization part 163 Variable length encoding part 164 Quantization table 165 Compression rate calculation determination unit 17 Picture memory 18 Corresponding decompression processing unit 19 Motion compensation prediction unit 21 Second variable length coding unit 22 Multiplexing unit 3 Decoding device 31 Separation unit 32 First variable length decoding unit 33 Decompression processing Unit 34 addition unit 35 lossy compression processing unit 36 picture memory 37 corresponding decompression processing unit 38 motion compensation prediction unit 39 second variable length decoding unit

Claims (7)

A memory for storing image data for prediction reference in order to perform predictive encoding between frames for moving image information composed of a plurality of temporally continuous frames;
An irreversible compression processing unit that performs predetermined irreversible compression processing on the image data in order to store the image data in the memory;
A decompression processing unit that inputs compressed image data from the memory and performs decompression processing corresponding to the lossy compression processing;
An input device that inputs one moving image or a plurality of moving images that are processed in parallel, and performs predictive coding processing,
The irreversible compression processing unit includes the decoding unit corresponding to the irreversible compression processing unit so that each moving image data is surely stored in a region for each moving image that is secured in advance in a memory on a decoding device corresponding to the memory. The quantization coefficient for quantization in the irreversible compression processing unit on the encoding device side is determined and output for each frame, and each quantization coefficient is output together with the image data after compression encoding. Encoding device.   The irreversible compression processing unit calculates a compression rate for each frame based on encoded image data obtained by compression encoding based on the updated quantization coefficient, and based on the compression rate, the encoded image A compression rate calculation determination unit that determines whether or not the data fits in a moving image area including the frame, which is secured in advance in the memory on the decoding device side, and the compression rate calculation determination unit does not 2. The encoding apparatus according to claim 1, wherein if it is determined that there is not, the quantization coefficient is sequentially updated in a direction in which the amount of information is reduced, and compression encoding is repeated.   The lossy compression processing unit calculates a compression rate for each frame based on encoded image data obtained by compression encoding based on a quantization coefficient, and based on the compression rate, the encoded image data is When determining whether or not to fit within the moving image area including the frame, which is secured in advance in the memory on the decoding device side, while updating the quantization coefficient, The encoding apparatus according to claim 1, wherein the quantization coefficient is updated based on a binary search method.   4. Each of the quantization coefficients is stored in an option header for each frame in an output format in which encoded image data to be output is arranged in a predetermined manner. Encoding device. A decoding device that inputs encoded image data output from the encoding device according to claim 1 and performs a decoding process corresponding to a predictive encoding process in the encoding device,
The irreversible compression processing unit and the decompression processing unit corresponding to the irreversible compression processing unit and the decompression processing unit in the encoding device, both of which are input together with the encoded image data, and the quantization for each frame A decoding apparatus characterized by performing a quantization process or an inverse quantization process for each frame based on a coefficient. A memory for storing image data for prediction reference in order to perform predictive encoding between frames for moving image information composed of a plurality of temporally continuous frames;
An irreversible compression processing unit that performs predetermined irreversible compression processing on the image data in order to store the image data in the memory;
A decompression processing unit that inputs compressed image data from the memory and performs decompression processing corresponding to the lossy compression processing;
An input device that inputs one moving image or a plurality of moving images that are processed in parallel, and performs predictive coding processing,
The irreversible compression processing unit includes the decoding unit corresponding to the irreversible compression processing unit so that each moving image data is surely contained in a region for each moving image that is secured in advance in a memory on the decoding device side corresponding to the memory. Quantization coefficients for quantization in the irreversible compression processing unit on the encoding device side are determined and output for each block constituting the frame, and each quantization coefficient is output together with the image data after compression encoding. An encoding apparatus characterized by that. A decoding device that inputs encoded image data output from the encoding device according to claim 6 and performs a decoding process corresponding to a predictive encoding process in the encoding device,
A block comprising the irreversible compression processing unit and the decompression processing unit corresponding to the irreversible compression processing unit and the decompression processing unit in the encoding device, both of which are input together with the encoded image data and constitute the frame A decoding apparatus characterized by performing a quantization process or an inverse quantization process for each block based on a quantization coefficient for each block.

Images