JP3507990B2 - Moving image reproducing apparatus and moving image recording / reproducing apparatus - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital video disk reproducing apparatus using a read-only disk medium such as a CD-ROM and a digital video disk recording / reproducing apparatus using a rewritable disk medium such as a magneto-optical disk. it relates moving image reproduction apparatus and a moving image recording and reproducing apparatus such as a.

[0002]

2. Description of the Related Art Digital moving image data is recorded on a magnetic disk,
When recording on a disk medium such as an optical disk or a magneto-optical disk, by encoding image data with high efficiency, the amount of data to be recorded can be reduced without visually impairing the image quality.

By reducing the amount of data to be recorded by using the high efficiency encoding, recording can be performed for a long time as compared with the case where the high efficiency encoding is not used, and a medium having a low data reading speed or a low data writing speed can be used. Can also be used.

When the methods for encoding image data with high efficiency are roughly classified, intra-frame encoding or intra-field encoding for removing the spatial redundancy in the data of one screen, and a plurality of temporally continuous ones. There is interframe coding or interfield coding that uses image data and also removes temporal redundancy. In general, interframe coding and interfield coding also use temporal redundancy, and thus more efficient coding is possible. Here, a frame refers to one screen, and a field refers to a frame selected every other line and divided into two screens. The difference between the inter-frame coding and the inter-field coding is only the difference in whether the unit used for coding is a frame or a field, and the principle used for coding is the same. Therefore,
Only inter-frame coding will be described below.

As an example of the conventional technique of the inter-frame coding system, there is the MPEG1 system described in "Multimedia New Era" (PP25-31, "Multimedia and Package", Senoo, TV Society Special Workshop). . MPEG is
It is a moving image compression method that is being standardized by ISO / IEC JTC1, and is a moving image compression method that is 2
There is a method. Hereinafter, MPEG1 and MPEG2 are collectively referred to as MPEG.

FIG. 16 shows the classification of images and the arrangement of encoded data in MPEG. In this method, a unit such as 12 frames or 15 frames is used as a GOP.
(Group Of Pictures).

One frame of GOP is called an I-picture (Intra Picture). In the I-picture, intra-frame coding is performed.

Frames at regular intervals such as every three frames in a GOP are called P-pictures (Predictive Pictur).
e). In the P-picture, a case where inter-frame predictive coding and a case where intra-frame coding is performed are used by adaptively switching, for each macroblock, a decoded image of a past I-picture or P-picture as a reference image. A macroblock is a unit obtained by dividing a frame into rectangles (for example, a 16 × 16 pixel unit), and interframe predictive coding is for coding the difference between a reference image and a target image.

A frame between I-pictures and P-pictures is called a B-picture (Bidirectional Picture), and intra-frame coding, preceding / posterior interframe predictive coding, and bidirectional interframe predictive coding are performed for each macroblock. The conversion is adaptively switched and used.

In the preceding interframe predictive coding, decoded images of past I-pictures and P-pictures are used as reference images,
The post-interframe predictive coding uses a decoded image of a future I-picture or P-picture as a reference image, and the bidirectional interframe predictive coding uses two past I and a future I as reference images.
-Decoded images of pictures and P-pictures are used. Furthermore, in the case of these inter-frame predictive coding, front / back inter-frame predictive coding, and bidirectional inter-frame predictive coding, motion compensation can be used, and more efficient coding is possible. Motion compensation is a method in which a motion vector between a reference image and a target image is obtained, and the reference image is shifted by that vector to obtain a difference between frames.

Since the MPEG system uses bidirectional prediction, the order of input data and the order of output encoded data are different, as shown in FIG. For example, in order to encode a certain B-picture, two decoded images of I and P-pictures before and after are required. Therefore, first, the preceding and following I and P-pictures are encoded and decoded, and then the B-picture is encoded. In addition, the amount of encoded data generally decreases in the order of I-picture, P-picture, and B-picture.

Here, consider the case where high-speed reproduction is performed using high-efficiency coded data.

In high-speed playback on a disc medium,
Data read and high-speed jump are repeated. Since the data cannot be read during the jump, the average amount of data obtained per unit time is smaller than that during normal reproduction, and the data becomes discontinuous.

In the high-efficiency coding method, a difference value from past data may be coded, and a reproduced image cannot be obtained unless the past data is obtained. That is, the reproduced image can be obtained only from a part of the data obtained discontinuously.

In the case of the MPEG system, encoding is performed in GOP units, and in order to reproduce this GOP, first the G
The I-picture, which is the only intra-frame encoded frame in the OP, must be reproduced. Since all the frames other than the I-picture are encoded with the difference value with the decoded image of the I-picture or the difference value with the image reproduced using the decoded image of the I-picture, the I-picture cannot be reproduced. , It becomes difficult to decode the GOP data. Therefore, it is necessary not only to simply jump at the time of high-speed reproduction, but also to access data from which a reproduced image can be obtained even when discontinuity occurs due to the jump.

[0016]

In the MPEG system, data for which a reproduced image can be obtained without using a reference image is I
-It is encoded data of a picture. This is because the I-picture is intra-frame coded. In addition, the P-picture can be played back if one reference image is obtained. On the other hand, in order to reproduce the B-picture, two reference images before and after are required, which makes it considerably difficult to reproduce the discontinuous data.

Therefore, in the case of the MPEG system, during high-speed reproduction, coded data of I-pictures or coded data of I-pictures and P-pictures following it are randomly accessed in sequence, and the coded data are consecutively accessed. You should read it.

However, in the case of the MPEG system, since variable length coding is used, the code amount of each picture generally changes, and the positions where I-pictures and P-pictures are recorded are not at equal intervals. Therefore, it is difficult to accurately and randomly access the I-picture or P-picture, and it becomes impossible to read only the encoded data of the I-picture or P-picture. For this reason, the number of I-pictures and P-pictures reproduced per unit time becomes very small.

The present invention has been devised in view of such circumstances, and when high-efficiency coded data is recorded, no reference frame is used or one reference frame is used. It is an object of the present invention to provide a moving image recording / reproducing apparatus that enables accurate access to encoded data that can be reproduced by itself.

[0020]

According to a first aspect of the present invention, there is provided a moving image reproducing apparatus for reproducing a moving image from a recording medium in which encoded moving image data is recorded, which is intra-frame encoded. Means for writing the management data recorded in a predetermined position of the recording medium to the memory, which indicates the recording position of the frame or the forward predictive coded frame, and intra-frame coding using the management data on the memory. Means for reading a data group including data or forward predictive encoded data from a recording medium, and detecting a header indicating the start of a frame from the read data group
The unnecessary data that can be obtained before the start of the frame.
Except that it is provided with means for selecting only intra-frame coded data or forward predictive coded data, means for decoding the selected data, and means for outputting the decoded data. is there.

According to the second invention of the present application, the code
Frame at the beginning of each frame of the encoded moving image data
A header is added and it is the read data group.
Intraframe coded data or forward predictive coded data.
Use the frame header when selecting only data
It is characterized by that.

Further , according to the third invention of the present application,
The management data is the first table that manages the video sequence.
And a second table for managing the recording area of the recording medium.
And intra-frame coding and forward prediction coding
A third table for managing the recording position of the encoded data of the frame
Cable sequence, which indicates the playback order of the video sequence.
Indicates the concatenation information and the playback order of the discontinuously recorded recording areas
Concatenation information, intra-frame coded frames and forward prediction
Including the connection information that indicates the connection of encoded frames
It is a feature.

According to a fourth invention of the present application, the pipe
Physical data is the first table that manages video sequences
And a second table for managing the recording area of the recording medium
And the recording position of the encoded data of the intra-frame encoded frame
A third table for managing the location and the forward predictive coding frame.
4th table for managing the recording position of encoded data of the system
Connected information indicating the playback order of the video sequence.
Information and concatenation indicating the playback order of the recording areas recorded discontinuously
A sequence showing the connection between the information and the intra-coded frame.
Connection information indicating the connection between the connection information and the forward prediction frame
It is characterized by including information.

According to the fifth invention of the present application, each of the above-mentioned tables
Add a flag to each word in
It is characterized by that.

According to the sixth invention of the present application, moving image recording / reproducing for encoding moving image data and recording it on a recording medium, and reproducing the moving image from the recording medium on which the encoded moving image data is recorded. The device is a device for inputting moving image data, a device for encoding the input moving image data, and a recording position of an intra-frame encoded frame or a forward predictive encoded frame as management data. Means for writing in the memory, means for recording the encoded data in the recording medium, means for recording the management data on the memory in the recording medium, and a recording medium recording the encoded moving image data. Means for writing the management data into a memory, and intraframe coded data or forward prediction coded data using the management data in the memory. Means for reading the data set from the storage media including, from the read out data group, frame
The header that indicates the start of the frame
A means for removing unnecessary unnecessary data and selecting only intraframe coded data or forward predictive coded data, means for decoding the selected data, and means for outputting the decoded data It is characterized by.

[0026]

According to the present invention, the recording medium is placed at a predetermined position.
Keyframes based on recorded management data
(Intra-frame coded data or forward predictive coded data
Data group including
Of the read data group or the data of the key frame part
Decrypt only. By acting in this way, the key
High-speed playback in which only frames are played sequentially, and recording media
Random access to play from any position on the
You can access exactly the data you want, even when you do
It

[0027]

[0028]

EXAMPLES In describing the present invention , a record and
Example of a moving image recording / playback device having both playback functions
And will be described in detail with reference to the drawings.

In the following embodiments, the MPEG method described in the section of the prior art is used as the encoding method. Also, I
-For each frame of picture and P-picture,
Data that manages the first sector number and the last sector number on the disc on which encoded data of one frame is recorded is called management data. Also, the management data is recorded in the TOC (Table Of Contents) portion of the disc. The TOC refers to a portion that stores information about the disc. Hereinafter, the management data recorded in the TOC will be referred to as TOC data.

[First Embodiment] This first embodiment is an embodiment of a first moving image recording / reproducing apparatus (claim 1) according to the present invention. FIG. 1 is a block diagram of a first moving image recording / reproducing apparatus. In FIG. 1, 1 is an encoder, 2 is a data processing unit, 3 is a disk controller, 4 is a memory unit, 5 is a selection circuit, 6 is a decoder, and 10 is a disk. The moving picture recording / reproducing apparatus of the first embodiment has an encoder 1 for highly efficiently encoding input data of a digital moving image, and the encoded data inputted from the encoder 1 at the time of data recording is divided into sector units. Then, the data processing unit 2 outputs the encoded data output from the disc 10 to the selection circuit 5 at the time of reproduction as recording data, and outputs the encoded data to the selection circuit 5, the disc controller 3 controlling the disc 10, and the TOC. A memory unit 4 for recording the TOC data stored therein,
A selection circuit 5 that selects necessary data from the data input from the data processing unit 2 and a decoder 6 that decodes the encoded data input from the selection circuit 5 and outputs a reproduced image.

First, the disk 10 is added to the moving image recording / reproducing apparatus.
The operation when is inserted will be described. The disk controller 3 outputs control data to the disk 10 so as to read the TOC portion of the disk 10. Then, the TOC data recorded in the TOC portion of the disc 10 is input to the data processing unit 2. The data processing unit 2 outputs the input TOC data to the memory unit 4. The memory unit 4 is composed of a plurality of memories and records the input TOC data in a predetermined memory position.

In this embodiment, the T recorded in the memory unit 4 is recorded.
The OC data includes not only the first sector number and the last sector number on the disc on which the encoded data of I-picture and P-picture used for high-speed reproduction are recorded, but also the data used for normal reproduction. Will be described later. Note that, below, the I-picture and the P-picture are collectively referred to as a key frame.

Normally, various information such as free area information on the disc may be recorded in the TOC portion of the disc 10 in addition to these data, but it is omitted here.

Next, the operation at the time of recording the encoded data will be described.

At the time of recording, digital moving image data is input to the encoder 1 and encoded by the MPEG system. Then, the encoded data is output to the data processing unit 2. The data processing unit 2 divides the encoded data input from the encoder 1 into sectors.

Then, the encoded data is output so as to be recorded in a predetermined sector on the disk 10 according to the sector number input from the disk controller 3. Further, when recording the encoded data of the video sequence, the data processing unit 2 stores the first sector number and the last sector number of the continuous recording area in the memory unit 4.
Output to.

Further, the data processing unit 2 is provided with a circuit for detecting a code indicating the start and end of the encoded data of the key frame (I, P-picture), and the encoded data of the key frame is recorded. The first sector number and the last sector number are output to the memory unit 4. Alternatively, a flag indicating the start and end of the encoded data of the key frame may be input from the encoder 1 without providing a circuit for detecting the code indicating the start and the end of the encoded data of the key frame. .

In the memory section 4, the sector number input from the data processing section 2 is recorded in a predetermined address of each table. The table will be described in detail later.

By this operation, when the encoded data is recorded on the disk 10, the disk 10 is always stored in the memory section 4.
The latest TOC corresponding to the encoded data recorded in
The data is recorded.

When new encoded data is recorded on the disk 10, the latest TOC data is recorded in the memory section 4, but the TOC of the disk 10 has not been updated yet. Therefore, for example, the TOC data recorded in the memory unit 4 needs to be recorded in the TOC of the disk 10 at the time when the recording is completed or when the disk 10 is taken out from the apparatus. Further, even if the encoded data is being recorded on the disc 10, the disc 10 should be updated periodically to update the TOC.
May be controlled.

In this procedure, the disk controller 3 controls the disk 10 so as to write data in the TOC of the disk 10, and controls the data in the memory section 4 to be sequentially input to the data processing section 2. In the data processing unit 2,
The data input from the memory unit 4 is output to the disk 10.

Next, the operation during normal reproduction will be described. Here, it is assumed that the memory unit 4 has already recorded the TOC data of the disc 10.

First, the first sector number and the last sector number of the area in which the encoded data is recorded are input from the memory section 4 to the disk controller 3. The encoded data may be divided and recorded in a plurality of areas on the medium (disk). In that case, the first sector number and the last sector number of the area are stored in the memory unit 4 for each area. Is input to the disk controller 3.

FIG. 2 shows an example in which the encoded data is divided and recorded in two areas indicated by diagonal lines. The actual recording area of the disk 10 has a plurality of concentric circles or spirals, but here, the horizontal axis represents the time axis. If the encoded data is read in the order of 1 and 2 in FIG. 2, the sector number at the position a is first output from the memory unit 4, and then the sector numbers are output in the order b, c, d. It

The disk controller 3 outputs disk control data so as to access a predetermined position of the disk 10 according to the sector number input from the memory section 4.

Continuous encoded data is input to the data processing unit 2 from the disk 10 and output to the selection circuit 5.

The selection circuit 5 is a circuit for removing unnecessary data from the input data.

Since data is normally obtained from the disk 10 in sector units, for example, in the case of reproducing from the middle of the recorded data and when the data of the frame for starting reproduction is recorded from the middle of the sector. , First, unnecessary data will be obtained. In the MPEG system, a fixed-length header indicating the start of encoded data of a frame is added to the beginning of the encoded data for each frame. The selection circuit 5 detects this fixed-length header, removes unnecessary data obtained before the start of the frame,
That is, the encoded data at the head of the frame is output to the decoder 6.

The decoder 6 is a circuit for decoding the data encoded by the encoder 1, and decodes the input data and outputs a reproduced image (digital moving image data).

The function of the selection circuit 5 may be provided in the decoder 6 without providing the selection circuit 5.

Next, the operation during high speed reproduction will be described.

During high speed reproduction, the disk controller 3
Controls the disk 10 to continuously access the encoded data of the key frame (I, P-picture). At that time, the TOC data recorded in the memory unit 4 is used.

First, when high speed reproduction is designated, TOC data corresponding to the encoded data to be reproduced from the memory section 4, that is, the sector number and the trailing end of the beginning of the encoded data of the key frame are recorded. The assigned sector number is output to the disk controller 3. The disk controller 3 uses the input sector number to control the disk 10 to access only the sectors in which the encoded data of the key frame (I, P-picture) is recorded.

Since the coded data of the key frame can be obtained in the data processing unit 2, the input coded data is output to the selection circuit 5 as in the normal reproduction. The selection circuit 5 and the decoder 6 also perform the same operation as the normal reproduction, and the reproduced image is output from the decoder 6.

Next, the structure of the memory section 4 will be described in more detail.

FIG. 3 is a block diagram of the memory unit 4 and corresponds to the above-described modified mode 1. The memory unit 4 is
It is composed of three memories 12, 13, 14 and a memory controller 11 for controlling them.

FIG. 4 shows an example of the contents of TOC data recorded in each memory 12, 13, 14.

Here, each data of the sequence table is recorded in the memory 12, the sector table is recorded in the memory 13, and the key frame table is recorded in the memory 14. The contents of each table will be described below.

The sequence table of the memory 12 has one word of data for one video sequence. Here, the video sequence indicates, for example, temporally continuous data from the start of recording to the end of recording. One-word data consists of a next sequence, which is a pointer indicating the next sequence table, and a table number, which is a pointer indicating the position of the sector table. The next sequence shows the reproduction order of the video sequence, and the table number shows the address of the memory 13 (that is, the sector table) in which the sector number corresponding to the video sequence is recorded. Here, the next sequence indicates the address of the memory 12, but the last data connected in the next sequence as a pointer has a special value indicating "end".

The sector table of the memory 13 has 1-word data for each continuous recording area on the disk 10. 1-word data includes a next area that is a pointer indicating the next sector table, a start sector number that is the start sector number of the continuous area, an end sector number that is the end sector number of the continuous area, and a start address of the key frame table. It consists of four start sector addresses as pointers indicating The sector table corresponding to 1-word data of the sequence table is a series of sector tables connected by the next area as a pointer from the sector table address indicated by the table number of the sequence table. The next area indicates the address of the memory 13, but the last data connected in the next area, which is a pointer, has a special value indicating "end". If no key frame (I, P-picture) is recorded in the continuous area indicated by 1-word data of the sector table, the start sector address has a special value indicating "no data".

The key frame table of the memory 14 has one key frame (I, P-picture) for each frame.
There is word data. One word of data is the next frame which is a pointer indicating the next key frame table,
It is composed of a K-start sector number indicating the sector number in which the head of the encoded data of the key frame is recorded and a K-end sector number indicating the sector number in which the rear end is recorded.
The next frame indicates the address of the memory 14, but the last data connected in the next frame as a pointer has a special value indicating "end".

In the example of FIG. 4, there is no information indicating whether the data recorded in the key frame table of the memory 14 indicates the I-picture encoded data or the P-picture encoded data. If the structure of the GOP does not change, since I-pictures are present at regular intervals and the video sequence starts from I-picture, which data is I-picture?
It can be determined whether the coded data of the picture is indicated. However, if the structure of the GOP changes, a flag indicating whether the encoded data of the I-picture or the encoded data of the P-picture is shown for each word data of the key frame table, as will be described later. Alternatively, a method of separating the key frame table into an I-picture table and a P-picture table is used.

FIG. 5 illustrates the relationship between the TOC data recorded in each table of the memory unit 4 and the recording area.

A plurality of video sequences can be recorded on the medium (disc), and one video sequence may be recorded in a continuous area on the disc or divided into a plurality of areas. Sometimes In the example of FIG. 5, three video sequences are recorded, and the video sequence 1 is divided into three areas and recorded. The reproduction order is assumed to be the order of the video sequences 1, 2 and 3.

In the case of the example in FIG. 5, the sequence table has three 1-word data corresponding to each of the video sequences 1, 2, and 3. The first 1-word data is the data corresponding to the video sequence 1, and the next sequence of the video sequence 1 is the video sequence 2.
1 word data corresponding to the video sequence 2
The next sequence of 1 indicates 1-word data corresponding to the video sequence 3. Then, the table number of the sequence table indicates the address of the first sector table corresponding to the video sequence. This management is performed by the memory 12 (sequence table) in FIG.

In the sector table, one word data corresponds to a continuous recording area on the medium (disk). That is, in the case of FIG. 5, there are three 1-word data in the sector table corresponding to the video sequence 1. As described above, when there are a plurality of sector tables, the area next to the sector table sequentially indicates the next sector table.
Of the 1-word data of the sector table, the start sector number indicates the first sector number of the continuous recording area, and the end sector number indicates the last sector number. Encoded data of a plurality of key frames (I, P-pictures) is recorded in this continuous recording area. Among these, the sector number in which the encoded data of the first key frame is recorded is managed by the key frame table indicated by the start sector address.

Among the 1-word data of the key frame table, the K-start sector number indicates the first sector number in which the encoded data of one key frame is recorded, and K-
The ending sector number indicates the last sector number. Then, the next frame sequentially shows the next key frame table data.

FIG. 6 shows an example of a special case where coded data of a key frame extends over an area managed by a plurality of sector tables. In this example, the coded data of the key frame i is recorded over three recording areas of the recording area 1, the recording area 2, and the recording area 3.

In the case of FIG. 6, the sector number in which the first encoded data of the key frame i is recorded is recorded in the K-start sector number of the last key frame table corresponding to the recording area 1, and K-end. The sector number stores the sector number in which the last encoded data is recorded,
The K-end sector number is the sector number of the recording area 3.
Further, since there is no key frame table corresponding to the recording area 2, data indicating "no data" is recorded at the start sector address of the recording area 2. Then, the start sector address of the recording area 3 indicates the key frame table of the key frame i + 1.

At the time of high speed reproduction, the sector number recorded in the key frame table of the memory section 4 is input to the disk controller 3, and the disk controller 3 outputs K-
The disk 10 is controlled to read data from the start sector number to the K-end sector number. At this time, when the key frame i shown in FIG. 6 is reproduced, a recording area having different K-start sector numbers and K-end sector numbers corresponding to the key frame i of the key frame table is shown as described above. It will be. However, even in such a case, it can be known that the recording area 1, the recording area 2, and the recording area 3 are reproduced in this order by the information recorded in the sector table. Therefore, the disk controller 3 uses the recording area 1, the recording area 2, and the recording area 3 in this order.
To read data up to the K-end sector number.

The operation of the memory section 4 of FIG. 1 will be described in more detail below.

When the disk 10 is inserted into the device,
The TOC data recorded on the disk 10 is sequentially input to the memory unit 4 and recorded in a predetermined memory of the memory unit 4.

FIG. 7 is a flow chart showing the operation of the memory section 4 when the encoded data is recorded on the disc 10. This shows updating of the TOC data in the memory unit 4 at the time of recording. In the sequence table of the memory 12 of FIG. 3, 1 is added to the trailing end of the already recorded data.
The word data is added, and the data of the sector table of the memory 13 and the key frame table of the memory 14 corresponding to the data of the added sequence table are additionally recorded.

In the following description, ad12 and adl12 are the addresses of the memory 12 shown in FIG.
Indicates the address of the memory 13, and ad14 and adn14 indicate the address of the memory 14, respectively. Each table is based on the structure shown in FIG. When recording, the disk controller 3 of FIG. 1 manages the free area of the disk 10 based on the TOC data in advance, and controls the recording of data in each sector of the free area.

First, in steps 101 to 104, the sequence table of the memory 12 is set. Memory 1
A free area of 2 and a free area of the memory 13 are searched for and designated as ad12 and ad13, respectively (step 101). Then, the table number of ad12 is set to ad13, and a
The sequence next to d12 is set to "end" (step 102). Next, the last data address of the data connected in the next sequence, which is the pointer of the sequence table of the memory 12, is set to adl12 (step 10).
3). Then, the sequence next to adl12 is set to ad12 (step 104). By the above operation, new 1-word data ad12 is added to the rear end of the sequence table.

Next, the sector table of the memory 13 is set. Since the first sector number of the continuous recording area on the medium (disk) is input to the memory unit 4, this input data is recorded at the start sector number of ad13 (step 105). Then, a free area of the key frame table of the memory 14 is searched and set as ad14 (step 106). Furthermore, the start sector address of ad13 is set to a
It is set to d14 (step 107).

Next, the key frame (I, P-picture)
Two sector numbers indicating the beginning and end of the encoded data of are input. Set these two sector numbers to K in ad14.
Write to the start sector number and K-end sector number (step 108). Then, it is judged whether or not the continuous recording area of the medium (disk) has ended (step 10
9) If not finished, a free area in the memory 14 is searched for and set to adn14 (step 110). And a
The frame next to d14 is set as adn14, and ad14 is updated to indicate adn14 (step 111).
Then, the process returns to step 108. In the loop of steps 108 to 111, the sector number indicating the encoded data of the key frame recorded in the continuous recording area on the medium (disk) is written in the key frame table.

If the continuous recording area of the medium (disk) is finished in step 109, the last sector number of the area which is the input sector number is set to ad13 of the sector table.
Is recorded in the end sector number of (step 112), and the frame next to ad14 in the key frame table is set to "end" (step 113). Step 105 to Step 11
In 3, the 1-word sector table data is set.

Next, in step 114, it is judged whether or not the input data is finished. If it is not finished, a free area in the sector table is searched and set to adn13 (step 11
5), the area next to ad13 is set as adn13, and ad1
3 is updated to indicate adn13 (step 11
6). Then, the process returns to step 105. Step 105
~ In the loop of step 116, a new sector table connected by a pointer from the previous sector table is created,
The sector number is recorded in a series of key frame tables indicated by the sector table.

If it is determined in step 114 that the data has ended, the area next to ad13 is set to "end" (step 117), and the operation ends.

Next, the operation during normal reproduction will be described.

At the time of normal reproduction, the start sector number and end sector number of the sector table corresponding to the video sequence to be reproduced are output from the memory section 4 to the disk controller 3.

FIG. 8 is a flowchart showing the operation of the memory section 4 during normal reproduction. Here, two pointers, that is, the current sequence address and the current sector address, which respectively indicate the addresses of the sequence table and the sector table corresponding to the data being reproduced, are used. Each table is based on the structure shown in FIG.

First, the current sequence address and the current sector address are set (steps 121 and 12).
2). Next, the values of the start sector number and end sector number of the current sector address are output to the disk controller 3 (steps 123 and 124). The disk controller 3 controls the disk 10 to read the data recorded in the continuous recording area by using the first sector number and the last sector number of the continuous recording area input from the memory unit 4.

Then, it is determined whether or not the area next to the current sector address is completed (step 125). If it is determined that the processing is not finished, the sector table corresponding to the current sequence address remains, so the pointer is advanced with the current sector address as the area next to the current sector address (step 126), and the process returns to step 123.

In the loop of steps 123 to 126, the first sector number and the last sector number of the area are sequentially output for each continuous recording area on a plurality of disks corresponding to one word data of the sequence table.

If it is determined in step 125 that the processing has ended,
It is determined whether the next sequence of the current sequence address is finished (step 127). If it is determined that the processing has not ended, there is still a video sequence recorded on the disc, so the pointer is advanced (step 128) with the current sequence address as the next sequence of the current sequence address, and the process returns to step 122. In the loop of steps 122 to 128, the first sector number and the last sector number are sequentially output for each continuous recording area until the video sequence recorded on the disc is completed.

FIG. 9 is a flowchart showing the operation of the memory section 4 in the case of high speed reproduction. At the time of high speed reproduction, the sector number in which the encoded data of the key frame on the disc recorded in the memory unit 4 is read out and output to the disc controller 3. Here, three pointers are used, which are the current sequence address, the current sector address, and the current frame address, which indicate the addresses of the sequence table, the sector table, and the key frame table corresponding to the data being reproduced.

First, the current sequence address, current sector address, and current frame address are set (steps 131-133). Next, the start sector number and end sector number of the current frame address are output to the disk controller 3 (step 134). The disk controller 3 controls the disk 10 to read the data recorded in the continuous recording area by using the first sector number and the last sector number of the continuous recording area input from the memory unit 4.

Then, it is judged whether or not the next frame of the current frame address is "end" (step 13).
5). If it is determined that the processing is not finished, the sector number to be accessed next is recorded in the next key frame table, so the current frame address is set as the next frame of the current frame address and the pointer is advanced (step 1
36) and returns to step 134. In the loop of steps 134 to 136, a plurality of sector numbers recorded in the key frame table corresponding to 1-word data of the sector table are output to the disk controller 3.

If it is determined in step 135 that the process has ended,
It is determined whether the area next to the current sector address is finished (step 137). If it is determined that it is not finished,
Since the sector table corresponding to the current sequence address remains, the pointer is advanced with the current sector address as the area next to the current sector address (step 13).
8) and returns to step 133. In the loop of steps 133 to 138, the sector numbers in which the encoded data of the key frame corresponding to the 1-word data of the sequence table are recorded are sequentially output.

If it is determined in step 137 that the processing has ended,
It is determined whether the sequence next to the current sequence address is "end" (step 139). If it is determined that the recording has not ended, there is still a video sequence recorded on the disc, so the pointer is advanced (step 140) with the current sequence address as the next sequence of the current sequence address, and the process returns to step 132. In the loop of steps 132 to 140, the sector numbers in which the coded data of the key frames are recorded are sequentially output until the video sequence recorded on the medium (disk) is completed.

In the example of FIG. 9, all the sector numbers continuously recorded in the key frame table are output.
That is, all key frames (I, P-pictures)
However, if the high-speed playback speed is high, some frames may be thinned out and output.

Here, since a decoded image of a past I-picture or P-picture is required to decode a P-picture, when a certain I-picture or P-picture in the GOP is thinned out, In the GOP, P-pictures after the thinned frame cannot be decoded. For example, in the case of the GOP structure as shown in FIG. 10, when the frame (c) is thinned out, the frame (d) cannot be decoded because the decoded image of the reference (c) cannot be obtained. Similarly, when the frame (b) is thinned out, the frames (c) and (d) cannot be decoded, and when the frame (a) is thinned out, the frames (b), (c) and (d) are decoded. It will not be possible.

As described above, since the number of frames that cannot be decoded differs depending on the thinned-out frame, it is necessary to calculate which frame is thinned out from the high-speed reproduction speed.

Here, when the GOP structure is unchanged, it is possible to know which key frame is the I-picture, but when the GOP structure changes in the middle of the video sequence, which key frame is the I-picture. I don't know. Therefore, for example, as shown in FIG. 11, a 1-bit flag indicating whether the data corresponds to the encoded data of the I-picture or the data corresponding to the encoded data of the P-picture for each word data of the key frame table. If added, it is possible to know whether the key frame is an I-picture or a P-picture, regardless of the GOP structure.

Alternatively, the memory section 4 may be constructed as shown in FIG. FIG. 12 is a modification example 2 described above.
The memory unit 4 is composed of four memories 12, 13, 15, 16 and one memory controller 11. In FIG. 12, the memory 1
2 and 13 are the same as the memories 12 and 13 shown in FIG.
The memory 15 is an I-picture key frame table, and the memory 16 is a P-picture key frame table.

FIG. 13 shows an example of data recorded in the I-picture key frame table and the P-picture key frame table. Here, the I-picture key frame table includes, in addition to the key frame table shown in FIG. 4, a pointer P-start address indicating the address of the P-picture key frame table for each 1-word data. The keyframe table corresponding to 1 GOP is the 1-word data of the keyframe table of the I-picture keyframe table and the P-picture keyframe table connected from the P-start address of the data in the P-next frame. It is word data.

In the moving picture recording / reproducing apparatus of the present invention, when changing or erasing the reproduction order of a video sequence or a part thereof, only the sequence table or sector table of the memory unit 4 needs to be changed. There is no need to change the recorded encoded data. For example, when erasing one video sequence, it is only necessary to remove one-word data in the sequence table corresponding to the video sequence to be erased from the pointers connected in the next sequence. In addition, in order to write new data in each table, it is necessary to manage unused addresses in each table.

FIG. 14 shows a memory section 4 corresponding to the third modification. In this mode, a 1-bit flag indicating whether it is used or not is added to each word data of each table.

The method of managing unused addresses is shown in FIG.
In addition to the mode of 4, the method of providing a table for managing free addresses for each table, or the method of finding an unused address by checking the used address by tracing the pointer of each table can be considered. Further, the flag indicating whether it is used or not can be applied to the tables of FIGS. 11 and 13 described above.

[Second Embodiment] The second embodiment is an embodiment of a second moving image recording / reproducing apparatus (claim 2) according to the present invention. FIG. 15 is a block diagram of the second moving image recording / reproducing apparatus. In FIG. 15, 1 is an encoder, 2 is a data processing unit, 4 is a memory unit, 5 is a selection circuit, 7 is a disk controller, 8 is a decoder, and 10 is a disk. Figure 15
The configuration is the same as that of FIG. 1 except for the following points.
That is, in FIG. 15, the decoder 8 inputs the flag indicating the end of the encoded data of the key frame to the disk controller 7. Further, in the case of FIG. 1, the first sector number and the last sector number in which the encoded data of the key frame is recorded are recorded in the memory unit 4, but in the case of FIG. 15, only the first sector number is recorded. It is recorded.

At the time of high-speed reproduction, in the embodiment of FIG. 1, the key frame (I,
The first sector number and the last sector number in which the encoded data of (P-picture) is recorded are input, and the disc controller 3 controls the disc 10 to read the sectors between the designated sector numbers. However, in the case of the embodiment of FIG. 15, only the first sector number in which the encoded data of the key frame is recorded is input from the memory unit 4.

In the embodiment of FIG. 15, the end of the coded data of the key frame is detected by the decoder 8 and a flag indicating the end of the coded data of the key frame is input to the disk controller 7. Upon receiving the flag input from the decoder 8, the disk controller 7 controls the disk 10 to access the sector number input next from the memory unit 4.

The second moving picture recording / reproducing apparatus shown in FIG. 15 has the decoder 8 having a function of detecting the end of the encoded data of the key frame. A function of detecting the end of the encoded data of the key frame may be added to.

The above embodiments are for accessing discrete frames, that is, discretely recorded data during high speed reproduction. Therefore, the present invention can also be applied to special reproduction involving random access as described below.

For example, the function of displaying a list of reduced images at regular intervals on a single screen and overviewing the contents of the disc requires random access to many images in a short time in order to create reduced images. Alternatively, for editing a scene of a moving image, the necessary cuts are connected by a logical pointer and continuously played back without copying the actual data recorded on the disc. Random access is made at the joint between the pointers. There is a need.

In this embodiment, the sector number on the disk is used as the information indicating the position on the disk, but the present invention is not limited to this. Depending on the disc, the expression may be different, and for example, it may be a combination of a track number and a sector number. In that case, an expression suitable for the disc may be used. The management data is T
Although it is written in the OC area, it may be an area other than the TOC area. Further, in this embodiment, the management data is I
-The sector number in which the picture or P-picture is recorded is used, but only the sector number in which the encoded data of the I-picture is recorded may be used. That is, I
-Only the picture may be managed.

[0109]

As described above, according to the present invention, in contrast to moving picture coding using intra-frame coding or inter-frame coding, management data is used during high-speed playback , for example,
The data group is read in units that correspond to the
Since a necessary frame is selected from the stored data, decoded and output , unnecessary access and unnecessary decoding are not performed. Therefore, a larger number of reproduced images can be output per unit time. Further, when random access is performed, desired data can be accurately accessed and useless data is not accessed. Therefore, it is also effective when performing special reproduction involving random access.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a moving image recording / reproducing apparatus according to a first embodiment of the present invention.

FIG. 2 is a diagram illustrating a position on a medium of a sector number input from a memory unit.

FIG. 3 is a block diagram of a memory unit according to a modification 1;

FIG. 4 is a diagram showing an example of TOC data recorded in a memory unit.

FIG. 5 is a diagram showing a relationship between TOC data recorded in a memory unit and a recording area on a medium.

FIG. 6 is another diagram showing the relationship between TOC data to be recorded in the memory section and the recording area on the medium.

FIG. 7 is a flowchart showing the operation of the memory unit when recording encoded data.

FIG. 8 is a flowchart showing the operation of the memory unit during normal reproduction.

FIG. 9 is a flowchart showing the operation of the memory unit during high-speed reproduction.

FIG. 10 is a diagram illustrating thinned frames and frames that cannot be decoded.

FIG. 11 is a diagram showing another example of data recorded in a key frame table in the memory unit.

FIG. 12 is a block diagram of a memory unit according to a modification 2;

FIG. 13 is a diagram showing data to be recorded in an I-picture key frame table and a P-picture key frame table according to Modification 2;

FIG. 14 is a block diagram of a memory unit according to a third modification.

FIG. 15 is a block diagram of a second moving image recording / reproducing apparatus according to the present invention.

FIG. 16 is a diagram illustrating an example of an interframe coding method.

[Explanation of symbols]

1 Encoder 2 Data processing unit 3 Disk controller 4 Memory unit 5 Selection circuit 6 Decoder 7 Disk controller 8 Decoder 10 Disk 11 Memory Controller 12 ... Memory (sequence table) 13 ... Memory (sector table) 14 ... Memory (keyframe table) 15 ... Memory (I-picture keyframe table) 16 ... Memory (P-picture keyframe) table)

─────────────────────────────────────────────────── --Continued from the front page (56) References JP-A-4-72976 (JP, A) JP-A-4-318375 (JP, A) JP-A-4-38679 (JP, A) JP-A-5- 30454 (JP, A) (58) Fields investigated (Int.Cl. ⁷ , DB name) H04N 5/76-5/95

Claims (6)

(57) [Claims] 1. A moving image reproducing apparatus for reproducing a moving image from a recording medium recording encoded moving image data, wherein a recording position of an intra-frame encoded frame or a forward predictive encoded frame. And means for writing management data recorded in a predetermined position of a recording medium to a memory, and recording a data group including intraframe coded data or forward predictive coded data using the management data on the memory. Means for reading from the medium, and indicating the start of a frame from the read data group
Detects headers and can be obtained unnecessarily before frame start
It is characterized by comprising means for removing data and selecting only intra-frame coded data or forward predictive coded data, means for decoding the selected data, and means for outputting the decoded data. Video playback device. 2. A frame header is added to the beginning of each frame of the encoded moving image data, and only intraframe encoded data or forward predictive encoded data is extracted from the read data group. When choosing
The moving picture reproducing apparatus according to claim 1, wherein the frame header is used. 3. The management data includes a first table for managing a video sequence, a second table for managing a recording area of a recording medium, and encoded data of an intra-frame encoded frame and a forward predictive encoded frame. Which is composed of a third table for managing the recording position of the video sequence, the concatenation information indicating the reproduction order of the video sequence, the concatenation information indicating the reproduction order of the recording areas recorded discontinuously, the intra-coded frame and the forward direction. 3. The moving image reproducing apparatus according to claim 1 or 2, wherein the moving image reproducing apparatus includes connection information indicating a connection of predictive coded frames. 4. The management data includes a first table for managing a video sequence, a second table for managing a recording area of a recording medium, and a first table for managing a recording position of encoded data of an intra-frame encoded frame. 3 table and a fourth table that manages the recording position of the encoded data of the forward predictive encoded frame, the connection information indicating the reproduction order of the video sequence, and the reproduction of the recording area recorded discontinuously. The moving image reproducing apparatus according to claim 1 or 2, further comprising: connection information indicating a sequence, connection information indicating a connection of intra-coded frames, and connection information indicating a connection of a forward prediction frame. . 5. The moving image reproducing apparatus according to claim 1, wherein a flag indicating whether the word is used or not used is added to each word of each table. 6. A moving image recording / reproducing apparatus for encoding moving image data, recording the encoded moving image data on a recording medium, and reproducing the moving image from the recording medium recording the encoded moving image data. Means, a means for encoding the input moving image data, a means for writing the recording position of the intra-coded frame or the forward predictive-coded frame in the memory as management data, and Means for recording data in the recording medium; means for recording the management data in the memory in the recording medium; and means for writing the management data in the memory from a recording medium recording encoded moving image data. , A data group including intraframe coded data or forward predictive coded data is recorded using the management data on the memory. Means for reading from the A, from the read out data group indicating the start of a frame
The header is detected and unnecessary data obtained before the start of the frame is detected.
Data removing means for selecting only intra-frame coded data or forward predictive coded data, means for decoding the selected data, and means for outputting the decoded data. A moving image recording / reproducing device.