JPH0984011A - Moving image coding system converter - Google Patents

Abstract

PROBLEM TO BE SOLVED: To convert moving image data coded based on a system into moving image data coded based on another system representing a moving image of the same content at a high speed while suppressing deterioration in the image quality at a high speed. SOLUTION: At first, a format analysis section 12 extracts a coding block and a coding parameter from moving image data coded by the system A and a Huffman decoding section 13 and an inverse quantization section 14 of the system A decode sequentially the extracted result. Then the decoding result is coded sequentially by a quantization section 15 and a Huffman coding section 16 of the system B encode sequentially the decoding result, and a format configuration section 17 arranges the coding result in the same form as the moving image data coded by the system B. When the cross reference between the inverse quantization method of the system A and the quantization method of the system B is clear, a Huffman code replacement section 19 converts the Huffman code of the system A directly into the Huffman code of the system B based on the cross reference.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention encodes moving image data encoded according to a certain moving image encoding system according to another moving image encoding system which represents a moving image having the same contents as the moving image data. The present invention relates to a moving picture coding system conversion device for converting moving picture data.

[0002]

2. Description of the Related Art Generally, when storing (recording) moving image data having a huge amount of data in a storage medium such as a hard disk or transferring it through a communication line such as ISDN,
The moving image data is encoded in advance to reduce the data amount. This encoding method is internationally standardized,
Among them, used for personal computers,
MPEG1 (Moving Pict) for the purpose of storing moving image data
ure Experts Group phase 1) and recommendations H. 261 is widespread. Both of these coding methods are interframe predictive coding including motion compensation, DCT (Discrete Cosine Transform:
Discrete cosine transform), quantization, and Huffman coding are the data compression methods for moving image data.

By the way, in a system using a plurality of encoding systems, moving image data encoded according to one encoding system is encoded according to another encoding system that represents a moving image having the same content as that of the moving image data. It will be necessary to convert it to moving image data. For example, H.264 received via a communication path. 26
There is a case where moving picture data coded based on 1 is converted into moving picture data coded based on MPEG1 and stored in a storage medium, or vice versa. Conventionally, the H.264 standard is used to perform this conversion. Code based on 261
Using a decoding device and a coding / decoding device based on MPEG1, a moving picture signal before coding showing an original image is once reproduced, and another coding is performed on the reproduced moving picture data. Specifically, for example, moving image data encoded based on MPEG1 is converted into H.264. When converting into moving image data encoded based on H.261, the conversion is performed according to the following procedure.

First, a kind of inter-frame predictive coding including motion compensation, Huffman decoding, inverse quantization, and orthogonal transformation, which is defined by MPEG1, is applied to moving image data encoded based on MPEG1. IDCT (Inverse Discret
e Cosine Transform: Inverse Discrete Cosine Transform) to reproduce digital video data before data compression,
It is converted into an analog video signal by a D / A converter. Next, the converted video signal is converted into digital moving image data again by an A / D converter, and the conversion result is recommended by H.264. Encoding is performed again by sequentially performing inter-frame prediction including DCT, quantization, Huffman coding, and motion compensation defined by H.261. Thus, the above M
H.264, which represents a moving image having the same content as moving image data encoded based on PEG1. Video data encoded based on H.261 is obtained. In addition, moving picture data encoded based on MPEG1 is converted into H.264. Also when converting to data encoded based on H.261, the conversion is performed in the same procedure as above.

[0005]

However, in the conventional technique, the conversion processing includes processing that requires a long time, such as interframe predictive coding including DCT, IDCT, and motion compensation. There is a problem that the conversion process cannot be performed at high speed. In the conversion process, DCT, D / A conversion, A / D conversion, I
Since a lot of processing such as DCT is included, an error is likely to be accumulated in the converted moving image data, which causes a problem that the image quality of the moving image data is deteriorated.

Therefore, the present invention is an apparatus for converting moving image data encoded according to a certain encoding system into moving image data encoded according to another encoding system that represents a moving image having the same content as that of the moving image data. In order to speed up the conversion process, it is possible to suppress the deterioration of the image quality of the moving image data due to the conversion process.

[0007]

In order to achieve the above object, the present invention provides a first moving picture coding in which at least orthogonal transformation, quantization and Huffman coding are sequentially performed at the time of coding. Input means for inputting moving image data encoded based on the method, an encoded block encoded in pixel block units from the input encoded moving image data, and various types used for the encoding Format analysis means for extracting parameters, Huffman decoding means for performing Huffman decoding on the extracted coding block based on the first method, and Huffman decoding for the coding block Inverse quantization means for performing inverse quantization based on the first method, and at the time of encoding, at least orthogonal transformation, quantization, and Huffman encoding are sequentially performed, and the orthogonal transformation is performed by the first transformation. Together is identical to that performed in the formula, the quantization based on a different second video coding scheme and format the first type of video data after encoding,
Quantization means for performing the dequantized coding block, Huffman coding means for performing the Huffman coding on the quantized coding block based on the second method, and Huffman coding Format constructing means for constructing moving image data in the same format as the moving image data encoded based on the second method, which includes an encoding block and parameters equivalent to the extracted various parameters, and constructed moving image data And an output means for outputting.

[0008]

According to the above construction, the encoded moving image data can be converted without performing interframe prediction processing including DCT, IDCT, motion compensation, etc., so that the conversion processing can be speeded up. At the same time, reduction in deterioration of the image quality of moving image data due to the conversion process is achieved.

[0009]

Embodiments of the present invention will be described below with reference to the drawings.

<First Embodiment> FIG. 2 is a block diagram showing the arrangement of a moving picture coding system conversion apparatus according to the first embodiment of the present invention. In the figure, 1 is a CPU, 2 is a DSP (Digital S
A programmable data processing unit such as an ignal processor, 3 is a main storage device, 4 is a communication control unit, 5 is a data storage unit including a hard disk, 6 is a communication path such as ISDN or LAN, 7 is a bus, and 8 is a code. It is a conversion method conversion unit. C
The PU 1 controls the data processing unit 2, the communication control unit 4, the storage unit 5, and the encoding method conversion unit 8 via the bus 7 according to the control program stored in the main storage device 3.
The device can be connected to a device that reproduces or generates image data via an interface circuit (not shown).

FIG. 1 is a diagram showing the configuration of the above-mentioned encoding system conversion circuit 8.

In the figure, 11 is a data input section, 12 is a format analysis section, 13 is a Huffman decoding section, 14 is a dequantization section, 15 is a quantization section, 16 is a Huffman coding section, and 17 is a format construction section. , 18 is a data output section,
Reference numeral 19 is a Huffman code replacement unit. The conversion circuit 8 in the figure is
It is for converting moving image data encoded by a specific encoding method into moving image data encoded by another encoding method.

Next, the operation of the above-mentioned encoding system conversion section 8 will be described.

In the description here, the data input unit 11 is used.
The encoding method applied to the moving image data input to the above is referred to as the A method, and the encoding method applied to the moving image data output from the data output unit 18 is referred to as the B method.

In FIG. 1, first, the data input unit 11
Captures moving image data encoded by the A method from the storage unit 5 or from the communication path 6 via the communication control unit 4, and outputs the captured data to the format analysis unit 12. The format analysis unit 12 analyzes the data format of the input moving image data based on the A method, extracts the moving image data (encoded block) encoded in block units in block units according to the analysis result, The extracted data is output to the Huffman decoding unit 13. The Huffman decoding unit 13 performs Huffman decoding on the input data in block units according to the A method, and outputs the decoding result to the inverse quantization unit 14 for each block. The inverse quantizer 14
The Huffman-decoded block-unit data is dequantized according to the A scheme, and the result is output to the quantization unit 15. At this time, the parameters such as the quantization characteristic value necessary for performing the inverse quantization have been clarified by the format analysis performed by the format analysis unit 12, and therefore they are used.

Next, the quantizing unit 15 quantizes the inputted block-unit data according to the B method, and outputs the result to the Huffman coding unit 16. At this time,
The values of parameters such as quantization characteristic values used for quantization can be arbitrarily selected. The Huffman coding unit 16 performs Huffman coding on the input quantized block according to the B method. The format construction unit 17 selects a B block from an encoded block based on the B format.
The moving image data encoded based on the method is generated, and the generation result is output to the data output unit 18. Then, the data output unit 18 stores the moving image data encoded based on the B system in the storage unit 5, or the communication control unit 4
To the communication path 6 via.

Further, when the correspondence between the A-system inverse quantization method and the B-system quantization method is clear, or the A-system quantization method and the B-system quantization method are similar. In this case, the Huffman code replacing unit 19 directly replaces the Huffman code of the A system with the Huffman code of the B system without performing the processing of the processing units 13 to 16.

Next, the operation of the encoding system conversion section 8 will be described more specifically in correspondence with the actual encoding system.

FIG. 3 shows that the A method is an H.264 standard. 261 is a flowchart showing the operation of the encoding system conversion unit 8 when the B system is MPEG1. Each processing S10 to S1 in the figure
Reference numeral 8 denotes each processing unit 11 to 1 of the above-mentioned encoding system conversion unit 8.
Since the processings 9 and 9 are made to correspond to each other, the description will be given below without specifying the processing section.

The encoding method used for moving image data is H.264. In the case of converting from H.261 to MPEG1, the encoding method conversion unit 8 first receives the H.264 signal from the storage unit 5 or the communication path 6.
The moving image data encoded by the H.261 method is imported (S
10). And the H. The format of the data of 261 is analyzed, the encoded moving image data is extracted in macroblock units, and at the same time, the parameter information such as the quantization characteristic value is extracted for each frame of the encoded moving image data (S11). Here, the macro block is
The minimum unit of a pixel block for which Huffman coding is performed. In this embodiment, a macro block is 16 × 1.
It is composed of 4 luminance signal blocks obtained by dividing a 6 pixel luminance signal block in units of 8 × 8 pixels, and 2 sub-blocks of 2 color difference signals corresponding to the luminance signal blocks. . It should be noted that macroblock type information indicating the type of the macroblock is set at the beginning of the data extracted in units of macroblocks, and this macroblock is intra-coded or i-coded in this information.
If the data includes data indicating which of the inter coding is used, and the data indicates inter coding, the data further indicating whether to include a motion vector (MVD) and 6 sub-blocks. A coded block pattern (CBP) indicating which of these is subjected to Huffman coding, data indicating whether or not the quantization characteristic value (MQUANT) is changed, and the like are included. Then, based on this macroblock type information, Huffman decoding is performed for each block of 8 × 8 pixel units forming the macroblock (S12), and the decoding result is H. Inverse quantization is performed according to the H.261 method (S1
3).

Next, the dequantized block-unit data is quantized in accordance with MPEG1 (S14).
The value of the quantization parameter used in this quantization can be arbitrarily selected, and the parameter value used at this time is the macroblock type information at the time of Huffman encoding, which will be described later, or the format information when the format is constructed. Is set. Then, Huffman coding according to MPEG1 is performed on the quantized data in macroblock units (S15). At this time, H. The information such as the presence / absence of a motion vector, the coded block display, and the motion vector value included in the macroblock type information extracted by the format analysis (S11) of H.261 is MPEG1.
It is set to the macroblock type information of. Then, a format is constructed in frame units based on the Huffman-encoded data in macro block units (S
16). At this time, the macroblock included in the frame is H.264. In the case where all of the H.261 codings are intra-coded, the frame is converted to an MPEG1 I-picture, and if inter-coded is included, the frame is converted to an MPEG-1 P-picture. To do. Then, finally, the moving image data converted into the MPEG1 coding system by the processing of S10 to S16 is output to the storage unit 5 and the like (S17).

Further, in the process of constructing the above-mentioned format (S16), in order to make the frame rate at the time of reproduction of the converted moving image data constant,
Insert dummy B picture data. Since information indicating that prediction has not been performed from bidirectional frames to be predicted is set in the data of this dummy B picture, the insertion does not affect the image quality at the time of reproduction. .

FIG. 4 shows a specific example of inserting a dummy picture.

H. When the image data encoded in H.261 is decoded, for example, the frames 20, 21, and 2 shown in FIG.
As in 2, the frame intervals during playback are not constant. Here, the frame 20 is a frame composed of all intra-coded macroblocks, and the frame 2
Assuming that 1 and 22 are frames including inter-coded macroblocks, the above-described processing of S16 causes
The frame 20 is converted into a frame 23 of I picture, and the frames 21 and 22 are converted into frames 26 and 30 of P picture, respectively. However, in MPEG1, the H.264 standard is used. Since the frame rate is higher than 261 and the frame rate at the time of reproduction must be constant (for example, 30 frames per second), the time interval of the frames 23, 26, 30 is checked to make the frame interval constant. As described above, the dummy frames 24, 25, 27, 28, 29 of the B picture are inserted. As a result, moving image data having a constant frame rate based on MPEG1 can be obtained.

As is clear from the above, according to this embodiment, DCT, IDCT, interframe prediction, motion compensation, D
Since the conversion processing can be performed without performing the A / A conversion processing and the A / D conversion processing, high-speed conversion processing and reduction in deterioration of image quality are simultaneously achieved.

By the way, the quantization method of MPEG1 is H.264.
Since it is very similar to that of H.261, H.264. The Huffman code of H.261 is directly replaced with the Huffman code of MPEG1 (S
18) It can be done. When this replacement process is performed, the processes of S12 to S15 described above are not necessary, so the conversion process is performed faster than the above-described process.

The Huffman code replacement process will be described in detail below. However, in the following description, the value of the Huffman code itself is not used, but the value of the Huffman code is used.

First, H.264. A basic method of Huffman coding of H.261 and MPEG1 will be described.

In both encoding methods, as shown in FIG. 5, by scanning the quantized block 40 in 8 × 8 pixel units in a zigzag manner in the order shown by the arrows, each coefficient (0 ˜63) (hereinafter referred to as level). When a non-'0 'level coefficient is detected, the level of that coefficient and the number of' 0 'level coefficients detected immediately before the detection of that coefficient
Two of them (hereinafter referred to as "run") are encoded into Huffman code. Then, the preprocessing is repeated to add the coding result, and when all the coefficients from a certain coefficient to the end of the block are at the “0” level, the EOB (End Of Bloc)
The Huffman code indicating k) is added, and the encoding ends. FIG. 6 shows a specific example of this encoding. In the figure, the number in each lattice of the block 41 indicates the level of each coefficient included in the quantized block, and the blank indicates the “0” level. In this case, the scan of the block 41 produces the scan result 42 shown in the figure. Then, according to the scan result, the combination of the run and the level and the EOB
A Huffman code is selected to uniquely indicate and. On the contrary, in Huffman decoding, this run and level are decoded as a set.

Next, H. 261 dequantization and MPEG
The quantization of 1 will be described.

FIG. Intra in 261
It is a figure which shows the dequantized value of the level other than the DC value of the encoded macroblock. Dequantized (ie,
A macroblock (with DCT) is composed of one DC component and multiple AC components.

As can be seen from FIG. 7, the inverse quantized value is uniquely determined from the level and the quantized characteristic value. Also, in
The DC value of the tra-coded macroblock is obtained by multiplying the Huffman-decoded value by 8. On the other hand, MP
The quantization of the coefficients other than the DC value of the intra coding of EG1 is performed using an 8 × 8 quantization matrix, and
Set the level of each coefficient included in a block of 8 pixel units to 16
After multiplication, it is divided by the value of the corresponding position in the quantization matrix, and further divided by the value of twice the quantization characteristic value, and the resulting first decimal place is rounded to an integer.
When the absolute value of the integer exceeds 255, 25
Limit to 5. The DC value of the intra-coded macroblock is divided by 8 and then rounded to an integer.

FIG. It is a figure which showed the specific example of the process (S18) which replaces the Huffman code obtained by Huffman coding of H.261 with the Huffman code of MPEG1.

In the figure, the scan result 50 is H.264. Two
It is a Huffman code of a block including a macroblock which has been subjected to inter coding of 61, and the quantization characteristic value at this time is “2”. In this case, the MPE
Considering the quantization method of G1, if the quantization matrix of MPEG1 is set as the quantization matrix 51 in the figure, H.264. The Huffman code of H.261 is approximately MPEG without changing the quantization characteristic value, run, and level.
1 Huffman code (scan result 52). Here, the quantization matrix set in the above process is incorporated in the format when the format is constructed as described above, and is used when decoding the moving image data.

In addition to setting the quantization matrix of MPEG1 like the quantization matrix 51, the quantization characteristic value is set to 1/2 of the value in the scan result 50, and the scan level is set to the level in the scan result 50. 2
By multiplying and adding 1 (scan result 5
3), which is a permutation that allows a closer approximation.

The intra-coded macroblock may be replaced in the same manner as above.

<Second Embodiment> Next, a moving picture coding system conversion apparatus according to a second embodiment of the present invention will be described.

This embodiment has the same configuration as that of the first embodiment (FIGS. 1 and 2), and the reverse conversion (that is, the encoding system is converted from MPEG1 to H.261). )
Is to do.

In the following, description of the configuration will be omitted, and only the conversion process will be described.

9A and 9B, in the above-mentioned encoding method conversion unit 8 (FIG. 1), the A method is MPEG1 and the B method is the recommended H.264 standard. It is a flow chart which shows operation when it is set to 261.

In the figure, the encoding system conversion unit 8 first takes in moving image data encoded by the MPEG1 system from the storage unit 5 and the like (S20). Then, the format of the captured data is analyzed, and the encoded moving image data representing the moving image information is extracted in macro block units (S21). Further, in this processing, the amount of data taken out is checked at regular time intervals, and if the value is larger than a specific value, it is included in the MPEG1 P picture or B picture frame among the moving picture data taken in. To delete. This is an H.264 standard used for communication purposes. 261
This is because the bit rate needs to be lower than in the case of MPEG1, and the amount of data to be converted is reduced by deleting the frame. Then, the information of the quantization characteristic value is obtained for each frame of the moving image data and used in the subsequent processing. In addition, there is information indicating the type of the macroblock at the head of the data in units of macroblocks, and Huffman decoding of MPEG1 is performed on the data in units of macroblocks based on the information on the macroblock type, Decoding is performed for each block of 8 × 8 pixel units forming the block (S22). Then, the inverse-quantization of MPEG1 is performed on the decoded block-unit data (S23).

Next, for the dequantized block,
H. The quantization of 261 is performed (S24). The value of the quantization parameter used in this quantization can be arbitrarily selected, and the parameter value used at this time is the macroblock type information at the time of Huffman encoding, which will be described later, or the format information when the format is constructed. Set. Then, with respect to the quantized block, H.264 is performed in macro block units. Huffman coding of H.261 is performed (S2
5). At this time, format analysis of MPEG1 (S2
The information such as the presence / absence of a motion vector, the coded block display, the motion vector value, and the like included in the macro block type information extracted in 1) is described in H. 261 is set in the macroblock type information. Also, a quantization parameter arbitrarily selected by inverse quantization can be included in this macroblock. Then, a format is constructed in frame units based on the Huffman-encoded data in macro block units (S26). Further, in this processing, the amount of moving image data after the format construction is checked, and if the value is larger than a specific value (that is, the bit rate becomes high), the above quantization processing (S24)
Change the quantization characteristic value used in step 1 to a larger value. As a result, the compression rate of moving image data to be encoded thereafter becomes high, and as a result, the bit rate is reduced. Further, the bit rate can be suppressed by deleting the data in frame units as in the format analysis process (S21) described above. Finally, the H.264 format was constructed. The moving image compressed data 261 is output to the communication control unit 4 or the like (S27).

As is clear from the above description, according to this embodiment, similarly to the first embodiment, the speeding up of the conversion process and the reduction of the deterioration of the image quality are simultaneously achieved.

By the way, the size of the frame at the time of displaying the moving image data is freely set in MPEG1 and is set to H.264. 26
In 1, it is fixedly set. For example, the pixel size of a standard frame of MPEG1 is 352 [pixel / line] ×
240 [line], H.264. 261 CIF (Common Interm
352 [pixels / line] x 288 in editable Format)
[Line] is set, and the numbers of lines of both do not necessarily match. In order to prevent inconvenience from occurring during reproduction of a moving image due to this inconsistency, in the process of constructing the above-described format (S16, S26), a process for matching the frame sizes may be performed. The H.264 coding system conforms to MPEG1 to CIF. A specific example of this processing will be described below by taking the case of conversion into H.261 as an example.

FIG. 10 shows a frame 60 of 352 [pixels / line] × 240 [line] of MPEG1 and H.264 based on CIF. It is the figure which showed the frame 61 of 261.

In the figure, a frame 60 of MPEG1
Is composed of 22 macroblocks in the horizontal direction and 15 macroblocks in the vertical direction. H. The frame 61 of 261 is composed of 22 macroblocks in the horizontal direction and 18 macroblocks in the vertical direction. In this case, for example, the first and the first of the frame 61
Dummy macroblocks are set in the macroblock columns of the 7th and 18th rows, and the macroblocks of the frame 60 of MPEG1 are made to correspond to the macroblock columns of the 2nd to 16th rows. As a result, the frame 60 of MPEG1 is
H. It is converted into a frame 61 of 261.

Not limited to the above example, a dummy macro block can be added at an arbitrary position (including a column) of the frame 61. In contrast to the above, MPEG
The size of the frame of H.1 is H.264. If it is larger than the frame size of H.261, the macroblock at an arbitrary position of the MPEG1 frame may be deleted according to the size.

The conversion process of the encoding system conversion unit 8 described in the above two embodiments is performed by the DSP of the data processing unit 2.
May be performed by the CPU 1, or by the CPU 1. In addition, the encoding method is H.264. 261 to MPE
You may make it possible to perform both the process of converting to G1 (first embodiment) and the reverse process (second embodiment), and selectively execute one of them.

Further, in the above embodiment, MPEG1 and H.264 are used as the moving picture coding systems. Although H.261 has been taken as an example, it can be easily applied to other encoding methods such as Motion-JPEG.

[0050]

As described above, according to the present invention, moving image data encoded according to one encoding system is encoded according to another encoding system that represents moving images of the same content. In the device for converting the converted moving image data, it is possible to speed up the conversion process and suppress deterioration of the image quality of the moving image data due to the conversion process.

[Brief description of drawings]

FIG. 1 is a configuration diagram of an encoding method conversion unit in FIG.

FIG. 2 is a configuration diagram of a moving picture coding and converting apparatus according to an embodiment of the present invention.

[Fig. 3] Fig. 3 is a diagram illustrating an H.264 coding method performed by a coding method conversion unit. 261 to M
It is a flowchart of the conversion process to PEG1.

FIG. 4 is a diagram illustrating a method of adjusting a frame rate.

FIG. 5 shows MPEG1 and H.264. 26 is a diagram for explaining zigzag scanning performed in Huffman encoding of H.261.

FIG. 6 shows MPEG1 and H.264. [Fig. 26] Fig. 26 is a diagram for describing a basic method of Huffman encoding of H.261.

7: H. It is a figure for demonstrating the dequantization of 261.

FIG. 8 is a diagram for explaining a specific example of Huffman code replacement processing.

FIG. 9 is a block diagram showing an example of MPEG1 to H.264 in the encoding conversion unit. 261
It is a flowchart of the conversion process to.

[Fig. 10] Fig. 10 is a diagram for describing a method of adjusting a difference in frame size depending on an encoding method.

[Explanation of symbols]

 1 ... CPU 2 ... Data processing unit 3 ... Main memory 4 ... Communication control unit 5 ... Storage unit 6 ... Communication path 7 ... Bus 8 ... Encoding method conversion Part 11 ... Data input part 12 ... Format analysis part 13 ... Huffman decoding part 14 ... Dequantization part 15 ... Quantization part 16 ... Huffman coding part 17 ... Format constructing unit 18 ... Data output unit 19 ... Huffman code replacing unit 20 ... Intra-encoding only frame 21, 22 ... Frame including inter encoding 23 ... I-picture 26, 30 ... P picture 24, 25, 27, 28, 29 ... Dummy B picture 40 ... Zigzag scan 41 ... Block example 42 ... Scan for Huffman code 50 ... Huffman decoding Value skiing 51: MPEG1 quantization matrix example 52, 53: MPEG1 Huffman code replacement example 60: MPEG1 frame 60: Recommendation H.264 261 frames

 ─────────────────────────────────────────────────── ─── Continuation of front page (72) Inventor Toru Nonomura 1099, Ozenji, Aso-ku, Kawasaki-shi, Kanagawa Inside the Hitachi, Ltd. System Development Laboratory (72) Inventor Takao Shimada 1099, Ozen-ji, Aso-ku, Kawasaki, Kanagawa (72) Inventor, Kazuaki Tanaka, 5030 Totsuka-cho, Totsuka-ku, Yokohama-shi, Kanagawa, Ltd. (72) In-house, Hitachi, Ltd. Software Development Division (72) Kuji Matsumura, 810, Shimoimaizumi, Ebina, Ebina, Kanagawa Hitachi Systems Office Systems Division

Claims (8)

[Claims] 1. Input means for inputting moving image data encoded on the basis of a first moving image encoding method in which at least orthogonal transformation, quantization and Huffman encoding are performed at the time of encoding. From the input encoded moving image data, an encoding block encoded in pixel block units, a format analyzing unit that extracts various parameters used in the encoding, and an extracted encoding block ,
Huffman decoding means for performing Huffman decoding based on the first method, dequantization means for performing inverse quantization on the Huffman-decoded coding block based on the first method, and At the time of encoding, at least orthogonal transformation, quantization, and Huffman encoding are sequentially performed, the orthogonal transformation is the same as that performed in the first method, and the format of the encoded moving image data is Quantizing means for performing quantization based on a second moving image coding scheme different from the first scheme on the dequantized coding block; and for the quantized coding block, Huffman coding means for performing Huffman coding based on the second method, Huffman-coded coding blocks, and codes based on the second method including parameters equivalent to the extracted various parameters A moving picture coding method conversion device, comprising: a format constructing means for constructing moving picture data in the same format as the transformed moving picture data; and an output means for outputting the constructed moving picture data. 2. An input means for inputting moving picture data coded based on a first moving picture coding method for carrying out at least orthogonal transformation, quantization and Huffman coding at the time of coding. From the input encoded moving image data, an encoding block encoded in pixel block units, a format analysis unit that extracts various parameters used in the encoding, and at least orthogonal transformation at the time of encoding , The quantization and Huffman coding are sequentially performed, the orthogonal transformation is the same as that performed in the first method, and the format of the moving image data after encoding is the first method. At least one of the extracted coding block and various parameters is converted according to a predetermined rule based on a correspondence relationship between a different second moving image coding method and the first method. Huffman code replacing means for converting, and a format constructing means for constructing moving image data in the same format as the moving image data encoded based on the second method, including the converted encoded block and various parameters,
An apparatus for converting a moving picture coding system, comprising: an output means for outputting the constructed moving picture data. 3. The moving picture coding system conversion device according to claim 1, wherein the format analysis unit or the format construction unit performs conversion or conversion according to the second system. In order that each of the frames composed of a plurality of Huffman-encoded blocks is displayed at a desired timing during reproduction, a frame that does not disturb the display image during reproduction is added to the encoded moving image data. Configure,
Insert multiple dummy coded blocks, or
A moving picture coding system conversion apparatus, characterized in that, from the coded moving picture data, a coding block included in a part of frames constituted by the coding blocks is deleted. 4. The moving picture coding system conversion device according to claim 1, wherein the format analyzing unit or the format constructing unit is moving picture data coded based on the second system. So as to be transferred at a desired data rate at the time of transfer, the coded blocks included in some of the frames composed of the coded blocks are deleted from the coded moving image data. A moving picture coding system conversion device characterized by the above. 5. The moving picture coding system conversion device according to claim 1 or 2, wherein the format constructing unit performs a plurality of conversions or Huffman codings corresponding to the second system. A dummy coded block that does not disturb the display image at the time of reproduction so that the pixel size of the frame composed of the coded blocks at the time of reproduction becomes closer to the pixel size of the frame defined by the second method. Is added to a predetermined position of the frame, or a coding block at a predetermined position is deleted from a plurality of coding blocks forming the frame. 6. The moving picture coding system conversion device according to claim 1, wherein the format constructing unit sets a data amount of a coding block Huffman-coded in a unit time based on the second system. A moving picture coding system conversion apparatus, characterized in that the value of a parameter used for processing by the quantizing means is changed so that the amount of data is closer to a desired value. 7. The moving picture coding method conversion device according to claim 1, wherein the first method is a moving picture coding method defined by MPEG1, and the second method is Recommendation H. ． A moving picture coding method conversion device characterized by being a moving picture coding method defined by H.261. 8. The moving picture coding system converter according to claim 2, wherein the first system is Recommendation H.264. 261 is a moving picture coding system defined by H.261, and the second system is a moving picture coding system defined by MPEG1.