JP2004032300A - Video compression encoding device - Google Patents

Abstract

Conventionally, it has been difficult to provide a moving picture compression encoding apparatus capable of controlling an encoded stream at a constant transfer rate without complicating the configuration.
A means for dividing each frame of video data into blocks and detecting a predicted motion difference between a current frame and a previous frame for each block, a result of the predicted motion difference or an original image Means 203 and 204 for performing variable-length coding by performing orthogonal transform and quantization specified for the data, and performing inverse quantization, inverse orthogonal transform, and inverse quantization specified for the variable-length coded data. Means 209, 210, 211 for performing a motion compensation process to generate a decoded image; coding terminating means 2071 for terminating the encoding of each block during the variable length encoding process; In response to the notification, there is provided a frame management unit 213 that controls so as to perform encoding without using motion prediction in encoding of the next frame image. To.
[Selection diagram] FIG.

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a moving image compression encoding apparatus, and more particularly, to a technique for controlling a generated code amount in an encoding apparatus that performs digital compression encoding on a video signal.
[0002]
2. Description of the Related Art In recent years, research and development of multimedia systems that integrally handle various types of data such as moving images, still images, sounds, and texts have been actively conducted in various fields. By the way, a video signal is one of important information sources, but it is necessary to efficiently compress a huge amount of data, and international standardization is progressing. ISO-IEC 14496-2 describes the syntax of a visual coded stream and its decoding method as an international standard for compression coding of video signals, and is deliberated by MPEG-4 (International Standardization Organization (ISO)). Of the MPEG (Moving Picture Experts Group), it is called Phase 4), which aims at a standardization technology for multimedia.
[0003]
Usually, in an application that handles a video having a small screen size at a low bit rate, a standard called a simple profile of MPEG-4 is adopted. In the simple profile, image coding in which each frame image that is a component of a moving image is compression-coded only from information of a current frame image called an I-VOP (Intra-Video Object Plane). And a picture coding type for compressing and coding a motion difference from the current frame picture using a previous frame picture called a P-VOP (inter-frame forward predictive coding VOP (Predicted Video Object Plane)) as a prediction value. Are defined.
[0004]
Generally, in moving picture coding, P-VOP coding of a plurality of frame images is continuously performed with respect to I-VOP coding of one frame image in order to improve coding efficiency based on video motion. Is done. Further, in an encoding device (moving image compression encoding device) that performs such encoding, the bandwidth of a communication path for transmitting an encoded video stream is limited, and thus the encoded stream has a constant transfer rate. Need to be controlled. Therefore, there is a demand for a moving image compression encoding apparatus capable of controlling an encoded stream at a constant transfer rate without complicating the configuration.
[0005]
[Prior art]
FIG. 1 is a block diagram showing an example of a conventional moving image compression encoding apparatus, and shows a configuration of an encoding apparatus used for MPEG-4 compression technology. In FIG. 1, reference numeral 101 denotes a current image memory, 102 denotes a motion difference detection unit, 103 denotes a DCT unit, 104 denotes a quantization unit, 105 denotes a DC / AC prediction unit, 106 denotes a rate control unit, and 107 denotes a variable length. 2 shows an encoding unit. Reference numeral 108 denotes a generated code buffer, 109 denotes an inverse quantization unit, 110 denotes an IDCT unit, 111 denotes a motion difference compensating unit, and 112 denotes a decoded image memory.
[0006]
The current image memory 101 is a memory for storing image data obtained by digitizing one frame of a video signal. Here, in MPEG-4, one frame image is handled by being divided into processing units called macroblocks of 16 × 16 pixels, and one macroblock is composed of four luminance signal blocks of 8 × 8 pixels. The color difference signal block of 8 × 8 pixels is divided into two blocks, that is, six blocks, and processed. Note that the color difference signal is processed at half the resolution in the horizontal and vertical directions as compared to the luminance signal.
[0007]
The motion difference detection unit 102 receives, from the current image memory 101, each block pixel data of the current image coded in the P-VOP type, and inputs the block pixel data of the previous frame stored in the decoded image memory 112. A motion vector that minimizes the motion difference value from the decoded image is detected, and the motion difference value is output to DCT section 103. Here, the motion vector is a two-dimensional vector indicating the displacement of the position between the current frame image and the immediately preceding frame image (previous frame image) in the horizontal and vertical directions, and the motion difference value is the current image. This is obtained by subtracting the corresponding pixel data of the 8 × 8 block of the previous frame image displaced by the motion vector from each pixel data of the block. Generally, the minimum motion difference value means that the sum of the absolute difference values of the pixels is minimum.
[0008]
For the I-VOP type frame image, the motion difference detection processing in the motion difference detection unit 102 is not performed, and the block pixel data read from the current image memory 101 is output to the DCT unit 103 as it is. Further, according to the MPEG-4 specification, it is possible to specify not to perform the motion difference detection on the macroblock in the P-VOP. This is effective for a block in which the motion between frame images is severe and the motion difference between the motion prediction block and the current image block is large. Note that a macroblock on which motion prediction is not performed is called an intra macroblock, and a macroblock on which motion prediction is performed is called an inter macroblock.
[0009]
FIG. 2 is a diagram for explaining processing in the motion difference detection unit. FIG. 2A shows a video sequence, and FIG. 2B shows block matching processing.
[0010]
As shown in FIG. 2A, image data of a plurality of frames is input to the moving image compression encoding apparatus with time, and is processed by a macroblock MB of 16 × 16 pixels as described above, for example. Is done. In FIG. 2A, reference numeral CF indicates a current frame (current frame), RF indicates an immediately preceding frame (reference frame), and SW indicates a search window (search area) in the immediately preceding frame RF. I have.
[0011]
As shown in FIG. 2B, in the macroblock MB of the current frame CF, in the search window SW in the immediately preceding frame RF, a motion vector is detected such that the motion difference value becomes minimum (P-VOP). Type encoding process).
[0012]
The DCT unit 103 performs a frequency transform called discrete cosine transform (discrete cosine transform) on the block pixel data of 8 × 8 pixels, and the result of the discrete cosine transform is 8 × 8, that is, 64. DCT coefficient data. Of these, the first coefficient is called a DC (direct current) component, and the other 63 coefficients are called AC (alternating current) components.
[0013]
The quantization unit 104 divides each input DCT coefficient data by a value called a quantization scale, thereby reducing the accuracy of the DCT coefficient and generating a quantized DCT coefficient. The generated quantized DCT coefficients are input to the DC / AC prediction unit 105 and the inverse quantization unit 109. In the processing in the quantization unit 104, if the quantization scale is arithmetically increased, the value of the quantized DCT coefficient decreases, and as a result, the generated code amount can be suppressed. Since the reproduction accuracy of the decoded image is reduced, the image quality of the decoded image is deteriorated.
[0014]
The DC / AC prediction unit 105 calculates and outputs a DC / AC coefficient prediction difference between each block in the intra macroblock and its neighboring blocks. As a block used as a prediction value, a block spatially left or above the current block is a candidate, and a block having a small difference between the current block and the DC coefficient is selected. The AC prediction is a process of calculating a difference between a selected prediction block and some AC coefficients of the current block. Here, DC prediction is an essential process in the MPEG-4 specification, whereas AC prediction is a selective process.
[0015]
Whether or not to perform the AC prediction processing is determined by performing the AC difference calculation on the six blocks in the macroblock and using the AC prediction result if the sum of the absolute values of the results is smaller than the case where the AC prediction is not performed. Is recommended. The arrangement of the coefficient data in the block output from the DC / AC prediction unit 105 is as follows.
(1) No AC prediction processing is performed.
[0016]
(2) Perform AC prediction processing and use the left block as prediction data.
[0017]
(3) Perform AC prediction processing and use the upper block as prediction data.
[0018]
It depends on the three cases.
[0019]
As described above, in the DC / AC prediction unit 105, the characteristics of the output data are determined depending on the statistics of the macroblock input data. Therefore, it is necessary to buffer the macroblock input data from the quantization unit 104 internally. There is.
[0020]
The variable-length coding unit 107 performs run-length coding on the quantized DCT coefficient input from the DC / AC prediction unit 105, assigns a variable-length code to each 0-run sequence (a sequence in which “0” is continuous), and performs coding. Output the stream. The generation code buffer 108 is a buffer for temporarily storing the coding result of the variable length coding unit 107, and absorbs a difference in performance between the generation performance of the coded stream and the output performance to the outside.
[0021]
If the generated code amount is larger than the target code amount based on the code amount generated by the variable length coding unit 107, the rate control unit 106 gives the quantization unit 104 a larger quantization scale value. In the case where the generated code amount is smaller than the target code amount, the quantization unit 104 is instructed to perform quantization with a smaller quantization scale value.
[0022]
The inverse quantization unit 109, the IDCT unit 110, and the motion difference compensating unit 111 perform a series of inverse transforms in order to generate a decoded image for the motion difference detection process of the next frame and store the decoded image in the decoded image memory 112. It is a processing system.
[0023]
The inverse quantization unit 109 multiplies the quantized DCT coefficient output from the quantization unit 104 by the quantization scale value to restore the DCT coefficient including the error. The IDCT unit 110 performs inverse discrete cosine transformation (DCT) on the DCT coefficients to restore motion difference pixel data of 8 × 8 pixels. The motion difference prediction unit 111 reads out the 8 × 8 pixel data of the previous frame from the decoded image memory 112 based on the motion vector, adds the input motion difference pixel data, reproduces the decoded image of the current frame, and , Stored in the decoded image memory 112. The decoded image memory 112 is a memory for storing the reproduced decoded image, and stores the decoded image data of two frames of the previous frame and the current frame while switching in units of frames.
[0024]
[Problems to be solved by the invention]
In the conventional moving image compression encoding apparatus having the configuration shown in FIG. 1, the rate control unit 106 performs feedback control based on the result of the generated code amount. There is a problem that the amount is generated and the generated code buffer 108 overflows.
[0025]
As a solution to such a problem, for example, Japanese Patent Application Laid-Open No. 9-261624 is known. Japanese Patent Application Laid-Open No. 9-261652 proposes an encoding device (moving image compression encoding device) having a function of stopping the output of the DCT section of a block when the generated code amount exceeds a predetermined value. .
[0026]
However, in the MPEG-4 specification, as described above, it is necessary to buffer the quantized DCT coefficients once in the DC / AC prediction unit 105 before performing variable-length coding. That is, in the variable length coding process for one macroblock, it is necessary to wait for the DCT process in the macroblock to be completely completed. It is impossible to abort. Further, even if the truncation process based on the generated code amount is performed on the output of the DC / AC prediction unit 105, in order to generate a decoded image inside the encoder device, the quantum that is subjected to the truncation process after the DC / AC prediction process is generated. The transformed DCT coefficient data must be returned to the original order and input to the inverse quantization unit 109. This not only significantly increases the time delay but also complicates the internal circuit as compared with the case where the quantized DCT coefficient data is directly output from the quantizer 104 to the inverse quantizer 109.
[0027]
SUMMARY OF THE INVENTION The present invention has been made in view of the above-described problems of the conventional moving image compression encoding apparatus, and has as its object to provide a moving image compression encoding apparatus capable of controlling an encoded stream to a constant transfer rate without complicating the configuration. I do.
[0028]
[Means for Solving the Problems]
According to the first aspect of the present invention, a motion difference detecting means for dividing each frame of video data into blocks and detecting a predicted motion difference between a current frame and a previous frame for each block; A variable-length encoding unit that performs a specified orthogonal transformation and quantization on the result of the motion difference or the original image data to perform variable-length encoding, and the orthogonal transformation and quantization are performed. Decoding image generating means for performing a specified inverse quantization, inverse orthogonal transform, and motion compensation processing on the data to generate a decoded image, and performing encoding of each block during the variable length encoding processing. Control to perform coding without using motion prediction in coding of the next frame image by being notified of coding cut-off means to be cut off by the coding stop means. Moving picture compression encoding apparatus is provided, characterized in that it comprises a frame management unit that, the.
[0029]
According to the moving image compression encoding apparatus according to the first embodiment of the present invention, the encoding discontinuation unit discontinues encoding of each block in the middle of the variable length encoding process, and the frame management unit In response to the notification of the coding cut-off from the coding cut-off means, control is performed so as to perform coding without using motion prediction in coding the next frame image.
[0030]
According to the second aspect of the present invention, a motion difference detecting means for dividing each frame of video data into blocks and detecting a predicted motion difference between a current frame and a previous frame for each block; A variable-length encoding unit that performs a specified orthogonal transformation and quantization on the result of the motion difference or the original image data to perform variable-length encoding, and the orthogonal transformation and quantization are performed. Decoding image generating means for performing a specified inverse quantization, inverse orthogonal transform, and motion compensation processing on the data to generate a decoded image, and performing encoding of each block during the variable length encoding processing. And a block position storing means for storing a position of a block in which encoding has been aborted by the encoding aborting means. In the motion prediction block in beam image, moving picture compression encoding apparatus, characterized in that no blocks the encoded truncation occurs as candidates for the motion prediction block is provided.
[0031]
According to the moving image compression encoding apparatus according to the second embodiment of the present invention, the encoding discontinuation unit discontinues the encoding of each block in the middle of the variable length encoding process, and the block position storage unit The position of the block in which the coding abortion has occurred is stored by the coding aborting means, and in the motion prediction of the block of the next frame image in which the coding abortion has occurred, the block in which the coding abortion has occurred is a candidate for a motion prediction block. And not to.
[0032]
Accordingly, it is possible to provide a moving image compression encoding device capable of controlling an encoded stream to a constant transfer rate without complicating the configuration. It should be noted that the moving picture compression / encoding device according to the present invention is not limited to the application of MPEG-4, but is also applicable to MPEG-1 and MPEG-2, and also to other types of moving picture compression / encoding device. be able to.
[0033]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the moving image compression encoding apparatus according to the present invention will be described in detail with reference to the accompanying drawings.
[0034]
FIG. 3 is a block diagram showing a first embodiment of the moving picture compression encoding apparatus according to the present invention, and shows the configuration of an encoding apparatus used for MPEG-4 compression technology. 3, reference numeral 201 denotes a current image memory, 202 denotes a motion difference detection unit, 203 denotes a DCT unit, 204 denotes a quantization unit, 205 denotes a DC / AC prediction unit, 206 denotes a rate control unit, and 207 denotes variable length coding. Reference numeral 2071 denotes a coding termination unit. Reference numeral 208 denotes a generated code buffer, 209 denotes an inverse quantization unit, 210 denotes an IDCT unit, 211 denotes a motion difference compensation unit, 212 denotes a decoded image memory, and 213 denotes a frame management unit.
[0035]
As is clear from the comparison between FIG. 3 and FIG. 1 described above, in the moving image compression encoding apparatus (encoding apparatus) of the first embodiment, the encoding cutoff means 2071 is provided in the variable length encoding unit 207. I have. When the quantized coefficient data in the block is subjected to variable-length encoding, if the amount of generated code exceeds a certain value, the coding terminating unit 2071 ignores the remaining quantized DCT coefficients in the block and changes the coding format. Then, the remaining block data is filled and output in a format that minimizes the amount of generated codes. Further, the variable-length coding unit 207 notifies the frame management unit 213 of an event that coding has been discontinued in the block. The frame management unit 213 performs an encoding process so as to perform compression encoding as an image type for which motion prediction is not performed, that is, an I-VOP in the case of MPEG-4, in the encoding of the next frame upon receiving the event of the encoding termination. Control.
[0036]
When the generated code amount of the block exceeds a certain value, the coding termination unit 2071 performs coding by regarding all the remaining quantized DCT coefficients in the block as zero. This means that every time a non-zero quantized DCT coefficient in a block is encoded, it is checked whether or not the amount of generated code exceeds a threshold. May be coded and output with a flag indicating that it is the last. Here, when the forced zero coefficient generation by the coding termination is performed, the coding termination unit 2071 notifies the frame management unit 213 and the rate control unit 206 of the event.
[0037]
The frame management unit 213 controls the encoding process so as to treat the next frame as an I-VOP when a forced zero coefficient occurs even in one block in the current (current) processing frame. In addition, since the rate control unit 206 needs to adjust the target value of the allocated code amount according to the VOP type, the rate control unit 206 receives the zero coefficient generation due to the coding termination, and receives the target code amount corresponding to the I-VOP for the next frame. Assign.
[0038]
FIG. 4 is a diagram for explaining the processing of the rate control unit in the moving picture compression encoding apparatus of FIG. 3, and FIG. 4 (a) shows a state in which normal processing (no encoding termination occurs) is performed. FIG. 4B shows a state in which the coding is discontinued and the processing is continued as it is, and FIG. 4C shows a state in which the coding is discontinued and the processing of the first embodiment is performed. Is shown.
[0039]
First, in a normal case in which coding termination does not occur, as shown in FIG. 4A, the rate control unit 206 transmits one I-VOP and a plurality of P-VOPs up to the next I-VOP. Are handled as one group unit (rate control group), and the generated code amount of each VOP is controlled so that each group has a substantially constant code amount. However, within one group, it is common to assign a larger amount of code to the I-VOP than to the P-VOP. Further, in order to simplify the calculation, the number of P-VOPs between two I-VOPs is determined to be a fixed number before encoding. In FIG. 4, one group includes one I-VOP and five P-VOPs.
[0040]
By the way, as shown in FIG. 4 (b), when the code generation of the block exceeds a certain value and the coding is discontinued and the processing is continued as it is, the I-VOP is forcibly stopped in the middle of the group. Thus, one or more I-VOPs having a relatively large code amount exist in one group. This is not a problem if the number of P-VOPs in the group is large, but if the number of P-VOPs in the group is small, the code amount to be allocated to the remaining P-VOPs due to the forcibly generated I-VOPs And the image quality may be degraded. (See IP in FIG. 4B)
Therefore, as shown in FIG. 4C, the rate control unit 206 according to the first embodiment sets the next frame after the frame including the coding cutoff block to be the first I-VOP of the rate control group. Is reassigned to the group bit amount. As a result, the increase in the bit amount caused by the forced insertion of the I-VOP can be gradually absorbed in the group unit, and the deterioration of the image quality can be prevented.
[0041]
A specific calculation method will be described with reference to FIG.
[0042]
First, assuming that the average target bit amount per frame given by the target transfer rate is Rpv, and the number of frames in a group defined in advance is Ngrp, the target code amount allocated to the group is Rpv × Ngrp It becomes.
[0043]
When the coding is discontinued in the group, if the generated code amount from the head of the group to the coding-discontinued frame is Ba, and the number of generated frames is Na (3 in FIG. 4C), Na sheets are generated. Since the code amount to be allocated on average is Rpv × Na, the difference Da from Ba (that is, Da = Ba−Rpv × Na) means the code amount generated extra for the transfer rate. Therefore, the extra code amount is subtracted from the target code amount assigned to the next rate control group. That is, Rpv × Ngrp-Da is assigned to the next rate group unit as a new target code amount. Accordingly, for example, it is possible to avoid the occurrence of a P-VOP in which the code amount to be allocated is small as shown by the reference code IP in FIG.
[0044]
FIG. 5 is a block diagram showing a second embodiment of the moving picture compression encoding apparatus according to the present invention. 5, reference numeral 301 denotes a current image memory, 302 denotes a motion difference detection unit, 303 denotes a DCT unit, 304 denotes a quantization unit, 305 denotes a DC / AC prediction unit, 306 denotes a rate control unit, and 307 denotes variable-length coding. Reference numeral 3071 denotes a coding terminating means. Reference numeral 308 denotes a generated code buffer, 309 denotes an inverse quantization unit, 310 denotes an IDCT unit, 311 denotes a motion difference compensation unit, 312 denotes a decoded image memory, and 314 denotes a block position storage unit.
[0045]
As is clear from the comparison between FIG. 5 and FIG. 3, the moving image compression encoding apparatus of the second embodiment includes a block position storage unit 314 instead of the frame management unit 213 in the first embodiment shown in FIG. It has become.
[0046]
The block position storage unit 314 stores the position of a block in which encoding has been discontinued, and the amount of information to be stored may be 1-bit information indicating whether or not truncation has occurred for each block position. It can be realized with a small-scale circuit. That is, when the frame next to the frame in which encoding has been discontinued is a P-VOP, the motion prediction detection unit 302 does not forcibly convert the frame into an I-VOP (see FIG. 4B). Referring to the block truncation information of the block position storage unit 314, when a motion prediction block candidate includes a coding truncation block as shown in FIG. 6 described later, the motion prediction block candidate is converted from the motion prediction candidate. Control to remove.
[0047]
FIG. 6 is a diagram for explaining the characteristic operation of the moving picture compression encoding apparatus shown in FIG. In FIG. 6, it means that 3 × 4 pixels (pixels indicated by black circles) for an arbitrary 8 × 8 pixel motion prediction block candidate have been coded and terminated in the previous frame.
[0048]
As shown in FIG. 6, the position of the block in which the coding is discontinued is stored in the block position storage unit 314, and the motion prediction detection unit 302 determines a part of the motion prediction block candidate in accordance with the stored position information. When the coding prediction block includes a coding cutoff block, control is performed such that the motion prediction block candidate is excluded from the motion prediction candidates.
[0049]
FIG. 7 is a diagram for explaining the characteristic operation of the moving picture compression encoding apparatus as the third embodiment according to the present invention. In FIG. 7, the original search area BL0 has four blocks in the horizontal direction and three blocks in the vertical direction. The rectangular area BL0 is defined as a first rectangular area BL1 having three blocks in the horizontal direction and one block in the vertical direction. It shows that the rectangular area is divided into three rectangular areas, that is, a second rectangular area BL2 having four blocks in the horizontal direction and one block in the vertical direction, and a third rectangular area BL3 having three blocks in the horizontal direction and one block in the vertical direction. .
[0050]
When detecting a motion prediction block, the range of the prediction block to be searched is limited in advance. For example, a search is performed in a rectangular area surrounding the current block position. Here, the motion prediction detection process is also performed on a region including two rectangular regions (for example, a region (R0) extending over the first rectangular region BL1 and the second rectangular region BL2).
[0051]
In the third embodiment, as shown in FIG. 7, in the motion prediction detection process, before the motion prediction detection unit 302 detects the motion prediction, the block position storage unit 304 is referred to and the rectangle to be searched originally is determined. The region is divided into a plurality of rectangular regions that do not include the coding cutoff block. As a result, in the motion detection process, it is necessary to perform a search process for each rectangular area, but there is no need to detect whether or not a coding discontinuation occurrence block is included in a motion prediction block candidate unit. Processing can be simplified.
[0052]
As described above, according to the moving image compression encoding apparatus according to each embodiment of the present invention, in a frame next to an image in which encoding has been discontinued, an image in which encoding has been discontinued is subjected to a prediction image for motion prediction difference detection. No need to use as Therefore, in the frame including the coding discontinuation occurrence block, even if the image decoded inside the encoder and the image decoded from the encoded stream in the receiving device are different, the same image is reproduced as the I-VOP in the next frame. Is done. As a result, the overflow of the generated code amount can be controlled without interrupting the flow of data from the quantization unit 205 (305) to the inverse quantization unit 209 (309).
[0053]
【The invention's effect】
As described above in detail, according to the present invention, it is possible to provide a moving image compression encoding apparatus capable of controlling an encoded stream at a constant transfer rate without complicating the configuration.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating an example of a conventional moving image compression encoding device.
FIG. 2 is a diagram for explaining processing in a motion difference detection unit.
FIG. 3 is a block diagram showing a first embodiment of the moving picture compression encoding apparatus according to the present invention.
FIG. 4 is a diagram for explaining processing of a rate control unit in the moving picture compression encoding apparatus of FIG. 3;
FIG. 5 is a block diagram showing a second embodiment of the moving picture compression encoding apparatus according to the present invention.
6 is a diagram for explaining a characteristic operation of the moving picture compression encoding device shown in FIG. 5;
FIG. 7 is a diagram for explaining a characteristic operation of a moving image compression encoding apparatus as a third embodiment according to the present invention.
[Explanation of symbols]
101, 201, 301 ... current image memory
102, 202, 302... Motion difference detection unit
103, 203, 303 ... DCT unit
104, 204, 304 ... quantization unit
105, 205, 305 DC / AC prediction unit
106, 206, 306 ... rate control unit
107, 207, 307 ... variable length coding unit
108, 208, 308... Generated code buffer
109, 209, 309: inverse quantization unit
110, 210, 310 ... IDCT unit
111, 211, 311... Motion difference compensator
112, 212, 312... Show decoded image memories.
213: Frame management unit
314: Block position storage unit
2071, 3071 ... coding termination means

Claims (5)

A motion difference detection unit that divides each frame of the video data into blocks and detects a predicted motion difference between the current frame and the previous frame for each block;
Variable-length encoding means for performing variable-length encoding by performing a prescribed orthogonal transformation and quantization on the result of the predicted motion difference or the original image data,
Decoded image generating means for generating a decoded image by performing a prescribed inverse quantization, inverse orthogonal transform, and motion compensation processing on the data on which the orthogonal transform and quantization have been performed;
At the time of the variable-length encoding process, encoding discontinuing means for discontinuing the encoding of each block in the middle,
A frame management unit configured to control to perform encoding without using motion prediction in encoding the next frame image by being notified of the encoding termination from the encoding termination unit. Moving image compression encoding apparatus. 2. The moving image compression encoding apparatus according to claim 1, further comprising a rate control unit configured to control a generated encoding amount of a group including a plurality of frames to be equal to a target encoded amount,
The moving picture compression method, wherein the rate control means re-assigns a target code amount by treating the next frame as the first frame of the rate control unit group by being notified of the coding cut-off from the coding cut-off means. Encoding device. A motion difference detection unit that divides each frame of the video data into blocks and detects a predicted motion difference between the current frame and the previous frame for each block;
Variable-length encoding means for performing variable-length encoding by performing a prescribed orthogonal transformation and quantization on the result of the predicted motion difference or the original image data,
Decoded image generating means for generating a decoded image by performing a prescribed inverse quantization, inverse orthogonal transform, and motion compensation processing on the data on which the orthogonal transform and quantization have been performed;
At the time of the variable-length encoding process, encoding discontinuing means for discontinuing the encoding of each block in the middle,
Block position storing means for storing a position of a block in which the coding abort has occurred by the coding abort means, wherein the coding abort occurs in a motion prediction of a block of a next frame image in which the coding abort occurs. A moving image compression encoding apparatus characterized in that the selected block is not used as a motion prediction block candidate. 4. The moving image compression encoding apparatus according to claim 3, wherein, based on the coding cutoff block information stored in the block position storage unit, the motion prediction of the block of the next frame image in which the coding cutoff occurs. Is divided into one or more areas that do not include any coding cutoff blocks. 5. The moving image compression encoding apparatus according to claim 1, wherein the encoding cutoff unit is configured such that data on which orthogonal transform and quantization have been performed by the variable length encoding unit are predetermined. 6. A moving image compression encoding apparatus characterized in that when a value is exceeded, encoding of the block is discontinued halfway.

Images