JP2000032446A - Dynamic image data compressor and storage medium thereof - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a dynamic image data compressor that satisfactorily conducts compression processing at a high speed, without deteriorating the image quality of a reproduced image. SOLUTION: When a series of input frames of video data consist sequentially of a picture I1 which is an I picture, pictures B1, B2, B3 being B pictures, a picture P1 which is a P picture, pictures B4, B5, B6 being B pictures, and a picture P2 which is a P picture,... being P picture, compressors 1, 2, 3 compress the pictures I1, P1, P2 in common, and any of the compressors 1-3 compresses the pictures B1-B6. Then the data after the compression are joined to form one stream to obtain data similarly, to the case with compression by one compressor and number of frames compressed by each of the compressors 1, 2, 3, in comparison with compression by one compressor.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a moving picture data compression apparatus for compressing moving picture data, and more particularly, to a moving picture data compression apparatus in consideration of high-speed compression.

[0002]

2. Description of the Related Art In recent years, with the diversification of communication devices, there has been proposed a technique of compressing moving image data, reducing the amount of data, and using the data for communication or the like. As a method of compressing the moving image data, for example, MPEG (Moving Picture Im)
Age Coding Experts Group) is known. In addition, various proposals have been made for a moving image data compression apparatus employing this type of system in order to speed up the compression processing. For example, Japanese Patent Application Laid-Open No. 8-205142 proposes dividing one screen into a plurality of screens and providing a compressor for each of the divided sections to compress moving image data.

[0003]

However, in the case of a compression method using an inter-frame difference such as MPEG, if one screen is divided into a plurality of screens, a motion vector extending over the divided portion cannot be used. For this reason, the compression ratio of the entire apparatus is reduced due to the division, and the image quality of the reproduced image is reduced, and the compression processing cannot be performed at a sufficiently high speed.

Therefore, an object of the present invention is to provide a moving picture data compression apparatus capable of satisfactorily speeding up the compression processing without deteriorating the quality of a reproduced image.

[0005]

Means for Solving the Problems and Advantages of the Invention The invention according to the first aspect of the present invention has been made to achieve the above object, and a moving image data compression method for compressing moving image data representing a moving image by a plurality of frames corresponding to the passage of time. The apparatus, comprising a plurality of compression means for compressing only the moving image data relating to a part of frames of the moving image data, wherein each of the compression means compresses moving image data, It is characterized by matching the entire data.

The present invention includes a plurality of compression means for compressing only moving image data relating to some frames. Since each of these compression means compresses only moving image data relating to some frames, the compression processing can be performed at a higher speed than in the case where all frames are compressed. In addition, since this compression means speeds up the compression processing by thinning out frames, so to speak, without depending on the screen division, motion vectors and the like can be sufficiently used, and the compression ratio does not decrease. For this reason, it is possible to prevent the image quality from lowering.

[0007] The union of frames for which each compression means compresses the moving image data coincides with the entire moving image data. For this reason, if the moving image data compressed by the respective compression means is appropriately connected, the entire original moving image data is compressed. Therefore, according to the present invention, it is possible to satisfactorily speed up the compression processing of the moving image data without deteriorating the image quality of the reproduced image.

According to a second aspect of the present invention, in addition to the configuration of the first aspect, each of the compression means compresses the moving image data by using an inter-frame difference of the moving image data, and compresses the moving image data with respect to another frame. The compression means commonly compresses the moving image data of a specific frame referred to when calculating the inter-frame difference.

In the case where moving picture data is compressed using the inter-frame difference, moving picture data of a specific frame (for example, I picture and P picture data in MPEG) to be referred to when calculating the inter frame difference relating to another frame is used. Although necessary, in the present invention, the compression means commonly compresses the moving image data of the specific frame. For this reason, each compression unit can independently compress the moving image data without exchanging data with other compression units.

Therefore, according to the present invention, in addition to the effect of the first aspect of the present invention, there is an effect that the speed of the compression processing of the moving image data can be further improved. Claim 3
According to the invention described in claim 1, in addition to the configuration according to claim 1 or 2, each frame constituting the moving image data is distributed to each of the compression units that compress the moving image data related to the frame, and is input to the corresponding compression unit. The distribution means
It is further characterized by being provided.

In the present invention, the distributing means distributes each frame constituting the moving picture data to each compressing means for compressing the moving picture data relating to the frame, and inputs it to the corresponding compressing means. For this reason, each compression unit does not need to determine whether or not to compress each frame constituting the moving image data by itself, and simply compresses the moving image data input from the distribution unit as usual. Good. For this reason,
The processing of the compression means is simplified, and an existing compressor can be used as the compression means.

Therefore, according to the present invention, in addition to the effects of the first and second aspects of the present invention, the processing of the compression means is further simplified to further speed up the compression processing of moving image data, and an existing compressor is used. As a result, there is an effect that the manufacturing cost can be reduced. According to a fourth aspect of the present invention, there is provided a storage medium storing a software program for controlling a plurality of compression means for compressing moving image data representing a moving image by a plurality of frames corresponding to a lapse of time. A software program for executing a selection process for compressing only the moving image data relating to a part of the frames of the moving image data, and a union set of the frames compressed by the respective compression means in accordance with the selection process Correspond to the whole of the moving image data.

If the plurality of compression means are controlled by the software program stored in the present invention, only the moving picture data relating to a part of the frames is transmitted to each compression means by the selection processing executed based on the software program. Can be compressed. Moreover, in the case of such control, the union of the frames compressed by the respective compression means coincides with the entire moving image data. In other words, if a plurality of compression means are controlled by a software program stored in the present invention, the plurality of compression means are stored in the storage medium.
It can be operated in the same way as the described compression means.

Therefore, according to the present invention, the same operation and effect as those of the first aspect can be produced by a plurality of general compression means.

[0015]

Next, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram schematically illustrating a configuration of a moving image data compression device to which the present invention is applied. This moving image data compression apparatus converts video data as moving image data read from a video tape or the like into an MPEG format.
This is a device that compresses data based on the standard No. 2 and outputs it as a stream to a communication line or the like.

As shown in FIG. 1, the present moving picture data compression apparatus includes three compressors 1, 2, and 3 as compression means, and inputs video data input from the outside to an input switch as distribution means. 5 to each of the compressors 1 to 3. The input switch 5 is a switch S1 for switching whether to input video data to the compressor 1, a switch S2 for switching whether to input video data to the compressor 2, and a video data to the compressor 3. Switch S3 for switching whether or not each switch S1 to S3
The state of S3 is controlled by an input controller 7 described later.

Here, the configuration of the compressor 1 and details of the compression operation thereof will be described with reference to FIG. The other compressors 2 and 3 have the same configuration as the compressor 1. As shown in FIG. 2, a compressor 1 includes a subtractor 11 for performing a subtraction process between input video data and reference data described later, and an orthogonal transform for performing a DCT (discrete cosine transform) process, which is a type of orthogonal transform. DCT converter 12 as a means and its DC
A quantizer 13 as quantization means for quantizing the DCT coefficient data output from the T transformer 12, and a variable length for performing variable length coding on the DCT coefficient data quantized by the quantizer 13; A variable length encoder 14 as an encoding means, and a DC quantized by the above-described quantizer 13
An inverse quantizer 15 for inversely quantizing the T coefficient data, and an inverse DCT for performing an inverse DCT (inverse discrete cosine transform) process on the DCT coefficient data inversely quantized by the inverse quantizer 15
DCT by the T converter 16 and its inverse DCT converter 16
An adder 17 for adding motion compensation to the data returned to the video data before being processed, a memory 18 for storing an output from the adder 17 for motion compensation prediction,
A motion compensator 19 that performs motion compensation prediction based on the data stored in the memory 18 and outputs it as reference data.
And

This compressor 1 can compress the input video data as follows and output it as a stream. First, the video data is read out to the subtractor 11 in synchronization with a synchronization signal accompanying the video data. Subtractor 1
1 also receives reference data (prediction value data) from the motion compensator 19, and outputs difference data (prediction error) between the video data and the reference data to the DCT converter 12.

The DCT converter 12 performs a DCT (Discrete Cosine Transform) process on the difference data input from the subtractor 11. The calculation unit of this processing is performed in a predetermined block unit, and a number of coefficients (hereinafter, also referred to as DCT coefficients) corresponding to the block unit is calculated. The DCT transformer 12 arranges the coefficients in the order of frequency components and outputs the coefficients to the quantizer 13. In the quantizer 13, the DC input from the DCT converter 12 at a preset quantization scale is input.
The T coefficient is quantized, and the quantized value is output to the variable length encoder 14. The variable-length encoder 14 determines a significant block, performs variable-length coding on the quantized DCT coefficient, and outputs it as a stream.

In this variable length coding, for example, after the output of the quantizer 13 is run-length transformed,
It is conceivable that the coefficient is converted into a variable length code and output. The run-length transform results in two coefficients, which are a set of the number of zero coefficients and the value of the next non-zero coefficient. However, since the output of the quantizer 13 tends to continue with many zero coefficients, the run-length The data amount is reduced by the conversion. Next, 2 obtained as a result of the run-length transformation
Compressed data can be obtained by collecting the coefficients and allocating one variable-length code to output.

On the other hand, the inverse quantizer 15 includes a quantizer 13
Since the quantized DCT coefficients output from are input, inverse quantization is performed using the quantization scale used in the quantizer 13, DCT coefficients are calculated, and output to the inverse DCT transformer 16. . The DCT coefficient has a value including an error corresponding to the quantization scale, and the compression rate of the compressor 1 can be adjusted by the quantization scale.

The inverse DCT transformer 16 performs an inverse DCT (inverse discrete cosine transform) process on the DCT coefficients,
Output to the adder 17. The adder 17 adds the output from the inverse DCT converter 16 and the output from the motion compensator 19 and inputs the result to the memory 18. The memory 18 is a frame memory for motion compensation prediction, and is configured by a frame memory for at least two frames. Then, at the same time as accumulating the data from the adder 17, the data of the previous frame for motion compensation prediction is output to the motion compensator 19. The motion compensator 19 performs a well-known motion compensation prediction using a motion vector or the like on the data of the previous frame input from the memory 18 and outputs the data to the subtractor 11 as prediction value data.

This is a schematic operation relating to data compression. As described above, by utilizing a difference between frames of video data (moving image data) and employing run-length transform and variable length code, efficient data compression is achieved. Can be realized. Therefore, the amount of video data can be reduced, and the transmission of the data can be facilitated.

Returning to FIG. 1, the compressed video data (hereinafter, also referred to as compressed data) output from each of the compressors 1, 2, and 3 is respectively divided into picture cut-out units 21, 22, 23.
After the data is cut out and the data is cut out as described later, the data is individually stored in the stream buffers 41, 42, and 43. The compressed data stored in the stream buffers 41, 42, and 43 is output as a stream to an external communication line or the like via the output switch 45. The output switch 45 includes a terminal T1 connected to the output side of the stream buffer 41, a terminal T2 connected to the output side of the stream buffer 42, and a terminal T3 connected to the output side of the stream buffer 43. Output controller 4
In response to a command from the terminal 7, compressed data output from one of the terminals T1 to T3 is output to the outside.

Next, the input controller 7, the picture cutout units 21, 22, 23, and the output controller 47
The microcomputer is mainly composed of a microcomputer having an OM and a RAM, and executes the following control based on a software program stored in the ROM. First, an outline of a process executed by control of each unit will be described.

In MPEG-1 and MPEG-2, a large number of frames constituting video data are sorted into three types of I picture, P picture, and B picture, so that a random access function and a high coding rate can be achieved. Has been realized. In the example of FIG. 3, a series of input frames constituting video data are represented by pictures I1, B
Pictures B1, B2, B3 which are pictures, picture P1 which is a P picture, picture B4 which is a B picture
B5, B6, and a picture P2, which is a P picture, are set. In this case, the frame that becomes the picture I1 is coded independently of the other frames, and the picture P1 is shifted from the picture I1 located in the past in time to the picture P1.
2 is predictively coded from a picture P1 located in the past in time. Also, pictures B1, B2, B3
Are predictively coded from both directions using pictures I1 and P1 located before and after in time, and pictures B4, B5, B6
Are predictively coded from both directions using pictures P1 and P2.

As described above, the data of the I picture and the P picture is referred to when encoding the B picture (that is, it corresponds to a specific frame), but the data of the B picture is not referred to at all when encoding other pictures. . Therefore, in the present moving image data compression apparatus, a compression process is performed on the input frames illustrated in FIG.
3 is assigned as follows.

As illustrated in FIG. 4, pictures I1, P
1 and P2 are commonly compressed by the compressors 1, 2 and 3, and the pictures B1 to B6 are compressed by any one of the compressors 1 to 3. For this reason, the number of frames to be compressed by each of the compressors 1 to 3 is reduced as compared with the case where the compression processing is performed by one compressor, and the compression processing can be sped up.

Next, the processing of each unit will be described in detail with reference to the flowcharts of FIGS. FIG. 5 is a flowchart illustrating a process of the input controller 7 that controls the input switch 5. The input controller 7 starts this processing when video data is input. When the processing is started, the input controller 7 turns on all the switches S1 to S3 in S1 (S represents a step: the same applies hereinafter), and waits until one frame of video data is transferred in S2. I do. Then, the frame that becomes the picture I1 is input to the compressors 1, 2, and 3, respectively. In the subsequent S3, the switch S1 is turned off.
N, turn off the others, and wait until one frame is transferred in S4. Then, a frame to be a picture B1 is input to the compressor 1. Similarly, in S5 to S8, the switch S2 is turned on and the others are turned off (S5), and the process waits until one frame is transferred (S6), and the switch S3 is turned on and the others are turned off (S7). It waits until it is transferred (S8). Then, the frame to be the picture B2 is in the compressor 2, and the frame to be the picture B3 is in the compressor 3.
Are sequentially input.

In S9 following S8, it is determined whether or not the input of the video data has been completed.
9: NO), and proceeds to S1 to repeat the above processing. By the above processing, data as illustrated in FIG. 4 is input to each of the compressors 1, 2, and 3. That is, S1 in FIG.
The processing of S8 corresponds to the selection processing. Then, when the input of the video data ends (S9: YES), the input controller 7 ends the processing. In S9, an affirmative determination is also made when the end of the compression process is instructed by an operation panel or the like (not shown).

Subsequent to this processing, the compressed data of each picture illustrated in FIG. 4 is output from the corresponding compressors 1 to 3, but the compressed data corresponding to the pictures I1, P1 and P2 are all compressed data. Are commonly output from the devices 1 to 3. Therefore, in the present moving image data compression apparatus, the compressed data output in common as described above uses the data output from the compressor 1 as a representative. For this reason, in the present moving image data compression apparatus, the processing of the picture clipper 21 and the processing of the picture clippers 22 and 23 are different as follows. When compressing the video data, the compressors 1 to 3 compress the I picture and the P picture prior to the B picture sandwiched between them, and return the order during reproduction. Therefore, the description is omitted.

FIG. 6 is a flowchart showing the processing of the picture clipper 21, and FIG. 7 is a flowchart showing the processing of the picture clipper 22. Note that the processing of the picture clipper 23 is the same as the processing of the picture clipper 22. As shown in FIG. 6, the picture cutout unit 21 reads the picture header output from the compressor 1 in S21, and outputs the picture header to the stream buffer 41 in subsequent S23. Furthermore, the following S
At 25, the data of one picture output from the compressor 1 is output to the stream buffer 41 as it is. That is, the compressed data output from each of the compressors 1 to 3 is composed of a picture header indicating which type of the picture is I, P, or B, and one picture of compressed data following the picture header. S21-
In S25, the compressed data output from the compressor 1 is output to the stream buffer 41 as it is.

At S27, picture information is output to the output controller 47 for the processing described later. The picture information includes information indicating the type of the picture, I, P, or B, and information indicating the size of data for one picture. In S29 following S27, it is determined whether the input has been completed. If the input has not been completed (S29: N
O), the process proceeds to S21 to repeat the above-described processing. If the processing has been completed (S29: YES), the processing is terminated as it is. The determination of the end of the input in S29 is made in the same manner as in S9, but is made later than the determination in S9 by a time corresponding to the compression processing of the compressor 1. By the above processing, all the compressed data output by the compressor 1 is output to the stream buffer 41 as it is, and is accumulated there.

On the other hand, the picture cutout unit 22
As shown in FIG. 7, after reading the picture header output from the compressor 2 in S31, it determines in S32 whether or not it represents a B picture. In the case of a B picture (S32: YES), the above-described S23 to S2
Steps S33 to S39 similar to step 9 are executed. That is, the picture header and the data of one picture outputted from the compressor 2 are outputted to the stream buffer 42 and the picture information is outputted to the output controller 47. on the other hand,
If it is determined in S32 that the picture is not a B picture (S3
2: NO), proceed to S41, discard the data of one picture, and proceed to S39.

With the above processing, the picture cutout unit 2
2 (and 23), of the compressed data output from the compressor 2 (or 3), discards I-picture and P-picture data and outputs only B-picture data to the stream buffer 42 (or 43) for storage. can do.
Therefore, in the case of the examples of FIGS.
1 stores compressed data of pictures I1, B1, P1, B4, and P2, compressed data of pictures B2 and B5 in the stream buffer 42, and compressed data of pictures B3 and B6 in the stream buffer 43, respectively. Will be accumulated. Therefore, the output controller 47
The output switch 45 is controlled as follows based on the picture information output from each of the picture clippers 21 to 23.

FIG. 8 is a flowchart showing the processing executed by the output controller 47. As shown in FIG. 8, the output controller 47 first sets a variable n to 1 in S51. At S53, the output switch 45 is switched to the terminal T1. As described above, the head of the input frame is compressed as an I picture. This compressed data is stored only in the stream buffer 41. Therefore, first, the output switch 45 is switched to the terminal T1 on the stream buffer 41 side, and preparation is made so that the data in the stream buffer 41 can be output to the outside. In S55, the picture information output from the picture clippers 21 to 23 is read.
The header and data of one picture are externally output from any of the stream buffers 41 to 43 according to the setting of 5.

At S61, based on the picture information read at S55, the compressed data output at S57 is
It is determined whether the data is picture data. As described above, since an I picture is output at first, a negative determination is made here and the process proceeds to S63. In S63, S29, S
It is determined that the input has been completed in the same manner as in step S39. If the input has not been completed (S63: NO), the process proceeds to S55.

In S55, the next picture information is read, and the data is output in S57. In S65, it is determined whether or not the compressed data output in S57 is B picture data. As described above, the compressors 1 to
3 is a B picture sandwiching an I picture and a P picture
Compress before pictures. For this reason, in the examples of FIGS. 3 and 4, the output controller 47 supplies the pictures I1, P1, B
Picture information is input in the order of 1, B2,. That is, secondly, P picture data is output from the output switch 45, so that a negative determination is made in S61 and S63 as in the previous time, and the process returns to S55.

The flow shifts to S55, where the next picture information is read, and data is output in S57.
Since it is a picture (S61: YES), the flow shifts to S65. In S65, the variable n is updated in the order of 1 → 2 → 3 → 1 → 2 →. In subsequent S67, the output switch 45 is switched to the terminal Tn, and the flow shifts to S63, where the above-described processing is repeated. After outputting the picture B1 by the above processing (S57), the output switch 45 is switched to the terminal T2 (S57).
67) After outputting the picture B2 (S57), the output switch 45 can be switched to the terminal T3 (S6).
7). Then, after outputting the picture B3 (S5
7) When the output switch 45 is switched to the terminal T1, pictures P2 and B4 can be output (S57). During this time, a negative determination is made once in S61 (picture P2).

As described above, in the present moving image data compression apparatus,
Despite the fact that the compression processing related to the B picture is shared by the three compressors 1 to 3, the output stream is the same as when the compression is performed by one compressor. For this reason, the compression processing can be sped up according to the above assignment. Also,
In the present moving image data compression apparatus, the compression processing is speeded up by thinning out the frames, so to speak, without depending on the screen division, so that the motion vector and the like can be sufficiently used, and the compression ratio does not decrease. For this reason, it is possible to prevent the image quality from lowering. Therefore, in the present moving image data compression apparatus, it is possible to satisfactorily speed up the compression processing of the video data without deteriorating the image quality of the reproduced image.

Further, in the present moving picture data compression apparatus, the compressors 1 to 3 commonly compress the I picture and the P picture. Therefore, each of the compressors 1 to 3 can independently compress video data without exchanging data with the other compressors 1 to 3. In addition, each frame constituting the video data is distributed by the input switch 5, and each of the compressors 1 to 3 does not need to determine whether or not each frame itself is compressed. For this reason,
Each of the compressors 1 to 3 only needs to compress the data of the frame input via the input switch 5 as usual. Therefore, in the present moving image data compression apparatus, the processing of the compressors 1 to 3 can be extremely simplified, and the compression processing of the video data can be further speeded up.
Alternatively, the manufacturing cost can be reduced by using an existing compressor.

Further, the moving picture data compression apparatus according to the present invention has a picture cutout unit 22 as a discarding means for discarding data commonly compressed by each of the compressors 1 to 3 except for the data compressed by one compressor 1. , 23, and an output switch 45 as rearrangement means for arranging the data sorted by the input switch 5 in the original order. For this reason, it takes a longer time to compress some frames (especially, frames that become B pictures) than the others,
Even if the order of the data output by 3 is different from the input order,
Compressed data can be output in the correct order. Therefore, in the present moving image data compression apparatus, the reliability of the compression processing can be further improved.

It should be noted that the present invention is not limited to the above-described embodiment, and can be implemented in various forms without departing from the gist of the present invention. For example, each of the input switch 5 and the input controller 7, or the output switch 45 and the output controller 47 can be configured by one computer. When the input switch 5 and the input controller 7 are configured by one computer,
Video data is input to an input port, and the compressors 1, 2, and 3 are connected to three different output ports. Then, if the processing of FIG. 5 is executed after replacing the ON / OFF of the switches S1 to S3 with the switching of the output port, the same operation and effect as those of the above-described embodiment occur. Further, the number of compressors may be smaller than the number of B pictures consecutively arranged (3 in the above embodiment). Further, the present invention relates to MPEG1, H.264. For example, the present invention can be similarly applied to an apparatus for compressing moving image data based on another standard such as H.261.

Further, as the storage medium according to the fourth aspect of the invention, various forms can be considered in addition to elements such as a ROM and a RAM (the ROM of the input controller 7 in the above embodiment). For example, a CD-ROM, a floppy disk, or the like, a program cartridge that can be inserted into a card slot, or a file server on the Internet may be used. Further, in the above embodiment, the case of the frame structure (frame picture) is taken as an example, but the present invention can be similarly applied to the case of the field structure (field picture). Furthermore, various forms are also conceivable for the distribution of frames to the compressors 1 to 3, for example, I,
The distribution may be performed for each GOP obtained by combining the P and B frames.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a moving image data compression device to which the present invention has been applied.

FIG. 2 is an explanatory diagram showing a configuration and an operation of a compressor of the moving image data compression device.

FIG. 3 is an explanatory diagram illustrating the distribution of input frames by the compressor.

FIG. 4 is an explanatory diagram illustrating the sharing of compression processing by each compressor.

FIG. 5 is a flowchart showing a process of an input controller of the device.

FIG. 6 is a flowchart illustrating a process of a picture cutout device of the device.

FIG. 7 is a flowchart showing a process of another picture cutout device of the above device.

FIG. 8 is a flowchart showing a process of an output controller of the device.

[Explanation of symbols]

1,2,3 ... Compressor 5 ... Input switch
7. Input controller 21, 22, 23 ... Picture cutout unit 45 ... Output switch 47 ... Output controller

Continued on the front page F term (reference) 5C059 KK13 KK15 LB07 MA00 MA05 MA14 MA23 MC11 MC38 ME08 PP05 PP06 PP07 SS06 SS20 UA02 UA32 UA33 UA39

Claims (4)

[Claims] 1. A moving image data compression apparatus for compressing moving image data representing a moving image by a plurality of frames corresponding to a lapse of time, wherein only the moving image data relating to a part of frames of the moving image data is compressed. A moving image data compression apparatus, comprising a plurality of compression means, wherein the union of the frames for which each compression means compresses the moving image data coincides with the whole of the moving image data. 2. The moving picture data of a specific frame referred to when each of the compression means compresses the moving picture data using an inter-frame difference of the moving picture data and calculates the inter-frame difference relating to another frame. 2. The compression means commonly compresses
The moving image data compression device according to the above. 3. The method according to claim 1, further comprising: distributing each frame constituting the moving image data to each of the compression units for compressing the moving image data relating to the frame, and inputting the input to the corresponding compression unit. 3. The moving image data compression device according to claim 1, wherein 4. A storage medium storing a software program for controlling a plurality of compression means for compressing moving image data representing a moving image by a plurality of frames corresponding to a lapse of time, wherein each of the compression means includes A software program for executing a selection process of compressing only the moving image data relating to some frames of the moving image data is stored, and the union of the frames compressed by the respective compression units according to the selection process is represented by A storage medium characterized by being coincident with the entire video data.