JP2008124931A - Data converting device, data converting program and data converting method, and data decoding device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a data converting device converting compressed encoded data into data which can be accessed fast at random without providing an index file as an independent file. <P>SOLUTION: The data converting device 1 includes a GOP detecting means 10 of detecting a GOP from the compressed encoded data, a frame detecting means 11 of detecting a frame in the GOP, a data length accumulating means 12 of accumulating a data length by a predetermined number, and a dummy data adding means 14 of adding dummy data to data of the group by the difference between a predetermined data length and the accumulated data length when the accumulated data length is less than the predetermined data length. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a data conversion device, a data conversion program and a data conversion method, and a data decoding device that convert compressed and encoded data into data for nonlinear editing.

In recent years, video and audio editing has been performed by playing back a tape on which video or the like has been recorded, copying the desired location to another tape, writing the video or the like as digital data in a storage device such as a hard disk, Shifting to non-linear editing that is performed by a computer.
This non-linear editing system performs non-linear editing such as “copy”, “cut”, “paste”, etc., on the GUI (Graphical User Interface) of a computer terminal by an operator. By doing the above, edit data composed of the start point and end point of video and the like is created. When the edited video or the like is reproduced, the nonlinear editing system reads out and outputs the video or the like between desired start points and end points from the storage device based on the edited data. Thus, the non-linear editing can reduce the time required for the editing as compared with the linear editing.
The video is generally data compression-coded in MPEG2 (Moving Picture Coding Experts Group 2) format. For example, in the HDV (digital high definition video) standard for recording and reproducing digital high-definition video, the MPEG2 Long GOP method is adopted.

Here, the MPEG2 encoding format will be briefly described with reference to FIG. FIG. 9 is a block diagram showing the structure of MPEG2 stream data.
As shown in FIG. 9, MPEG2 stream data includes 6 “sequence layer”, “GOP (Group of Picture) layer”, “picture layer”, “slice layer”, “macroblock layer”, and “block layer”. Consists of layers. Here, the GOP is composed of one or more pictures and includes an I picture (intra-frame encoded image) that is encoded without using a reference image. Further, GOP is a P picture (inter-frame encoded image) that is predictively encoded from a picture that is located in the past in time, or a B picture that is predictively encoded from pictures that are located in the past and future in time (both Direction-coded images) are also included.

In the MPEG2 Long GOP system of the HDV standard, generally, video is compression-encoded at about 25 Mbps so as to form a group (GOP) of compressed data for 15 frames. In this case, the size of the GOP varies with the difference in the compressed data size for each frame. Actually, the size of the GOP differs by about ± 15 to 20%.
As described above, since the encoding method of MPEG2 cannot uniquely specify the head address of each GOP due to the difference in size, when it is desired to access a desired GOP, the GOP is sequentially added from the head of the MPEG2 stream data. It is necessary to perform a search process such as searching or searching for a GOP before and after a position predicted in advance, which is not a method suitable for nonlinear editing.
Therefore, a technique is disclosed that enables direct access to a GOP or the like by creating an address of the GOP or the like as an index file in MPEG2 stream data (see Patent Document 1).
JP-T-2006-524410

However, the above-described method of creating the GOP position as an index file requires that the MPEG2 stream data and the index file be managed separately in association with each other. There's a problem.
Also, if the stream data and the index file are managed as separate files, the GOP address cannot be uniquely specified when the index file disappears, so it is necessary to detect the GOP from the beginning of the stream data, There is a problem that a desired GOP cannot be accessed at high speed.

  The present invention has been made to solve the above-described problems, and converts compression-encoded data into data that can be randomly accessed at high speed without providing an index file as a separate file. An object of the present invention is to provide a data conversion device, a data conversion program and a data conversion method, and a data decoding device for decoding data-converted editing data.

  The present invention was devised to achieve the above object. First, the data conversion apparatus according to claim 1, wherein a predetermined number of frames are grouped into a compression code that is compressed and encoded for each frame. A data conversion apparatus for converting the digitized data into editing data that can be randomly accessed, and includes a group detection means, a frame detection means, a data length accumulation means, and a dummy data addition means.

In such a configuration, the data conversion apparatus detects a group from the compression-encoded data based on the group header information assigned in advance for each group by the group detection unit. Then, the data conversion apparatus detects the frame by the frame detection means based on the frame header information assigned in advance for each frame in the group detected by the group detection means. As described above, by detecting the group and the frame based on each header information, the data conversion apparatus analyzes the compression encoded data having the variable length data structure.
When the compression encoded data is encoded by, for example, an MPEG2 encoding method, the group corresponds to a GOP and the frame corresponds to a picture.

Then, the data conversion apparatus accumulates a predetermined number of data lengths as the accumulated data length for each frame by the data length accumulating means. As a result, the data length of the entire group is accumulated.
When the data length of the entire group (cumulative data length) is less than a predetermined prescribed data length by the dummy data adding means, the data conversion device adds the prescribed data length and the accumulated data length to the group data. The dummy data of the difference is added. As a result, the data length of each group is fixed to the same length throughout the entire compression encoded data.

  The data conversion device according to claim 2 is the data conversion device according to claim 1, further comprising an offset information adding unit.

  In such a configuration, the data conversion apparatus adds offset information indicating an offset from the predetermined reference position of the group to the compressed encoded data for each frame in the group by the offset information adding unit. This makes it possible to recognize the position of the frame by referring to the offset information within each group.

  Furthermore, the data decoding device according to claim 3 is a data decoding device for decoding an arbitrary frame from editing data consisting of a plurality of groups having a predetermined prescribed data length configured by compressing and encoding a predetermined number of frames. And it was set as the structure provided with a group specific | specification means, an offset information reading means, and a frame decoding means.

In such a configuration, the data decoding apparatus specifies a group having the decoding target frame in the editing data based on the frame number of the decoding target frame by the group specifying unit.
Thus, since the number of frames included in the group is determined in advance, the group specifying means can specify the number of the group including the frame from the top by the frame number. Further, since each group has a specified data length and is fixed to the same length, the position (address) of the group in the editing data can be specified.
Then, the data decoding apparatus reads the offset information added in the group specified by the group specifying means by the offset information reading means.
Then, the data decoding apparatus decodes the decoding target frame based on the offset information by the frame decoding means.

  According to a fourth aspect of the present invention, there is provided a data conversion program for converting a compression-encoded data, which is compression-encoded for each frame into a predetermined number of frames as a group, into editing data that can be randomly accessed. Are configured to function as group detection means, frame detection means, data length accumulation means, and dummy data addition means.

In such a configuration, the data conversion program detects the group from the compression encoded data based on the group header information given in advance for each group by the group detection unit. Then, the data conversion program detects the frame by the frame detection unit based on the frame header information added in advance for each frame in the group detected by the group detection unit.
Then, the data conversion program accumulates a predetermined number of data lengths as the accumulated data length for each frame by the data length accumulating means. As a result, the data length of the entire group is accumulated.
Then, when the data length of the entire group (cumulative data length) is less than the predetermined prescribed data length by the dummy data adding means, the data conversion program adds the prescribed data length and the accumulated data length to the group data. The dummy data of the difference is added. As a result, the data length of each group is fixed to the same length throughout the entire compression encoded data.

  Furthermore, the data conversion program according to claim 5 is configured such that, in the data conversion program according to claim 4, the computer further functions as an offset information adding unit.

  In such a configuration, the data conversion program adds offset information indicating an offset from the predetermined reference position of the group to the compressed encoded data for each frame in the group by the offset information adding unit. This makes it possible to recognize the position of the frame by referring to the offset information within each group.

  The data conversion method according to claim 6, wherein a predetermined number of frames are grouped, and the compression encoded data compressed and encoded for each frame is converted into editing data that can be randomly accessed. The procedure includes a group detection step, a frame detection step, an offset addition step, a data length accumulation step, and a dummy data addition step.

In this procedure, in the data conversion method, in the group detection step, a group is detected from the compression-encoded data based on the group header information previously assigned to each group by the group detection means.
In the data conversion method, in the frame detection step, the frame detection unit detects a frame based on the frame header information added in advance for each frame in the group detected by the group detection unit.
In the data conversion method, in the offset addition step, offset information indicating an offset from the predetermined reference position of the group to the compressed encoded data for each frame in the group is added to the group by the offset information adding unit. .

Further, in the data conversion method, in a data length accumulation step, a predetermined number of data lengths are accumulated for each frame as the accumulated data length. As a result, the data length of the entire group is accumulated.
In the data conversion method, in the dummy data adding step, when the accumulated data length accumulated by the data length accumulating means by the dummy data adding means is less than a predetermined prescribed data length, The dummy data of the difference between the specified data length and the cumulative data length is added. As a result, the data length of each group is fixed to the same length throughout the entire compression encoded data.

The present invention has the following excellent effects.
According to the first or fourth aspect of the present invention, the compression-encoded data that is compression-encoded for each frame with a predetermined number of frames as a group becomes fixed-length data having the same length for each group. Can be converted to editing data. Since the editing data thus converted has the same group data length, it is possible to access the group at high speed without providing an index file.

  According to the invention of claim 2, claim 5 or claim 6, a predetermined number of frames are grouped, and compression encoded data that is compression encoded for each frame is fixed to the same length for each group. In addition to the increase in length, the offset information for the frames in the group can be converted into editing data describing the information. This makes it possible to randomly access groups and files at high speed without providing an index file.

  According to the third aspect of the present invention, since it is not necessary to search the group or frame by decoding the editing data from the beginning, the group or frame can be specified at high speed, and the desired frame can be decoded at high speed. it can.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[Outline of data converter function]
First, the outline of the function of the data conversion apparatus will be described with reference to FIG. FIG. 1 is an explanatory diagram for explaining an outline of functions of a data conversion apparatus according to the present invention.
The data conversion apparatus 1 converts compression-encoded data (stream data), which is compression-encoded for each frame, into a group that includes a predetermined number of frames as editing data that can be randomly accessed.
Here, MPEG2 Long GOP stream data, which is the HDV standard, is used as the compression encoded data. In this case, the GOP layer of the compression-encoded data is a series of a plurality of GOPs in which the GOP header is the head and the compressed data for 15 frames is one group (GOP).

As shown in FIG. 1, in compression-coded data, the data lengths (L _1A , L _2A , L _3A , L _4A ,...) Of each GOP differ in principle due to the difference in compressed data length for each frame. Yes. In this case, in order to access each GOP, it is necessary to search for a desired GOP sequentially from the first GOP or to search for a GOP before and after a position predicted in advance, and it takes time to edit the compression-encoded data. become.
Therefore, the data conversion apparatus 1 generates editing data by adding dummy data to each GOP of the compression-encoded data and fixing the data length to the same length.

As described above, as the editing data, the data length (L _1B , L _2B , L _3B , L _4B ,...) Of each GOP is fixed to the same length, so that the position (address) of the GOP is It becomes possible to calculate by a multiple of the fixed length size. This makes it possible to directly access the GOP and to perform editing at high speed.
Hereinafter, the configuration of the data conversion apparatus 1 that realizes this function will be specifically described.

[Configuration of Data Conversion Device: First Embodiment]
Here, the configuration of the data conversion apparatus according to the first embodiment of the present invention will be described with reference to FIG. FIG. 2 is a block configuration diagram showing the configuration of the data conversion apparatus according to the first embodiment of the present invention. The data conversion apparatus 1 includes a GOP detection unit 10, a frame detection unit 11, a data length accumulation unit 12, a data length difference calculation unit 13, and a dummy data addition unit 14.

The GOP detection means (group detection means) 10 analyzes the input compressed encoded data (stream data) and detects the GOP. Here, the GOP detection means 10 detects a GOP by detecting a unique code predetermined in MPEG2 as a GOP header (GOP header; group header information) of the GOP layer.
The GOP detection means 10 accumulates stream data in a storage means such as a memory (not shown), and 1 GOP is stored in the frame detection means 11 and the dummy data addition means 14 when 1 GOP of stream data is accumulated. Output stream data for minutes. Note that only the position (address) of the GOP in the memory may be output as GOP detection information to the frame detection unit 11 and the dummy data addition unit 14.

The frame detection means 11 detects a compression-coded frame in the GOP detected by the GOP detection means 10. Here, the frame detection means 11 detects a frame by detecting a unique code predetermined in MPEG2 as a header (picture header; frame header information) of a picture (frame) in a picture layer.
The frame detection means 11 outputs (notifies) the detected frame position (address) to the data length accumulation means 12 as frame detection information.

  The data length accumulating unit 12 accumulates a predetermined number of data lengths for each frame detected by the frame detecting unit 11. Here, the data length accumulating unit 12 accumulates the data length of the frame by the number of GOP frames (here, 15 frames) based on the frame detection information (frame position) notified from the frame detecting unit 11. To do. The accumulated data length indicates the data length (GOP data length) of the total GOP that has been accumulated. The GOP data length accumulated by the data length accumulation unit 12 is output to the data length difference calculation unit 13.

The data length difference calculating means 13 calculates a difference between a predetermined GOP prescribed data length and the GOP data length (cumulative data length) accumulated by the data length accumulating means 12.
Note that a data length longer than the maximum data length of the GOP predicted in advance is used as the specified data length. For example, when 1 GOP of compressed encoded data is configured at 25 Mbps in conformity with the HDV standard, the specified data is set so that 1 GOP is 30 Mbps in consideration of a difference of about ± 15 to 20% of compressed data. Set the length.
Then, the data length difference calculating unit 13 outputs the data length difference between the specified data length and the GOP data length to the dummy data adding unit 14.

  The dummy data adding means 14 adds dummy data of the data length difference between the specified data length calculated by the data length difference calculating means 13 and the GOP data length for each GOP detected by the GOP detecting means 10. . The GOP to which the dummy data is added is output to the outside as editing data. As the dummy data, a value “0” recommended in MPEG2 or the like is used.

  By configuring the data conversion apparatus 1 as described above, the data conversion apparatus 1 can convert the compression-encoded data into editing data in which the GOP is fixed to the same length. In this way, by fixing the GOP, when accessing a desired GOP, a GOP such as searching sequentially from the first GOP as in the past or searching for a GOP before or after a previously predicted position is used. It is possible to directly access a desired GOP by obtaining an address that is a multiple of the GOP data length.

[Outline of additional functions of the data converter]
Next, an overview of additional functions of the data conversion apparatus will be described with reference to FIG. FIG. 3 is an explanatory diagram for explaining the outline of the additional function of the data conversion apparatus according to the present invention.
In addition to the functions described with reference to FIG. 1, the data conversion apparatus 1B converts compression-encoded data (stream data) into editing data that allows random access to a desired frame within each GOP.

As shown in FIG. 3, each GOP of compression-encoded data is composed of data of 15 frames (I, B, and P pictures) having different data lengths. In FIG. 3, subscripts added to I, B, and P indicate the storage order of frames in the GOP. In this case, since each frame (picture) has a different data length, in order to access each frame, it is necessary to search for a desired frame sequentially from the _first frame (I ₁ ). It will take time.

Therefore, the data conversion apparatus 1B provides a user data area in the GOP of the compression encoded data, and describes the offset information (address) of each frame (picture) in the user data area, thereby editing the editing data. Is generated. For example, offset information of I picture (I ₁ ), B picture (B ₂ ), B picture (B ₃ ),... Is described in order from the top of the user data area.
Thus, by describing offset information of each frame in the GOP as editing data, each frame (picture) can be directly accessed, and editing can be performed at high speed.
Hereinafter, the configuration of the data conversion apparatus 1B that realizes this function will be described in detail.

[Configuration of Data Conversion Device: Second Embodiment]
Here, with reference to FIG. 4, the structure of the data converter which concerns on 2nd Embodiment of this invention is demonstrated. FIG. 4 is a block configuration diagram showing the configuration of the data conversion apparatus according to the second embodiment of the present invention. The data converter 1B includes a GOP detection unit 10, a frame detection unit 11, a data length accumulation unit 12, a data length difference calculation unit 13, a dummy data addition unit 14, an offset information generation unit 15, and an offset information addition Means 16. Since the configuration other than the offset information generation unit 15 and the offset information addition unit 16 is the same as the configuration of the data conversion apparatus 1 described with reference to FIG. 2, the same reference numerals are given and description thereof is omitted.

  The offset information generation unit 15 generates offset information in the GOP including the frame for each frame output from the frame detection unit 11. Here, based on the frame detection information output from the frame detection unit 11, the offset information generation unit 15 generates a relative address from a predetermined reference position in the GOP as offset information. The reference position may be anywhere as long as it is a fixed position in the GOP, such as the start address of the GOP and the start address of the user data area (see FIG. 3). The offset information generated by the offset information generating unit 15 is output to the offset information adding unit 16.

The offset information adding unit 16 adds the offset information generated by the offset information generating unit 15 to the user data area in the GOP. Here, the offset information adding means 16 sets the user data area in the stream data for 1 GOP output from the GOP detecting means 10, and sequentially writes the offset information output from the offset information generating means 15 to the user data area. .
Further, the offset information adding means 16 writes the offset information of the frame for 1 GOP in the user data area, and then outputs the stream data for 1 GOP to the dummy data adding means 14.

The offset information adding means 16 writes the offset information in the user data area if the user data area already exists in the stream data output from the GOP detecting means 10, and if the user data area does not exist, A GOP to which the area is added is generated, and offset information is written in the added user data area.
As described above, when a user data area is added, it is desirable that the specified data length used in the data length difference calculating unit 13 is also added for the data length of the user data area.

  By configuring the data conversion apparatus 1B as described above, the data conversion apparatus 1B converts the compression-encoded data into editing data in which the GOP is fixed to the same length, and in GOP units. , Offset information to a frame in the GOP can be added. As a result, when accessing a desired GOP, there is no need to perform a GOP search process such as sequentially searching from the first GOP as in the prior art or searching for a GOP before or after a previously predicted position. By obtaining an address that is a multiple of the data length, it is possible to directly access a desired GOP. Further, when accessing a frame in the GOP, it is not necessary to sequentially search from the first frame, and it becomes possible to directly access a desired frame based on the offset information.

[Operation of data converter]
Next, the operation of the data conversion apparatus will be described with reference to FIG. FIG. 5 is a flowchart showing the operation of the data conversion apparatus according to the second embodiment of the present invention.
First, the data conversion apparatus 1B inputs the compression encoded data by the GOP detection means 10 (step S1).

Then, the data conversion apparatus 1B detects the GOP by analyzing the compression-encoded data by the GOP detection means 10 and searching for a unique code predetermined as a GOP header (group header information) (step) S2: Group detection step). If no GOP is detected (No in step S2), the data conversion apparatus 1B determines that the input of the compressed encoded data has been completed, and ends the operation.
On the other hand, when a GOP is detected (Yes in step S2), the data conversion apparatus 1B initializes “GOP data length” and “number of frames i”, which are internally managed variables, to “0” (step S3). ).

  After that, the data conversion apparatus 1B uses the frame detection unit 11 to search for a unique code predetermined as a frame header (picture header; frame header information) in the GOP, and thereby compress the encoded frame. It detects (step S4; frame detection step).

  In the data conversion apparatus 1B, the offset information generating unit 15 generates offset information in the GOP including the frame, and the offset information adding unit 16 adds the offset information to the user data area in the GOP. (Step S5; offset adding step). At this time, the frame detection means 11 adds “1” to the variable “number of frames i” (step S6).

Further, the data conversion apparatus 1B accumulates the data length for each frame in the variable “GOP data length” by the data length accumulation means 12 (step S7; data length accumulation step).
Here, the data conversion apparatus 1B determines whether or not the variable “number of frames i” has reached a predetermined number (here, “15”) (step S8). If the variable “number of frames i” does not reach “15” (No in step S8), the data conversion apparatus 1B returns to step S4 and continues the operation. As a result, the data length of the GOP composed of 15 frames is accumulated in step S7.
On the other hand, when the variable “number of frames i” reaches “15” (Yes in step S8), the data conversion apparatus 1B uses the data length difference calculation means 13 to determine the predetermined data length of the GOP determined in advance in step S7. The difference from the accumulated “GOP data length” is calculated (step S9).

Then, the data converter 1B adds and outputs the difference calculated in step S9 as dummy data for each GOP detected in step S2 by the dummy data adding means 14 (step S10; dummy data adding step). .
Thereafter, the data conversion apparatus 1B returns to step S2 and continues the operation while the GOP is detected.

With the above operation, the data conversion apparatus 1B can convert the compression-encoded data into editing data that can be accessed randomly.
Here, the operation of the data conversion apparatus 1B of the second embodiment has been described, but the operation of the data conversion apparatus 1 (see FIG. 1) of the first embodiment omits only the process of step S5 in FIG. Therefore, the description is omitted.

  The configuration and operation example of the data conversion apparatus according to the embodiment of the present invention have been described above, but the present invention is not limited to this. Here, an example has been described in which MPEG2 Long GOP stream data is used as compression-encoded data. However, the compression-encoded data may be data that is compression-encoded for each frame and that is configured as a group of a predetermined number of frames. For example, the encoding method is not limited.

Further, the data conversion apparatuses 1 and 1B described with reference to FIGS. 2 and 4 function in a single configuration, but the apparatus can be configured by being incorporated in a nonlinear editing machine.
The data converters 1 and 1B can be configured by a general computer including a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), etc. (not shown). Can be operated by a data conversion program that functions as each means described above.

[Configuration of Data Decoding Device]
Next, the configuration of the data decoding device according to the embodiment of the present invention will be described with reference to FIG. FIG. 6 is a block diagram showing the structure of the data decoding apparatus according to the present invention.
The data decoding device 2 decodes the editing data converted by the data conversion device 1B (see FIG. 4) for each frame. Here, the data decoding apparatus 2 includes an editing data storage unit 20, a GOP specifying unit 21, and a decoding unit 22.

  The editing data storage means 20 stores editing data and is a general storage device such as a hard disk. The editing data storage means 20 stores the editing data converted by the data converter 1B (see FIG. 4).

The GOP specifying means (group specifying means) 21 in the editing data stored in the editing data storage means 20, based on a frame decoding instruction including a frame number input from outside, GOP including a frame) is specified. As described with reference to FIG. 1, the editing data has the GOP data length fixed to the same length. Therefore, the GOP specifying unit 21 specifies the GOP including the frame by calculating the absolute address of the GOP based on the frame number. Specifically, when the frame number is F _NUM , the number of frames in the GOP is F _MAX , and the data length (specified data length) of the fixed 1 GOP is L _CONST , the address AD of the GOP in which the frame is included _GOP is obtained by the following equation (1).

AD _GOP = (F _NUM ÷ F _MAX ) × L _CONST (1) Equation (1) However, in this equation (1), the part after the decimal point is rounded down.

  The GOP specifying means 21 outputs the GOP address calculated by the equation (1) to the decoding means 22 together with the frame number (instruction frame) as GOP specifying information.

  The decoding unit 22 decodes the frame in the GOP specified by the GOP specifying unit 21. The decoding unit 22 first decodes the first frame (I picture) of the GOP, and sequentially decodes the frames (B picture, P picture) according to a predetermined display order until the instruction frame is decoded. . This is because in MPEG2, the first frame of a GOP is an I picture (intra-frame encoded image) that can be decoded only with encoded data in the frame.

Here, the data structure of the GOP will be described with reference to FIG. FIG. 7 is an explanatory diagram for explaining the order of displaying pictures (frames) in an MPEG2 GOP.
As shown in FIG. 7A, in the MPEG2 Long GOP stream data, the first frame of the GOP is an I picture, and thereafter, a P picture (frame) predictively encoded from a picture located in the past in time. Inter-coded images) and B pictures (bi-directional coded images) predictively coded from pictures located in the past and the future in time. Note that B pictures and P pictures cannot be decoded by themselves, but are decoded sequentially after decoding an I picture. Therefore, in the MPEG2 Long GOP stream data, as shown in FIG. 7B, the picture storage order and the display order are different.

Returning to FIG. 6, the description of the configuration of the data decoding device 2 will be continued.
The decoding unit 22 includes an offset information reading unit 22a and a frame decoding unit 22b.

The offset information reading means 22a reads offset information corresponding to the frame number in the GOP specified by the GOP specifying means 21. Here, when the GOP is specified by the GOP specifying means 21, the offset information reading means 22a reads the offset information corresponding to the first frame in the user data area described with reference to FIG. 3, and sends it to the frame decoding means 22b. Output.
Further, the offset information reading unit 22a receives the frame number from the frame decoding unit 22b, reads the offset information corresponding to the frame number, and outputs it to the frame decoding unit 22b.

The frame decoding means 22b reads a frame (picture) corresponding to the offset information read by the offset information reading means 22a from the editing data storage means 20 and decodes it. When the decoded frame is not an instruction frame instructed from the outside, the frame decoding means 22b further notifies the offset information reading means 22a of the frame number of the frame to be decoded next after the decoded frame. The target instruction frame is decoded.
In MPEG2, by decoding a frame, it is possible to acquire information for specifying a frame to be decoded next. Therefore, the frame decoding unit 22b is based on the information until the instruction frame is decoded. A new frame number is output to the offset information reading means 22a.

  By configuring the data decoding device 2 as described above, the data decoding device 2 randomly selects a desired frame based on the frame number in the editing data converted by the data conversion device 1B (see FIG. 4). It becomes possible to access.

[Operation of Data Decoding Device]
Next, the operation of the data decoding apparatus will be described with reference to FIG. 8 (refer to FIG. 6 as appropriate). FIG. 8 is a flowchart showing the operation of the data decoding apparatus according to the present invention.
First, the data decoding apparatus 2 calculates the address of the GOP including the frame based on the frame number of the frame (instructed frame) instructed from the outside and the specified data length of the GOP by the GOP specifying unit 21. The GOP is specified by (step S21).

Then, the data decoding apparatus 2 sets the first frame (I picture) of the GOP as a decoded frame by the decoding unit 22 (step S22).
Then, the data decoding device 2 uses the offset information reading unit 22a of the decoding unit 22 to decode the decoded frame from the user data area of the GOP specified in step S21 in the editing data stored in the editing data storage unit 20. Is read out (step S23).

Thereafter, the data decoding device 2 reads the compressed data of the corresponding frame from the editing data storage unit 20 based on the offset information read in step S23 by the frame decoding unit 22b of the decoding unit 22 and decodes it ( Step S24).
Then, the data decoding device 2 determines whether the decoded frame is the same frame as the instructed frame (instructed frame) by the frame decoding means 22b (step S25).

Here, when the decoded frame and the instruction frame are different frames (No in step S25), the data decoding apparatus 2 sets the next frame to be decoded by the frame decoding unit 22b (step S26), and the process proceeds to step S23. Return and continue operation.
On the other hand, when the decoded frame and the instruction frame are the same frame (Yes in step S25), the target instruction frame has been decoded, and thus the operation ends.

  By the above operation, the data decoding device 2 accesses the compressed data of the frame based on the GOP address and the offset information in the GOP in the editing data converted by the data conversion device 1B (see FIG. 4). Therefore, the frame can be decoded at high speed.

The configuration and operation example of the data decoding apparatus according to the embodiment of the present invention have been described above, but the present invention is not limited to this.
Here, the data decoding device 2 has been described as functioning in a single configuration, but it is also possible to configure the device by incorporating it in a non-linear editing machine together with the data conversion device 1B (see FIG. 4).
The data decoding apparatus 2 can be configured by a general computer including a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), etc. (not shown). It can be operated by an editing data decoding program that functions as each means.

It is explanatory drawing for demonstrating the outline | summary of the function of the data converter which concerns on this invention. It is a block block diagram which shows the structure of the data converter which concerns on 1st Embodiment of this invention. It is explanatory drawing for demonstrating the outline | summary of the additional function of the data converter which concerns on this invention. It is a block block diagram which shows the structure of the data converter which concerns on 2nd Embodiment of this invention. It is a flowchart which shows operation | movement of the data converter which concerns on 2nd Embodiment of this invention. It is a block block diagram which shows the structure of the data decoding apparatus which concerns on this invention. It is explanatory drawing for demonstrating the order which displays the picture (frame) in GOP of MPEG2. It is a flowchart which shows operation | movement of the data decoding apparatus based on this invention. It is a block diagram which shows the structure of the stream data of MPEG2.

Explanation of symbols

1, 1B data conversion device 10 GOP detection means (group detection means)
DESCRIPTION OF SYMBOLS 11 Frame detection means 12 Data length accumulation means 13 Data length difference calculation means 14 Dummy data addition means 15 Offset information generation means 16 Offset information addition means 2 Data decoder 20 Editing data storage means 21 GOP specification means (group specification means)
22 Decoding means 22a Offset information reading means 22b Frame decoding means

Claims (6)

A data conversion apparatus that converts compression-encoded data compressed and encoded for each frame into a group that includes a predetermined number of frames as editing data that can be randomly accessed,
Group detection means for detecting the group from the compressed encoded data based on group header information assigned in advance for each group;
In the group detected by the group detection means, frame detection means for detecting the frame based on frame header information given in advance for each frame;
For each of the frames detected by the frame detection means, data length accumulation means for accumulating the predetermined number of data lengths as the accumulated data length;
When the accumulated data length accumulated by the data length accumulating means is less than a predetermined prescribed data length, dummy data of a difference between the prescribed data length and the accumulated data length is added to the group data. Dummy data adding means;
A data conversion device comprising:   The apparatus further comprises offset information adding means for adding offset information indicating an offset from the predetermined reference position of the group to the compressed encoded data for each frame in the group. The data conversion apparatus according to 1. A data decoding device that decodes an arbitrary frame from editing data composed of a plurality of groups having a predetermined prescribed data length configured by compressing and encoding a predetermined number of frames,
Group specifying means for specifying a group having the decoding target frame in the editing data based on a frame number of the decoding target frame;
Offset information reading means for reading offset information added in the group specified by the group specifying means;
Frame decoding means for decoding the decoding target frame based on the offset information read by the offset information reading means;
A data decoding apparatus comprising: In order to convert a predetermined number of frames into a group and to convert the compression-encoded data compression-encoded for each frame into editing data that can be accessed randomly,
Group detection means for detecting the group from the compressed and encoded data based on group header information assigned in advance for each group;
In the group detected by the group detection means, a frame detection means for detecting the frame based on frame header information given in advance for each frame,
Data length accumulation means for accumulating the predetermined number of data lengths as the accumulated data length for each of the frames detected by the frame detection means,
When the accumulated data length accumulated by the data length accumulating means is less than a predetermined prescribed data length, dummy data of a difference between the prescribed data length and the accumulated data length is added to the group data. Dummy data adding means,
A data conversion program characterized by functioning as The computer,
The offset information adding means for adding offset information indicating an offset from the predetermined reference position of the group to the compressed encoded data for each frame in the group is further made to function as offset information adding means. Item 5. A data conversion program according to Item 4. A data conversion method for converting compression-encoded data, which is compression-encoded for each frame into a predetermined number of frames as a group, and editing data that can be randomly accessed,
A group detection step of detecting the group from the compressed encoded data based on group header information assigned in advance for each group by a group detection unit;
A frame detection step of detecting the frame based on frame header information given in advance for each frame in the group detected by the group detection unit by a frame detection unit;
An offset addition step of adding offset information indicating an offset from the predetermined reference position of the group to the compression-coded data for each frame in the group by the offset information addition unit;
A data length totaling step of totaling the predetermined number of data lengths as the total data length for each frame by the data length totaling means;
When the accumulated data length accumulated by the data length accumulating means by the dummy data adding means is less than a predetermined prescribed data length, the difference between the prescribed data length and the accumulated data length is added to the group data. A dummy data adding step for adding the dummy data;
A data conversion method comprising:

Images