JPH11176137A - Optical disk medium and its recording method and device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a recorder that is the maximum and true application of a rewritable large-capacity optical disk DVD-RAM by recording a thumbnail for indicating the data between partial reproduction sections and control information with pointer information to the thumbnail at each partial reproduction section of moving image data. SOLUTION: A recording/editing/reproducing control part 105 first takes out a thumbnail image. For taking out the thumbnail image, a thumbnail file is guided from the body name of an AV file that a chip belongs to. The recording/editing/reproducing control part 105 requests the reading of thumbnail data to an AV file system part 103 with the thumbnail file and the address and size of the thumbnail to be read as arguments. The AV file system part 103 reads an MPEG program stream where the thumbnail is stored via a disk- reading part 101 and transmits the read MPEG program stream to an AV data reproduction part 130.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a readable / writable optical disk and a DVD recorder which is a recording method, a recording device, an editing method, an editing device, a reproducing method, and a reproducing device. The present invention relates to an optical disc on which multimedia data is recorded and a recording method, a recording device, a reproducing method, and a DVD recorder as a reproducing device.

[0002]

2. Description of the Related Art In the field of rewritable optical disks, the upper limit of which is about 640 MB, a phase-change optical disk DVD-RAM having a capacity of several GB has emerged, and is expected not only as a computer application but also as a recording / playback medium in AV. ing. In other words, it is expected to spread as a medium that replaces the magnetic tape, which is a conventional typical AV recording medium.

(Simple Example of DVD-RAM) In recent years, the density of rewritable optical disks has been increased, and not only recording of computer data and audio data but also recording of moving image data has become possible.

For example, an uneven guide groove is conventionally formed on a signal recording surface of an optical disk.

Conventionally, signals were recorded only on the convex or concave portions. However, signals can be recorded on both convex and concave portions by the land / groove recording method. As a result, the recording density was improved about twice (see, for example, JP-A-8-7282).

Further, a zone CLV system which simplifies the control of the CLV system (constant linear velocity recording) effective for improving the recording density and facilitates its practical use has been devised and put into practical use (for example, Japanese Patent Laid-Open No. -93873).

[0007] How to record AV data including moving images using an optical disk aiming at large capacity,
It is a major task in the future to realize performance and new functions that greatly exceed the equipment.

With the advent of such a large-capacity and rewritable optical disk, it is conceivable that AV recording and reproduction will be mainly performed by an optical disk instead of a conventional magnetic tape. The transition of recording media from tape to disk has various effects on the functions and performance of AV equipment.

[0009] The greatest feature in the transition to a disk is a significant improvement in random access performance. If a tape is randomly accessed, it usually takes several minutes to rewind one roll. This is orders of magnitude slower than the seek time (several tens of ms or less) in optical disk media. Thus, tape cannot be a random access device in practice.

In a specific example, if a two-hour broadcast is once recorded on a two-hour tape, the remaining one and a half hours are erased except for half an hour, and the erased part is replaced with another one-half hour. Consider the case of using for recording. 3 left on the tape
The first half and the second half of the fragment, which are divided by the 0-minute video, remain. Theoretically, it is impossible to perform continuous recording for one and a half hours, even though there is a free area for one and a half hours in theory.

There are two reasons for this. First, even with a digital VTR, video data on a tape is not managed as a file. Therefore, it is not possible to identify where the empty area is and which part is the recorded area. Second, AV data cannot be continuously recorded or reproduced in a remote area other than the continuous area due to low random access performance.

As described above, the physical structure of the tape is limited to either the method of rewinding and recording continuously from the beginning or the method of appending to the last recorded part. .

[0013] When additional recording is performed on a pre-recorded tape, it is often experienced that rewinding to the beginning substantially occurs, erasing the recorded video, rewinding time takes a long time, and missing a recording opportunity. Should be.

The improvement in random access performance brought about by an optical disk is not limited to a mere cueing speed.
It produces the three functional features described below. 1) AV data can be handled as a computer file. 2) Non-linear editing of AV data becomes possible. 3) Variable bit rate AV data can be recorded and reproduced in real time.

Hereinafter, these features will be described with reference to the prior art. (Prior Art 1: Recording AV Data as Computer File) Prior Art 1 with an example of a multimedia PC equipped with a phase-change optical disk DVD-RAM
Will be described.

When an optical disk is connected to a computer, writing and reading on the disk are performed through a recording medium management program called a file system in the OS. Since data mainly handled by a computer is character or code data, the capacity of one file is, for example, about 100 kb, which is smaller than the recording capacity (several GB) of an optical disc. For this reason, the file system is designed so that even if a large number of small files are repeatedly recorded and erased, no problem occurs.

More specifically, the entire disk is divided into small data blocks of several tens of KB, and if a file cannot be accommodated in one data block, the remaining data is recorded in another data block which is not necessarily in a continuous area. . Finally, the link information of the data block is recorded as file management information, and the writing of the file is completed. A data block in which nothing is recorded is managed as a free area, and is used for recording data as needed. When a file is deleted, a used data block is registered in a free area.

Consider a case where AV data is recorded / reproduced on / from an optical disk with such a conventional file system structure.

First, it is assumed that the format of AV data to be recorded on an optical disc is specified by MPEG2 (see ISO / IEC13318) and that the bit rate is 8 Mbps. That is, 1
It is necessary to record or read 8 megabits (= 1 megabyte) of data per second. Normally, a conventional multimedia PC uses AV data having a bit rate of about 1.5 Mbps called MPEG1, but in this example, description will be made based on MPEG2, which is standardly used in DVD.

Here, the data transfer speed and random access performance of the optical disk drive will be briefly described.

First, the transfer speed of a DVD-RAM is about 11 Mbps at the maximum. This is a value at the time of recording, and at the time of reading, it is possible to improve the reading speed at double speed or higher.
Random access performance varies greatly depending on the design of the disk drive mechanism and servo system. Normally, the time required to move between data blocks is 30 per cent for PC peripherals.
Although it is about 0 ms, a consumer player or the like requires a maximum time of about 1.5 seconds. In particular, it takes time to move from the innermost circumference to the outermost circumference. It is assumed that after recording on a DVD-RAM with a PC, playback is performed on a consumer device.
Suppose 1.5 seconds is needed.

Therefore, assuming that a data block, which is a continuous recording unit of the file system, is 32 KB, and that movement between data blocks occurs every 32 KB, it takes 31 times to record or read 1 MB which is one second worth of data. In the worst case, it takes 45 seconds or more. This is AV
There is no guarantee that data can be recorded and played back without interruption.

When recording AV data on a disk on which nothing is recorded, continuous data blocks can be allocated, so that continuous recording / reproduction is possible.
However, if various recording and (partial) erasures are repeated many times, data blocks will not be continuously allocated, and the same problem will occur.

As a solution, there is a method called garbage collection. This is a technique for changing the arrangement of data that becomes an obstacle when data blocks cannot be arranged continuously, and securing a continuous area. However, since the transfer speed of an optical disk is relatively low for its capacity, garbage collection requires several hours, which is not practical.

(Prior Art 2: Editing AV Data) A second feature obtained by handling AV data with a disc is that AV data can be nonlinearly edited. This can be regarded as an improvement in data processing performance such as freely rewriting necessary portions in AV data, erasing only unnecessary portions, and connecting arbitrary sections of a plurality of AV data.

First, editing using a tape will be briefly described. Consider cut editing of a recorded video as a typical process of AV data. Cut editing on a tape is a work of combining a video cut 1 and a video cut 2 to create a final video and recording it on a tape. When this operation is performed on a tape deck, at least two decks and at least two tapes (material and editing result) are required.

First, the tape on which the video cut 1 is recorded is set on the deck 1, and the tape on which the editing result is recorded is set on the deck 2. After cueing to the head of the video cut 1 on the deck 1, reproduction of the deck 1 and recording of the deck 2 are started simultaneously. Stop the two decks at the end of Cut 1. Subsequently, the head of the video cut 2 on the tape of the deck 1 is started, and the reproduction of the deck 1 and the recording of the deck 2 are simultaneously started again. When the recording is completed up to the end of the cut 2, the tape on the deck 2 is rewound and the editing operation is completed.

Editing a tape medium is a time-consuming and time-consuming operation, and requires two decks, so that ordinary consumers rarely edit except for business or professional use. .

Next, consider a case where editing is performed using a disc. Editing on tape was thought to copy the video. On the other hand, when editing on a disc, cut 1
The editing work is completed only by specifying the connectivity between and Cut2.

That is, the editing procedure is as follows. Here, it is assumed that the video cut 1 and the video cut 2 are recorded on the same disk as in the above example. 1) Specify the end of video cut 1 and the head of video cut 2 to instruct connection. 2) Eliminate unnecessary parts other than video cuts 1 and 2 (if necessary).

The cut 2 is successively read after the video cut 1 during reproduction, so that the edited result is output.

The operations 1) and 2) are completed within a very short time because it is not necessary to read a large amount of video data. However, there has not been a conventional one that performs such editing using an optical disc.

The biggest problem when using an optical disk in the above editing is that the random access performance of the optical disk is low. Performance that is orders of magnitude faster than tape is also insufficient for editing operations. For this reason, hard disks are used in non-linear editing machines for business use. Since the hard disk has a random access speed of several ms and a transfer speed of 30 Mbps or more, it is easy to record and reproduce AV data of about 8 Mbps in real time.

However, the biggest problem is that the hard disk is a fixed device and cannot be used as an input medium or a final output medium for editing. Therefore, the output of the conventional non-linear editing machine is a video tape to the last, and in the case of recording on an optical disc in the future, two discs are required: a hard disk for editing work built in the non-linear editing machine and an optical disc for writing the editing result.

(Prior art 3: Recording / reproducing variable bit rate AV data in real time) Further, as a third merit of a disk obtained by improving random access performance, variable bit rate AV data is recorded / reproduced.
It becomes playable. This can be said to be the difference between a tape mechanism that cannot spin and a disk mechanism that can spin, rather than improving random access performance.

The DVD uses a technique called intermittent transfer for reading data. In other words, while a CD continuously reads data at a constant speed, data is read at a transfer rate higher than the speed at which data is consumed by a reproduction system such as a decoder, and a margin generated by a difference between the read speed and the consumption speed. Is buffered by a semiconductor memory. When the buffer becomes full, the reading is interrupted and the idle operation called kickback is continued. When the buffer becomes empty, reading is performed again.

According to this method, a retry can be performed even if an error occurs in reading the disk due to some reason such as external vibration. Also important is the recorded AV
The point is that the data bit rate can be changed.

MPEG2 basically aims at compression / transfer / recording at a variable bit rate. In particular,
It analyzes the video signal input at the time of encoding and calculates the complexity of the image. More bits are assigned to complex scenes, and less bits are assigned to simple scenes. A bit amount several times larger is assigned to a scene with a lot of motion or a scene with many high frequency components than a stationary scene or a scene mainly composed of low frequency components.

With this technology, a fixed bit rate MPEG
The compression ratio can be increased up to twice as much as that of the stream, and long-time recording can be performed.

However, when AV data is recorded on a disk at a variable bit rate, there is a problem that it is not possible to know where to jump in at random access.

For this reason, in the DVD-ROM, the playback time of the internal video data in one or a plurality of GOPs is set at the beginning of a continuous section of 0.4 seconds or more and 1.0 seconds or less (exceptionally 1.2 seconds at the end of VOB). , NV pack, which is a pack containing DVD-specific information,
Information that refers to the pack and the size of data that must be read to display the first reference picture are recorded.

Note that the AV data from the NV pack to before the next NV pack is called a VOBU (abbreviation of Video Object Unit). The VOBU divides AV data continuously and without gaps.

As information for referencing a neighboring NV pack recorded in the NV pack, the addresses of the NV packs of the preceding and succeeding VOBUs located at a place separated by a certain time interval with reference to the time code at the beginning of the VOBU. Is indicated by a relative value from the address of the own NV pack. The time interval is from 1 second to 15 seconds every other second, 20 seconds, 60 seconds, 120 seconds, and 240 seconds.

Next, an example of the operation of special reproduction such as fast forward reproduction and rewind reproduction will be described. By reproducing only the reference picture of the VOBU at a certain time interval according to the reproduction speed, it is possible to realize the special reproduction at a substantially uniform speed. In order to read VOBUs one after another at regular time intervals, information indicating the address of a neighboring NV pack where the NV pack is located is used.

For each time code at a fixed time interval from the beginning of the AV data, time search map information indicating an address in the AV data of the VOBU corresponding to the time code is recorded. By referring to the time map information, the reproduction of the AV data can be started from the designated time code.

(Prior Art 4: Random Access) The features obtained by handling AV data on a disc are as follows.
That is, random access to AV data can be easily performed. However, how the contents of the random access target (video or audio) can be displayed to the user is closely related to the ease of use of the AV equipment and the AV recording format. In particular, this problem becomes more serious as the amount of data to be recorded increases.

For example, in the case of a digital still camera (still image camera by JPEG compression), which has become increasingly popular recently, a user can select and reproduce a large number of still images easily.
A method of displaying a menu using thumbnails is often used. In the case of a digital still camera, one still image data is recorded as one file, and at the same time, a thumbnail image is also recorded as one file on a recording medium (a memory card, a floppy disk, or the like).
At this time, a method of recognizing the relationship between the main image and the reduced image (thumbnail image) by providing a rule (for example, attaching the same prefix or suffix) to the file name is often used. In the case of a digital still camera, since only one image is handled, it is sufficient to create and record a thumbnail image after creating and recording the main image.

[0048]

SUMMARY OF THE INVENTION The present invention solves the following problems which are hindered in maximizing the random access performance of a disk described in the above-mentioned prior art, and provides a rewritable large-capacity optical disk DVD-RAM. It realizes a DVD recorder that is the largest and favorite application.

The biggest problem in performing random access to AV data on a DVD-RAM is that the AV data includes not only still image data but also moving image data.

For example, when several minutes of moving image data are recorded in one file, it is rare that the random access target desired by the user is one place. In particular, it is shot in small pieces like a camcorder (generally said to be about several tens of seconds)
This is especially true for devices that perform In a DVD recorder that handles moving image data in this way, a mechanism that can have a plurality of thumbnail images for one moving image (file), and a plurality of thumbnail images for a plurality of thumbnails are automatically generated during a series of operations during shooting. I needed a function to create images.

[0051]

According to a first aspect of the present invention, there is provided a rewritable optical disk medium on which at least moving image data and management information of the moving image data are recorded. An optical disc medium characterized by recording, for each partial playback section of recorded moving image data, a thumbnail indicating the content of the partial playback section and management information having pointer information to the thumbnail.

The invention according to claim 2 is the optical disc medium according to claim 1, wherein the data capacity of the thumbnail is a fixed value.

According to a third aspect of the present invention, in the optical disk medium according to the second aspect, the data capacity of the thumbnail is an integral multiple of an ECC block which is a unit of reading and writing to the optical disk medium. I have.

According to a fourth aspect of the present invention, there is provided a recording apparatus for recording moving image data and management information on the optical disk medium according to any one of the first to third aspects, wherein the still image related to the moving image data to be recorded is recorded. The optical disk recording apparatus has a frame memory for temporarily storing image data.

According to a fifth aspect of the present invention, there is provided a recording apparatus for recording moving image data and management information on the optical disk medium according to any one of the first to third aspects, wherein the content of a partial reproduction section of the moving image data is provided. An optical disk medium recording apparatus is characterized by having a still image library storing a plurality of still image data for thumbnails indicating.

According to a sixth aspect of the present invention, there is provided a recording method for recording moving image data and management information on an optical disk medium according to any one of the first to third aspects, wherein a moving image recorded on the optical disk medium is recorded. Extracting one frame of still image data from the image data as still image data, converting the still image data to thumbnail data, recording the thumbnail data on the optical disc medium, and converting the thumbnail data on the optical disc medium to the thumbnail data. An optical disk medium recording method is characterized by comprising a step of creating pointer information and a step of recording the pointer information.

According to a seventh aspect of the present invention, there is provided a recording method for recording moving image data and management information on the optical disk medium according to any one of the first to third aspects. When there is already a partial playback section that overlaps with the partial playback section, pointer information to thumbnail data of the overlapping partial playback section is read from the management information of the overlapping partial playback section. And recording the pointer information as pointer information to a thumbnail of the partial playback section.

[0058]

(Embodiment 1) Details of the present invention will be described using a DVD recorder which is an embodiment of the present invention.

1. System Configuration and Function of Each Unit The system configuration of the DVD recorder will be described with reference to FIG.

(Disc Recording Unit) Disc Recording Unit 100
Records a logical sector number and logical data (2048 bytes) in units of one or more sectors on a disk. Data recorded on the disk is processed in units of ECC blocks, which are blocks of 16 sectors.
When recording in some sectors of the block, the following processing occurs.

First, the ECC including the sector to be recorded is
Read the block once into the buffer memory. The data of the sector to be recorded is copied to the corresponding area of the buffer, and the data of the buffer is recorded in the ECC block.

When recording a large amount of continuous data such as AV data, recording is performed in ECC block units in order to remove the overhead of this processing.

(Disk Read Unit) When the disk read unit 101 receives a logical sector number and the number of sectors to be read in the same manner as the disk recording unit, the disk read unit 101 reads data in units of the ECC block, and after the ECC processing, only necessary sector data is stored in a file. Transferred to the system section. As in the case of the disk recording unit, when reading AV data, the overhead is reduced by performing reading in units of 16 sectors for each ECC block.

A DVD-RAM drive, which is a peripheral device for a PC, realizes functions of a disk recording unit and a disk reading unit, and is connected to a PC main body by a connector of an interface called SCSI or IDE. DVD based PC
Although the recorder may have slightly different specifications from the description, the description is omitted because it is not related to the gist of the present invention.

(File System Unit) The file system unit 102 is an AV file system unit 10 that handles AV data.
3 and a non-AV file system unit 104 that handles non-AV files such as control information. FIG. 1 shows a list of commands in the file system unit.

"CREATE" creates a new file on the disk and returns a file descriptor.

"DELETE" deletes a file existing on the disk. "OPEN" obtains a file descriptor for a file recorded on the disk in order to access the file.

"CLOSE" closes an open file. "WRITE" records a non-AV file on the disc.

[AV-WRITE] records an AV file on a disk. “READ” reads a file recorded on a disk.

“SEEK” moves in a data stream recorded on a disc. "RENAME" changes the name of the file recorded on the disc.

"MKDIR" creates a new directory on the disc. "RMDIR" deletes a directory existing on the disk.

"STATFS" inquires about the current status of the file system. "GET_ATTR" queries the attributes of the currently open file.

"SET_ATTR" changes the attribute of the currently open file. “IN_AV_BLK_BOUND” checks whether there is an AV block boundary in a designated section of the AV file.

[MERGE] merges two AV files on a disc with data in a memory.

[SPLIT] divides an AV file on a disk into two AV files. "SHORTEN" deletes the end of the AV file on the disc and deletes the unnecessary part of the AV file.

"REPLACE" replaces a part of the AV file with the data in the memory. The point to focus on here is A
The difference is that a command different from that used when recording non-AV data is used when recording V data. For details on how the recording state on the disc changes as a result of exercising these functions, see the operation example of the AV file system described later.

(User IF Unit) The user IF unit 106
In accordance with an instruction from the recording / editing / playback control unit 105 which controls the entire DVD recorder, a graphic display is displayed on the screen to prompt a user operation, display the progress of processing, and record the result of the user's remote control operation.・ Inform the editing / playback control unit.

(Recording / Editing / Playback Control Unit)
The reproduction control unit 105 is a part that controls the entire DVD recorder. The recording / editing / playback control unit 105 performs new recording and playback / editing of the recorded AV data according to a user operation.
0, request processing to the AV data editing unit 120. At this time, each processing unit independently requests the AV file system unit 103 to record and read AV data.
The control data management unit 107 reads the V data into the main memory in advance, and can provide information immediately in response to a request from each processing unit.

(AV Data Recording Unit) AV Data Recording Unit 1
Reference numeral 10 includes an AV data input unit, an AV file management information generation unit, an AV clip management information generation unit, and an overflow countermeasure unit. The AV data recording unit 110 performs recording and
A recording request is received from the editing / playback control unit 105. AV
The data input unit 111 converts input video and audio signals into MPEG data, that is, performs encoding in real time. Next, in order to write the encoded MPEG data to a disc as an AV file, A
The file is passed to the V file system unit 103. Also, the AV data input unit 111 determines, for each GOP, which is the minimum unit that can be encoded and decoded MPEG video data, the number of sectors required to record the data of the GOP on the disc and the number of sectors of the reference picture. Calculate and calculate the obtained value as A
It is passed to the V file management information generation unit.

The AV file management information generation unit 112
The number of sectors for each P and the number of sectors of the reference picture are stored in the main memory until the recording is completed. At the end of recording, AV file management information is generated from the stored GOP information and passed to the control data management unit 107.

The AV clip management information generating section 114 attaches a StartMark to the beginning of the AV data at the start of recording, and attaches an EndMark to the end of the AV data at the end of recording. The two marks are used as one clip as shown in FIG.
The clip management information is created and written as an AV clip management information file. Further, a new clip sequence composed of one clip is added to a file for managing the clip sequence. Clips and clip sequences will be described later in detail.

Next, the processing when a pause instruction is received from the recording / editing / reproduction control unit will be described with reference to FIG.
The AV data input unit shown in FIG. 25 includes a video encoder, a video buffer for storing the output of the video encoder, an audio encoder, an audio buffer for storing the output of the audio encoder, and a system for multiplexing video data and audio data. It is composed of an encoder, an STC (system time clock) that is a synchronous clock of the encoder, and an encoder control unit that performs overall control and management. The encoder control unit manages encoding particularly in the video encoder. Specifically, it manages GOP information and picture information of the encoded data, and passes necessary information to the AV file management information generation unit.

The pause command from the recording / editing / playback control unit is sent to the encoder control unit in FIG. The encoder controller first issues a pause command to the video encoder and the audio encoder. The video encoder and the audio encoder that have received the pause instruction perform encoding up to the frame (including the frame being input) buffered inside the encoder, and stop encoding. The video encoder notifies the encoding control unit of the suspension of the encoding at the same time as the suspension of the encoding. The encoding control unit that has received the suspension of the encoding from the video encoder sends a pause command to the system encoder and the STC (system time clock) to suspend the system encoder and the STC (system time clock). Also, the encoder control unit sends the time code (relative value from the start of encoding) of the paused frame to the AV clip management information generation unit. Thereafter, the video encoder, the audio encoder, the system encoder, and the STC (system time clock) resume encoding by receiving a pause release instruction from the recording / editing / playback control unit.

The AV clip management information generation unit records the time code received from the AV data input unit as a single mark in the AV clip management information during the pause.
Note that the clip may be divided at the pause time.

The overflow countermeasure unit 113 is adapted to output the MPEG data generated by the AV data input unit when some trouble occurs in the recording speed of the disk and the data cannot be transferred to the file system unit at the bit rate assumed in advance.
The AV data input unit is instructed to reduce the bit rate so that the data does not overflow into the buffer memory provided therein, and only the B-picture data is deleted from the already generated MPEG data. Perform the process of forcibly reducing the rate.

(AV Data Editing Unit) AV Data Editing Unit 12
0 includes an AV clip sequence editing unit 121, an AV clip editing unit 122, and an AV file management information editing unit 123.

Here, an AV clip and an AV clip sequence will be described. An AV clip is a partial section of AV data specified by an IN point (start point) and an OUT point (end point) set in cut editing of AV data. FIG.
6 shows an example of an AV clip and an AV clip sequence. The upper row is
AV file #j composed of MPEG data and AV file
An example of three AV clips specified in #k is shown.
For Clip # 1, Mark # 1 indicates the IN point and Mark # 2 indicates the OUT point. Mark # 1 and Mark # 2 are both indicated by time codes. In the clips # 2 and # 3, the IN point and the OUT point are similarly indicated by Mark.
This time code sets the beginning of the AV file at time 00: 00: 0
It is calculated as 0:00.

The lower part of FIG. 26 shows the three AVs shown in the upper part.
The clips are arranged in a sequence as a sequence.
First, clip # 1 of AV file #j, then clip # 1 of AV file #k, and finally clip # 2 of AV file #j
Indicate an AV data sequence to be arranged in order and reproduced as an editing result.

It should be noted here that the AV clip sequence editing section 121 merely specifies a mark, and does not perform any processing on the AV data itself. The clip string information is recorded on the disc and can be reproduced in the order specified by the AV data reproducing unit 130.

The AV clip editing unit 122 converts the AV clip designated by the AV clip sequence editing unit 121 into A
The AV data is cut out from the V file, and the AV data itself is processed so that connection can be made in the order of the specified AV clip sequence.

Here, the relationship between AV data and AV clips will be described with reference to FIG. First, since AV data is composed of MPEG, the data is a GOP (GOP
roup of Pictures). # 1 in the figure
Each VOBU from to # 8 indicates a GOP. VOBU is Video Object
This is an abbreviation of Unit, and usually consists of one GOP. On the other hand, as described above, the marks indicating the IN point and the OUT point of the AV clip are specified by the time code and specified completely independently of the boundary of the GOP (VOBU).

In FIG. 27, the hatched portions indicate the AV clip and A
It roughly shows the relationship with the corresponding video data in the V data. However, audio data is multiplexed out of video data. When video data is extracted from such MPEG data,
Audio data that is not synchronized with the video data (the playback time is earlier than the video) is simultaneously extracted. Just because an AV clip is specified in this way
Due to the nature of MPEG, it is difficult to simply extract AV data. For this reason, the extraction involves GO data related to MPEG data.
P (VOBU) needs to be reconfigured. The AV clip editing unit reconstructs such MPEG data. Details will be described in an operation example.

The AV file management information editing unit 123
The relative address information in the special reproduction file generated by the file management information generation unit 112 is corrected according to the deformation of the AV data accompanying the editing of the AV clip.

(AV Data Reproducing Unit) The AV data reproducing unit includes an AV clip string reproducing unit 132, an AV file reading unit 133, a read error countermeasure unit 134, and an AV decoder unit 131.

[0095] The AV clip sequence playback unit 132 controls the playback of the clip sequence passed from the recording / editing / playback control unit 105. It requests the AV file reading unit 133 to read the data of each clip constituting the clip sequence. It requests the AV decoder unit 131 to decode the read data.

The AV file reading section 133 calculates the address (file offset) of each clip constituting the clip sequence in the AV file, and requests the file system section 102 to read data from the AV file. I do. In the case of fast-forward playback or fast-reverse playback, a request is made to the file system unit 102 to search for a minimum necessary portion for playback and to read only necessary data.

The read error countermeasure unit 134 performs appropriate recovery and avoidance measures when an error cannot be corrected even after the ECC processing has been performed on the AV data recorded on the disk. More specifically, it instructs the next GOP (VOBU) instead, or requests the audio decoder of the AV decoder unit 131 to mute the audio.

[0098] The AV decoder 131 decodes the AV data read from the file system 102. FIG. 28 shows a decoder model. In FIG. 28, reference numeral 150 denotes input AV data; 151, a DeMUX (demultiplexer) for transferring data to a decoder based on an SCR and a stream ID described in a pack header in the AV data; 152, a video decoder; A decoder 154 is a seamless connection processing unit that performs processing at a data boundary generated by editing existing in the AV data, and 155 is a video signal output from a video decoder.
Reference numeral 156 denotes an audio signal which is an output of the audio decoder. This configuration is normal except for the seamless connection processing unit.
Same as MPEG decoder.

2. Disc Format Next, a format to be recorded on a disc will be described.

FIG. 2 shows the entire structure of the disk. In the figure, the horizontal axis indicates a physical sector address. At the head of the physical sector address, there is a lead-in area, in which a reference signal necessary for stabilizing the servo, an identification signal with other media, and the like are recorded. Following the lead-in area is a data area. Logically valid data is recorded in this part. Finally, there is a lead-out area where the same reference signal as the lead-in area is recorded.

Hereinafter, the data area which is the object of the present invention will be described in detail. (Data Area) First, the data area is divided into blocks of the minimum accessible size called sectors, and the use status is assigned to each sector and managed using the information management area. The form of the assignment management information area may be a list structure or a table. Here, description will be made using an example of management using a table called a sector bit map. FIG. 10 shows this state.

The size of one sector is 2 KB, and READ and WRITE to the disk are permitted only in units of an integral multiple of the sector size. Further, the data area is divided into a plurality of zones. In this example, the specific number of zones is 24. The meaning of zone introduction is to facilitate control during recording by rotating the disk at CAV (constant angular velocity, that is, constant rotation speed) in the zone. The same number of tracks is assigned to each zone. In this example, it is 1888 per zone. At the boundary of the zone, two zones (corresponding to 48 to 80 sectors) of buffer sectors are provided in each zone, and data cannot be recorded in this area. Further, the head of each zone is the head sector of the ECC block, thereby improving the accessibility.

As described above, there is a sector that cannot be recorded at the zone boundary, but it is inconvenient in use. Therefore, the logical sector number is calculated from the physical sector address using a table, so that the logical sector address becomes the data recordable area. It is only considered to show only continuous. Therefore, hereinafter, the discussion will proceed while ignoring the area existing at the boundary of the zone. However, when data is recorded or read continuously across the zone boundary, a delay of about several hundred ms is assumed.

3. AV File System Next, a file system that is a method of using a data area will be described.

The purpose of the file system is to enable an application to access a file as a unit.
In other words, the application can be realized without worrying about other files. This makes it possible to easily realize an application even if many files exist on the same disk. Furthermore, the data area can be effectively used even if the file is repeatedly added or deleted. This is because when recording logical data, which is the contents of a file, on a disk, it is divided into small data blocks and the link information between the blocks is managed together. At the beginning of the data area, volume information and file information for managing data through the file system are recorded. Following these management information, actual file data is recorded.

There are two types of file data, AV data and non-AV data, and the recording method on the disc is also different. This is to ensure the real-time property of recording and reading of AV data, and to maintain the use efficiency of the disk when a large number of files of non-AV data (usually having a small capacity) are recorded.

FIG. 3 shows details of the data area. In order to guarantee real-time performance, it is necessary to accurately estimate the overhead time generated when recording and reading data. First, the data area is divided into 24 zones as described above, and a delay of several hundred ms occurs when continuous recording / reading is performed across the boundary. One zone is divided into fixed-length logical blocks (hereinafter, referred to as AV blocks). However, the last AV block in the zone has a larger size than other AV blocks. Each AV block is EC
The ECC block is constituted by an integer multiple of the C block, and the ECC block is constituted by 16 sectors. Each sector can record 2048B of data. FIG. 3 shows an example in which AV data is recorded in each sector. A method for determining the size of the AV block will be described later.

(AV Block) Each AV block may record AV data or non-AV data. However, recording of AV data and non-AV data mixedly in the same AV block is not allowed. The AV block is also managed using the allocation management information area, similarly to the sector. The form of the allocation management information area for the AV block may be a list structure or a table, as in the case of the sector.

The management information area can be provided in a file system area like a volume management area of a disk, or can be processed by an application.
It can also be stored on disk as a single file.

Here, the AV structure shown in FIG.
A description will be given using a block management table as an example. This management table is recorded as a part of the volume information at the head of the data area.

The management table is used to identify whether the AV block corresponding to the AV block address is assigned data for AV, is assigned for non-AV, or is unused. As shown in the figure, the two bits of the table, 00, are unused, 01 is AV, 1
0 indicates non-AV. The length of each AV block is 224 in this example.
ECC block (approximately 7 MB), the final AV of each zone
Only the block has a different value for each zone shown in the table. This is to maintain the disc usage efficiency while avoiding the AV block from straddling the zone boundary.

Here, how the length of the AV block is determined will be described. 5 and 6 show buffering models at the time of recording and reading AV data. In FIG. 5, the MPEG data output from the encoder is a FIFO (First In First Out) called a track buffer.
After being temporarily stored in the memory, it is recorded on the disk. When recording data on a disc, by recording data in units of continuous ECC blocks as much as possible, useless rotation wait and seek time can be avoided. The AV block is a unit for continuously performing recording and reading at a time. FIG. 5 is a graph showing the occupation state of the track buffer at this time.
And FIG.

In FIG. 5, the maximum input rate to the track buffer is Vin, and the maximum allowable rate for disc recording is Vou.
Let t. Here, Vin <Vout. In this example, Vin = 8 Mbps and Vout = 11 Mbps are set as actual values. It is assumed that the track buffer has stored the AV data up to a certain value equal to or less than the buffer capacity as an initial value. This is because if recording is started without accumulation, the track buffer immediately underflows due to Vout> Vin, and continuous recording cannot be performed on the disc at the maximum allowable rate. In FIG. 5, if the encoding is continued while recording the AV block #j, the occupation amount of the buffer decreases at the rate of Vin-Vout during the writing time to the disk. When the recording of block #j is completed, the AV block #k which is the next AV block
Jump (seek and wait for rotation). During this time, the buffer occupancy increases with the encoding rate Vin. The condition to be satisfied here is to prevent the track buffer from overflowing and the AV data from being lost. This condition is: Vin x maximum jump time <track buffer capacity from the maximum value of the jump time and Vin regardless of the length of the AV block.

Now, Vin = 8 Mbps, maximum jump time = 1.5
Since it is seconds, the capacity of the track buffer is 1.5MB. Of course, if a disk recording device having a short maximum jump time during recording is used, the buffer capacity can be greatly reduced.

Next, a buffering model at the time of reading will be described with reference to FIG. In the figure, the data read from the disk goes through ECC processing to the track buffer.
Entered at Vin rate. The track buffer is the same FIFO as used for recording. As a worst case, reading of the AV block #j starts from a state where no data is accumulated in the track buffer. In this case, during reading, Vi
Data accumulates in the buffer at the rate of n-Vout.
Assuming that the length of the AV block is 224 ECC blocks (= 7.2 MB), Vin = 11 Mbps, and Vout = 8 Mbps, the read time of the AV block is about 5.2 seconds, and a track buffer of about 2 MB is required to prevent the buffer from overflowing. Become. However, even if the track buffer overflows, the data recorded on the disk is not lost, so there is no need to have a 2 MB buffer. A necessary condition is to secure a buffer capacity that does not cause an underflow in the track buffer during the jump after the reading of the AV block #j is completed.
This means that if the maximum jump time is 1.5 seconds and Vout = 8Mbps
1.5MB. In the case of a read device with a short maximum jump time, the required buffer capacity can be reduced.

Here, the lower limit value of the AV block length will be described. In the above buffering model at the time of reading, when all AV blocks are read at a time, a 2 MB track buffer is required, but a 1.5 MB track buffer is sufficient to prevent data from actually underflowing when jumping. . In other words, since the AV block length was set appropriately, 2 MB> 1.5 MB
Discussions can now proceed without being aware of the block length. Conversely, the track buffer clearly underflows when only the amount less than 1.5 MB can be accumulated in the track buffer even when the data is read all at once. Therefore, when data stored in the track buffer while reading data in the AV block is read at the Vin bit rate, it is sufficient that a time of 1.5 seconds or more is required. x (Vin -Vout) / (Vin x Vi
n)> 1.5 seconds In other words, the AV block length must be at least 5.5 MB or more.
The reason for setting the size to about 7 MB is to allow for a margin when a disk error occurs.

When recording an AV file using this AV block, if the other parts except the head and tail of the AV file are recorded by writing data completely in the AV block, the real-time Recording and reproduction can be guaranteed. This is because, as described above, when the AV data which is continuously recorded in one AV block is read out, data which can be reproduced for 1.5 seconds or more by decoding at a bit rate of 8 Mbps can be accumulated in the track buffer. Since 1.5 seconds is the worst-case jump time for consumer use, it is possible to continue recording / reproduction without any problem even if any jump occurs during data reading. In the playback, if the continuous recording length at the beginning of the AV file is shorter than the AV block length, it seems to be a problem, but if decoding is not started until the track buffer is full of data, jump is guaranteed from the initial stage of AV file playback And the problem of interruption of reproduction does not occur. By using the AV block and the buffering model,
In addition, it is possible to maintain the efficiency of use of the disk while guaranteeing real-time recording / reproduction.

As described above, in order to make the buffer full, 7.2 MB of continuous data is not required. It is possible to fill the buffer if 5.5MB of data is written continuously. For this reason, an AV data recording method that creates an unused area in an AV block if 5.5 MB of data is continuously written is conceivable, but from the standpoint of recording more data on a disc, the use of a disc is considered. This method of reducing efficiency is a recording method that has no advantage. Also, the AV file system must manage whether 5.5 MB of data can be continuously recorded in the AV block, and the processing of the AV file system becomes complicated.

(Hierarchization of AV Blocks and Sector Bitmaps) The above description has focused on only the AV blocks. However, the property of managing the allocation status of the data area is based on the fact that the AV block management table is also used in the sector bitmaps. When allocating a data area, it is necessary to operate the AV block management table and the sector bitmap in synchronization. In the AV block management table, the AV block is unused (00), for the AV file (01),
It manages which state is for non-AV files (10).

The sector bit map manages the state of the allocated (1) and unallocated (0) sectors.

Here, the relationship between the AV block management table and the sector bit map will be described. However, the process for the AV block management table and the sector bitmap accompanying the actual operation (READ, WRITE, etc.) will be described later when describing the file system command.

When an AV block is allocated for an AV file, all the sectors contained therein are registered as allocated on the sector bitmap. In FIG. 11, A
V block AV_BLK # 4 is an example. FIG. 12 (a)
FIG. 9 shows the change of the sector bit map in that case. Even if there is a sector in which data is not actually written, all sectors are already allocated. By doing so, even if this disk is accessed by a file system that supports only the sector bitmap and does not support the AV block management table, data is written in a file system that supports both the sector bitmap and the AV block management table. AV files can be protected.

An AV block allocated for an AV file is registered as an unused AV block in the AV block management table when all recorded AV data is deleted. Specifically, the value of 01 in the AV block management table is changed to 00. A changed to unused at the same time
The status of the sector included in the V block is also changed, and is changed to an unallocated sector on the sector bitmap. FIG.
2 (b) shows the processing contents at that time. Since it is not possible to determine from the sector bitmap whether all the AV data in the AV block has been deleted, the file Exte
The AV file system determines based on the nt information.

When an AV block is allocated for a non-AV file, 10 is registered in the AV block management table.
AV_BLK # 1 in FIG. 11 is an example. AV block non-A
In the case of allocating for the V file, the sector included in the AV block is changed to be already allocated on the sector bitmap only for the sector to which data is actually written. That is, unlike the case where an AV file is recorded, a sector in which data is not recorded is not assigned to a sector bitmap. By doing so, a plurality of non-AV files can exist in an AV block for a non-AV file, and the use efficiency of the entire disc can be improved. In this way, even in a file system that supports only the sector bitmap, the non-AV
A non-AV file can be written to the file AV block.

Further, when a non-AV file is written on a DVD-RAM file system using a file system that supports only the sector bit map, the sector bit map is first searched for
In the V block management table, if there is a sector that has been assigned on the sector bitmap even though it is a sector included in the AV block that is 00 in the V block management table, the AV block including that sector is set to 10 and the AV block By registering the disc in the management table, the validity of the disc on the DVD-RAM file system can be maintained.

In addition to the recording method of the AV block, data recording is performed by the alternative sector method for the AV block having the non-AV attribute. However, since there is no particular limitation on the recording method, an ECC block skip method (an ECC block having an address error is skipped and recording is performed on the next ECC block. See Japanese Patent Application No. 8-258078 filed by the present applicant). ), Data may be recorded. However, it is only necessary to avoid that the above two recording systems are mixed in the same ECC block.

The preceding is an explanation of the principle of the AV block. (AV file system AP that processes AV blocks
I) Of the commands of the file system shown in FIG. 1, the commands required for explaining the operation of the DVD recorder of FIG. 24 and relating to the AV block management table and the sector bitmap will be described here.

"WRITE" is executed when writing a non-AV file on a disc. "WRITE" is AV first
The block management table is searched to find an AV block having 10 states. Next, the state of the sector in the found AV block is searched from the sector bitmap, and if there is an unallocated sector, data is written to the unallocated sector, and the state where the data is written is changed to the allocated state on the sector bitmap. I do. If there is not enough free space in the AV block having the state of 10, or if there are no 10 AV blocks, the 00 AV block is newly assigned as an AV block for a non-AV file, and the non-AV Write the AV file. In this case, the processing for the sector is performed in the same manner as described above. No non-AV file is written to the AV block in the 01 state. As described above, the AV file system uses Extent
When writing non-AV files, if the sectors to be recorded become discontinuous, a new Exte
nt will be generated.

“AV-WRITE” is executed when writing an AV file on a disc. "AV-WRITE"
The AV block management table is searched for an AV block with a status of 00 from the beginning, and the AV
An attempt is made to secure a new AV block in a form that is continuous with the blocks. All sectors existing in the secured AV block are changed to assigned on the sector bitmap. Then, an AV file is written into the secured AV block from the beginning. FIG. 12 shows how the AV block management table and the sector bitmap change in “AV-WRITE”. Contrary to "WRITE", "AV-WRITE" does not record an AV file for an AV block having 10 states. In the AV file system, files are managed as Extent link lists. If the AV blocks become discontinuous while writing the AV file, a new extent is generated. When an AV file is recorded across zones, since a buffer sector exists at the boundary of the zone of the disk, the extent is always cut off at the zone boundary.

"READ" is executed when data recorded on the disk is read out by a specified size. “READ” reads about 32 KB of data at a time. This is because the minimum unit ECC block for error correction is composed of 16 sectors, which is 32 KB. When the data reaches the boundary of the extent while “READ” is continuously reading data, the CPU jumps to the extent where the next data is recorded, and reads data again from there. The jump time is 1.5 seconds in the worst case. However, since the AV file is continuously recorded in the sector in the AV block, by reading all the sectors in the AV block continuously, the jump time is 1.5 seconds. During this time, the data provided to the AV decoder can be stored in the track buffer, and seamless reproduction in the AV file can be guaranteed. "READ" itself does not process AV blocks or sector bitmaps.

[0131] Note that "READ" contains A in the designated data.
If there is a boundary between V blocks, the data read time can be optimized by first reading out the data recorded in the AV block closer to the current head position.

"DELETE" is executed when deleting the entire file existing on the disk. "DELETE" will follow all Extent links in the file,
Delete the tent to delete the file. "DE
The processing for the AV block and the sector bitmap changes depending on whether "LETE" is executed for an AV file or a non-AV file. Each case will be described.

First, the case of deleting an AV file will be described with reference to FIG. FIG. 13 shows an example of deleting AVfile # 2. AVfile # 2 is recorded in AV_BLK # 11 and AV_BLK # 14, and AV_BLK # 11, # 1 in the AV block management table.
4 (shaded area) is registered as 01. When the “DELETE” command is executed for AVfile # 2, the AV_BLK # 11 and # 14 parts of the AV block management table are changed from 01 to 00,
It is managed as an unused AV block. AV_BLK
The sector bitmap is changed from the assigned state (each bit is 1) to the unassigned state (each bit is 0) for all the sectors included in # 11 and # 14.

Next, a case of deleting a non-AV file will be described with reference to FIG. Non-AV file to delete
# 3. file # 3 is composed of one Extent,
The contents of the Extent are as shown in FIG. Now, if file # 3 is deleted by the "DELETE" command, the contents of the sector bitmap (the part enclosed by the thick line) from sector # 100 to sector # 110 where file # 3 was recorded will be changed from the allocated state to the unallocated state. Is changed to The processing of the state of the AV block management table of AV_BLK # 11 changes depending on the state of the sector bitmap at that time. AV other than file # 3
_ If a non-AV file is recorded in BLK # 11, AV_
The AV block management table is not changed for BLK # 11. However, if there is no longer any file in AV_BLK # 11 by deleting file # 3, the AV block management table for AV_BLK # 11 is changed from 10 to 00, and AV_B
Make LK # 11 unused.

"SHORTEN" is executed when an unnecessary portion of the AV file is deleted. “SHORTEN” is a command valid only for AV files, and an error occurs when executed for non-AV files. "SHORTEN" has A
There are two ways of use, one for deleting the beginning of the V file and the other for deleting the end of the AV file.

[0136] Partial deletion of an AV file is realized by changing Extent information if the processing can be completed within one AV block. In other words, it does not actually delete the data from the disk, but changes the Extent information so that the unnecessary data is not visible from the file system. In the case of an AV block of AV, the file system extends the release of the AV block.
Judge based on information. Because the AV block is A
When allocated for a V file, all sectors included in the AV block are determined to have been allocated on the sector bit map, and are allocated to a location where no data actually exists. It is not possible to determine where data exists. If the deletion of the data results in an AV block having no valid (recognized by the file system) data, the AV block is changed to an unused state. This process is called "DELET
Do the same as described in section "E".

Referring to FIG. 15, “SHORTE
The concept when “N” is executed will be described. However, FIG.
In FIG. 5, for simplicity, it is written that data is actually deleted from the AV block, but it should be noted that data is actually left as described above. FIG. 15A shows an example of deleting a portion before the edit point of AVfile # 1. By executing "SHORTEN",
The data before the edit point of AVfile # 1 becomes invisible from the file system. FIG. 15 (b) shows “SHORTEN” for deleting unnecessary parts at the end of the AV file.
This is an example of executing a command. In FIG. 15B, AV
When the part before the edit point of file # 1 is deleted, AV_BLK # k becomes an AV block having no valid data. In this case, it is necessary to change AV_BKL # k to an unassigned state as described above.

“SPLIT” is executed to divide an AV file into two. “SPLIT” is a command valid only for an AV file, and an error occurs if executed for a non-AV file.

An example of dividing AVfile # 1 will be described with reference to FIG. Here, the name AVfile # 1 is used even after AVfile # 1 is divided, but this is done so that the file before division inherits the management information of the original file as it is. By “SPLIT”, AVfile # 1 is divided into AVfile # 1 (of reduced size) and AVfile # 3. At this time, two files exist in the AV block AV_BLK # m including the edit point. When recording AV data in an AV block, since only one AV file can be recorded in one AV block, data of coexisting AVfile # 3 must be moved to another unallocated AV block. Therefore, find AV_BLK # n in the state of 00 from the AV block management table, change the state to 01, and
Move # 3-1 from AV_BLK # m. Since the AV_BLK # n is an AV block for an AV file, all the sectors included in the AV_BLK # n are assigned and the sector bitmap is changed.

"MERGE" is executed when two AV files on a disc are connected to data on a memory to create one AV file. The movement of data in "MERGE" will be described with reference to FIG. FIG. 17 shows AVfi
This is an example in which le # 1, AVfile # 2, and data in the memory are connected in the order of AVfile # 1, data in the memory, and AVfile # 2.
In this case, the continuous recording length of AVfile # 1 immediately before the out point is longer than the AV block length, and a vacant area equal to or larger than the data size in the memory after the out point of AV_BLK # n (already assigned to the vector bitmap. ), The data in the memory is written there, and the merge processing is completed by setting the jump destination from the end of the data in the memory to the in point of AV_BLK # n.

However, in the case of FIG. 17B, that is, when the data size k in the memory is larger than the size i of the free area after the out point, the data in the memory cannot be written in the free area. The data of AVfile # 1 is moved to the free area, and the data length is A together with the data in the memory.
It is necessary to create continuous recording data longer than the V block length. This is because there is a restriction that continuous recording data other than the end of the AV file must be longer than the AV block length. As described above, the concept of determining the length of the AV block is that if this constraint is not satisfied, continuous reproduction in the file cannot be guaranteed. In the case of FIG.
Since # 1 already met this condition, we did not move the data.

The limitation that the continuous recording data length of the AV file must be equal to or longer than the AV block length is as shown in FIG.
This applies to the case where the sum of the data size 1 of the AV file # 1-1 and the data size k in the memory is equal to or less than the AV block length, and data movement occurs. If the data in memory
When writing to AV_BLK # n, the data length in AV_BLK # n becomes shorter than the AV block length, and continuous playback in the file cannot be guaranteed. In order to create continuous data longer than the AV block length, AVfile # 1-1 and AVfile # 1-2 are grouped into two in a region where two free AV blocks are continuous, and memory data is added after that. Create continuous recording data longer than the block length. This makes it possible to guarantee seamless playback within a file. AV
If the continuous data length after the in point of file # 2 is less than the AV block length, this also corresponds to the case where the continuous recording data length in the file is less than the AV block length. Must be configured.

It should be noted that in FIG.
If the size j of the free area immediately before the point is larger than the size k in the memory, the data in the memory is
A method of performing a special recording method such as writing to the empty area immediately before the in point from the middle of the AV block and merging the file may be considered. In the following, several methods for writing data in memory before AVfile # 2 will be described. In this case, note that data in the memory is written in AVfile # 2 in a continuous manner. This is also a direct cause of writing from the middle of the free space. This is because the AV data must be recorded in a continuous form as much as possible so that an extra jump does not occur in the same AV block. In some cases, continuous playback of recorded AV data cannot be performed due to an extra jump.
If the data in the memory can be written immediately before the in point, FIG.
In the case of (b), no extra data movement occurs. Of course, a condition is required that the continuous recording data length generated by combining the in-memory data and AVfile # 2 is equal to or longer than the AV block length.

When the sum of the free areas i and j is larger than k in FIG. 17B, as shown in FIG.
As long as the data in the memory can be written in the free area of size i as much as possible, and the remaining data in the memory can be written in the free area of size j in such a way as to be connected to AVfile # 2,
There is no need to move extra data. However, also in this case, the two generated AV data need to satisfy the condition that the length is equal to or longer than the AV block length. The procedure for writing data in the memory is based on the size j existing before the in point.
It is also conceivable to write data in a free area after the out point and then write the remaining data into a free area existing after the out point.

Even if the method of FIG. 22A is considered, if the sum of i and j is smaller than the data size in the memory, as shown in FIG. You have to move. However, in this case, as shown in FIG. 17B, the data in the AV file # 1 is not moved by one AV block, and
Note that only the data to be supplemented is moved to make the data of the block length.

As a special case of FIG. 22B, there is a case where two data adjacent to the in point and the out point both have a size smaller than the AV block length as shown in FIG. As shown in (a), if there is an empty area before the in point and after the out point, and if the data in the memory is divided well,
If both data can have a size equal to or larger than the AV block length, extra data movement does not occur if data is written in the memory as shown in FIG.

Even in the case where there is no empty area adjacent to the in-point and the out-point (b), the data adjacent to the in-point and the out-point and the data in the memory are added together to determine the AV block length. If the above data can be constructed, these three data can be moved by allocating a new free space.
Data on other disks does not need to be moved.

The various "MERGE" methods have been described with reference to the drawings. What is common in either case is that at worst, the moving data size can be suppressed to 2 AV block length or less. It is. However, as an exception, when two consecutive free AV blocks are required but only one isolated free AV block exists, data of one AV block is moved to create two continuous free AV blocks. It is necessary. Only in this case, extra data movement of one AV block occurs, and as a whole, data movement of three AV blocks occurs.

Note that “MERGE” can also merge one file and data in the memory. FIG.
If you do not specify AVfile # 2 in (in the case of NULL),
Merging of AVfile # 1 and data in the memory is performed, that is, data in the memory is added after the AV file. If AVfile # 1 is not specified, data in the memory will be added before AVfile. Further, if no data in the memory is specified, it is possible to perform a merge of only the file.

In the foregoing, only seamless playback within a file has been considered. However, considering seamless playback between files, the "MERGE" process considered so far is insufficient. In order to realize seamless reproduction between files, the conditions that must be satisfied in recording an AV file become more severe. If only seamless playback within a file is guaranteed, the continuous recording length of data is longer than the AV block length in all AV blocks except the first and last AV blocks among the AV blocks that record a series of AV files. I wish I had a size.
However, in order to guarantee seamless playback between files, all AV blocks that record an AV file use A
The continuous recording length of the V data must be equal to or larger than the AV block length, and the data movement as shown in FIG. 17B and FIG. 18 is required even at the end of the AV file. .

“IN_AV_BLK_BOUND” is executed when it is determined whether or not there is an AV block boundary between two designated points. If an AV block boundary exists in the specified section, TRUE is returned; otherwise, FALSE is returned. In FIG. 19, AV_BLK # 1 is used for both points a and b.
00, so specify these two points and select “IN_AV_BLK_BO
UND "returns TRUE. Since there is a boundary between AV_BLK # 100 and AV_BLK # 101 in the section defined by points a and c, executing "IN_AV_BLK_BOUND" with points a and c specified returns FALSE. This command is meaningful in the case of trick play (trickle play), and will not be described in detail in the trick play information.

“SEARCH_DISCON_AV_BLK” checks whether there is an AV block discontinuous boundary in the designated section.
Returns TRUE, otherwise returns FALSE. Here, an AV block discontinuous boundary is defined. An AV block discontinuous boundary is two AV data recording continuous data of an AV file.
A state in which another AV block is sandwiched between blocks or a zone boundary is sandwiched.

FIG. 20 shows an example of an AV block discontinuous boundary. In (a), AVfile # 1 records data continuous to AV_BLK # m and AV_BLK # n. A plurality of AV blocks exist between AV_BLK # m and AV_BLK # n, and an AV block discontinuous boundary occurs. Extent is also divided here.
In the case of (b), AVfile # 2 is continuously recorded in AV_BLK # k and AV_BLK # l. However, since these two AV blocks belong to different zones #i and #j, an AV block discontinuous boundary occurs between the AV blocks.

"SEARCH_DISCON_AV_BLK" is executed by designating two points. The determination that the AV blocks are discontinuous is made based on the Extent information of the AV file. If the specified two points are included in the same Extent, there is no AV block discontinuous boundary between the two points. 2 specified
If a point is included in different extents, and there is a gap longer than the AV block length between the extents, or if the extent is included in different zones, an AV block discontinuity boundary must exist between the two specified points. Will exist.

Another case where the extent is divided is a skip of an ECC block. When recording an AV file, when an address error occurs, a recording method of skipping the ECC block and recording data in the next ECC block (described in Japanese Patent Application No. 8-258080).
Is executed, the ECC block is skipped and Exte
nt discontinuity occurs. Therefore, when recording an AV file by skipping an ECC block due to an address error, the Extent is divided, but in this case, the Extent is a discontinuous boundary of the ECC block within the AV block and does not correspond to the AV block discontinuous boundary. And

When editing the AV data using the primitive AV file system commands introduced up to this point, extra data may be moved.

For example, there is a command "SPLIT" in the command for dividing the AV data.
Due to the condition that "V data occupies one AV block", a file generated after the edit point by being divided by "SPLIT" must be moved to another area. However, in editing, "SPLIT" is rarely used alone, and there are many cases in which there is a subsequent process.

For example, when deleting an unnecessary part such as a CM cut, it is necessary to divide the AV file in the unnecessary part in order to execute “SHORTEN”. In this case, “SPLIT” is executed in the area to be cut, and the newly created A
The data of the V file is moved. If data is moved in spite of unnecessary data to be deleted by “SHORTEN” executed in the subsequent flow, unnecessary data movement occurs one more time. If "SH
If two files created after removing unnecessary parts to be deleted by "ORTEN" exist in different AV blocks, no data movement occurs at all. Considering this, if the processing from “SPLIT” to “SHORTEN” is implemented as one command, it is possible to reduce the movement of unnecessary data.

In the execution of "SPLIT", if the processed data does not have to satisfy the definition of the AV file, no extra data movement occurs. This can reduce the amount of data movement, but complicates AV block management somewhat. When an AV file has data in a part of an AV block, the AV block cannot be immediately unused just because it is deleted. This is because data of another AV file may exist in the remaining free area of the AV block. You can do this by examining the Extents of all files, but doing this can be cumbersome. As a solution to this, a method of giving the start address and the size of the AV file recorded in the AV block management table can be considered.
If only the AV file system accesses this disk, the sector bitmap is stored in the AV attribute of the AV attribute.
By registering only blocks in which data has been recorded as allocated, it is possible to confirm the existence of data in an AV block, thereby removing the restriction that an AV file occupies one AV block. I can do it.

4. Presentation Data Here, an AV file containing an MPEG stream (hereinafter, also referred to as “presentation data”) will be described.

(Directory Structure) FIG. 29 shows the structure of directories and files required in the DVD recorder.

All files of the DVD recorder are recorded in a dedicated directory called DVD_RAM_AV under the ROOT directory. AV_File_ # with extension ".avf" in the figure
1.avf ,,, AV_File_ # n.avf indicates an AV file in which actual MPEG data is recorded. These files are continuously recorded using the above-mentioned AV block. In the figure, the extension “.ifo” is an AV file management information file described later, and “.clp” is an AV Clip part of the AV clip management information file described later. An AV file, an AV file management information file, and an AV clip management information file are associated with each other by a body name without an extension. Also, the file name is “Clip_S
“equence” is an AV Clip Sequence of an AV clip management information file described later.

Hereinafter, the logical format of the AV file will be described. First, MPEG data will be briefly described, and then the data structure and creation method of an AV file that can be edited will be described.

(Video Data) First, MPEG video data will be described.

In MPEG video compression, in order to increase the compression ratio, not only compression using spatial frequency characteristics in a frame but also data compression using time correlation characteristics from the past and the future are performed. Referring to FIG. 30, MP
In the EG, there are three picture types: a B picture that refers to the past and the future, a P picture that refers only to the past, and an I picture that does not use temporal correlation. Since P and B pictures using compression by time correlation are decoded based on information of a picture of a reference destination, decoding of these pictures requires decoding of a picture of a reference destination. Therefore, a B picture that is being referenced from the future cannot be decoded until after the future picture that is the reference destination has been decoded. Therefore, in the MPEG compressed data (stream), unlike the picture reproduction order (display order), the B picture is arranged after the reference picture (coding order) as shown in FIG. This change of order is called reordering.

P and B pictures using time correlation are
Since decoding is performed using the picture of the reference destination, decoding alone cannot be performed. Therefore, if only P and B pictures are used consecutively, a problem occurs when decoding is performed from the middle of a stream in special playback or the like.
Generally, an I picture is inserted approximately every 0.5 seconds. A GOP (Group of Pictures) is referred to as a GOP (Group of Pictures) from the top of the I picture, and is treated as one compression unit in MPEG.

In MPEG, since the compression using time correlation is performed, in the GOP, the last frame in the display order ends with an I or P picture, and the coding ord
There is a restriction that the first frame in er must be I.

(Multiplexing) Next, multiplexing of AV data will be described.

Video and audio data is about 2 KB
And stored in a 2 KB fixed-length pack.
The video pack and the audio pack are arranged in the order of input to the decoder buffer, and the pack header is an SCR that is a time stamp indicating the input time to the decoder buffer.
(System Clock Reference) is added. A PTS (Presentation Time Stamp) indicating the playback time of video and audio data is included in the packet header in the pack.
And a DTS (Decoding Time Stamp) indicating the decoding time.

The video multiplexed in this way,
Audio data has the following features.

In MPEG video, in order to realize high-efficiency compression, a decoder has an input buffer so that a dynamic code amount can be dynamically allocated according to the complexity of an image in units of frames, and data is previously stored in this decoder buffer. By storing the data, a large amount of data can be allocated to a complicated image that is difficult to compress. Therefore, it is necessary to input video data to the decoder buffer earlier by the time for storing data in the decoder buffer in advance (hereinafter referred to as "VBV delay"). On the other hand, since the audio data is encoded at a fixed size in each frame, it is not necessary to input data particularly earlier than the decoding time unlike video data. Therefore, the video data is multiplexed before the audio data.

FIG. 31 shows an example of multiplexing of video data and audio data. FIG. 31 shows, from the top, (a) video data, (b) video buffer status, (c) AV data obtained by multiplexing video data and audio data (packed and multiplexed, respectively), ( d) shows audio data. The horizontal axis shows a time axis common to each figure, and each figure is drawn on the same time axis. In the state of the video buffer, the vertical axis indicates the buffer occupancy (the amount of data stored in the video buffer).
The bold line in the figure indicates the temporal transition of the buffer occupancy. The slope of the bold line corresponds to the video bit rate, indicating that data is being input to the buffer at a constant rate. The fact that the buffer occupancy is reduced at regular intervals indicates that the data has been decoded. The intersection of the dotted line (diagonal line) and the time axis indicates the start time of data transfer of the video frame to the video buffer.

Hereinafter, a complicated image A in video data will be described as an example. As shown in FIG. 31B, since the image A requires a large amount of code, the data transfer to the video buffer must be started earlier than the decoding time of the image A by VBV delay in the figure. As a result, the position (time) of the shaded video pack is used as the AV data.
Are multiplexed. On the other hand, since the transfer of the audio data is not particularly advanced than the decoding time, it is multiplexed at the position (time) of the shaded audio pack in the figure. Therefore, the video data and the audio data reproduced at the same time are multiplexed in a state where the video data precedes. In MPEG, the time during which data can be stored in the buffer is limited, and it is specified that all data must be output from the buffer to the decoder within one second after being input to the buffer. Therefore, the difference in multiplexing of the video data and the audio data is 1 second at the maximum (strictly speaking, there may be a further shift by the reordering of the video data).

In this example, the video precedes the audio, but it is theoretically possible that the audio precedes the video. When a simple image having a high compression ratio is prepared for video data and audio data is transferred unnecessarily early, such data can be intentionally created. However, due to the restrictions of MPEG, the maximum advance can be 1 second.

(Edited AV File) Next, the structure of the AV file as a result of editing will be described.

Description will be made using three AV clips shown in FIG. 26 as an example. First, before editing, A
Both the V files #j and #k are recorded in the standard MPEG data format shown in FIG. In other words, time stamps that are temporally continuous are assigned with the SCR of the first pack set to 0. Such MPEG data is called a VOB (Video Object) in DVD.
The exact definition of VOB is the MPEG program stream specified in ISO / IEC13818-1, and finally program_end_code
Is not attached.

When three clips are edited to form one AV file, one AV file composed of three VOBs corresponding to the three clips is created as a result. At the boundary of each VOB, the time stamp at the beginning differs depending on the position cut out from the original VOB, so the time stamp is not continuous at the boundary of the VOB.

FIG. 32 shows a case where time stamps are not continuous.
5 is an example of an AV file composed of one VOB. At the VOB boundary point created by editing, discontinuity of the time stamp occurs. Since the decoder model (hereinafter referred to as “STD”) defined by MPEG is a model that specifies the operation for a stream with continuous time stamps,
When the time stamp discontinuity occurs as in this example, VO
Continuous playback (hereinafter referred to as “seamless playback”) cannot be performed without breaking the B boundary.

In the (E-STD) DVD, an extended STD model (hereinafter referred to as “E-STD”) shown in FIG. 33 is defined, and seamless playback between VOBs having discontinuous time stamps is realized.
E-STD is an adder that adds an offset to the system time clock (hereinafter referred to as “STC”) that ticks the reference time compared to the STD defined by MPEG, and one of the output value of the STC and the output value of the adder Is added to the switch. With this configuration, by adding a unique offset (called “STC offset”) between VOBs,
As shown in (1), it is possible to make time stamps continuous between VOBs. STC offset is defined as follows. Expressed as a timestamp in the stream
The VOB video display start time is defined as VOB_V_S_PTM, and the end time is defined as VOB_V_E_PTM. At this time, the STC offset is obtained from the following equation.

STC offset = VOB_V_E_PTM of front VOB-VOB_S_PTM of rear VOB Next, the operation of the E-STD will be briefly described. What is important in the operation of the E-STD is that each switch (SW1, SW2,
SW3 and SW4). Among them,
The switching timing of the demultiplexer switch SW1, which needs to be switched first, is particularly important because it can be a switching advance notice for other switches.
Since the switch SW1 is switched from the start of input of the last pack of the front VOB to the start of input of the first pack of the rear VOB, S indicating the input start time of the last pack of the front VOB
The CR (LAST_SCR) and the SCR (FIRST_SCR) indicating the input start time of the first pack of the rear VOB may be given to the E-STD as the SW1 switching timing. This enables switch switching during normal reproduction.

The ADPI (Audio Decoder Pa) shown in FIG.
Use Information) is information for temporarily stopping the audio decoder, and will be described in detail later in the audio gap.

Since the basic operation of the E-STD is described in detail in International Patent Publication No. WO97 / 13364, further description in this example will be omitted.

(Audio Gap) There are several conditions for realizing seamless playback between VOBs. The first is the audio gap. AV data handled by DVD is NTSC and PAL for video, and Dolby AC3, MPEG and LPCM for audio. All of these AV data have the concept of a frame, and are configured with different frame periods. Specifically, NTSC
One frame of about 33msec (accurately 1 / 29.97sec), PAL
One frame is 40 msec, and one frame of Dolby AC3 is 32 m
sec, 1 frame of MPEG is 24 msec, 1 frame of LPCM is about
It is 1.67 msec (exactly 1/600 sec). For this reason, since the frame cycle is different for any combination of video and audio, when editing a VOB by cutting out an VOB from an AV file and connecting two different VOBs, if a VOB is connected based on video, a gap in the reproduction of audio data is generated. Or, an overlap occurs, and V
If the OBs are connected, a gap or overlap in reproduction occurs in the video data. In a DVD, it is assumed that an audio gap is generated based on video.

FIG. 34 is a diagram showing an audio gap of this example. Although a time stamp indicating a decoding time or a reproduction time is added to the MPEG data, it is not added to every frame. For example, in the case of video data, it is often attached once per GOP, that is, once every 0.5 seconds. Therefore, in the decoder, AV
Taking advantage of the fact that the data has a constant frame rate,
AV data is continuously decoded at a constant period. When AV data having an audio gap is input to such a decoder, an audio pause is instructed from the system control unit (recording / editing / playback control unit in FIG. 24) to the decoder, and the audio decoder is temporarily stopped. There is a need. This is the ADPI (Audio Deco
der Pause Information).

In this way, the audio gap can be processed. However, the audio gap can be processed in a short period of time because the system controller composed of a general-purpose microcomputer and software controls the stop of audio decoding. If the process is continued, there is a problem that the processing of the system control unit cannot be completed in time. Therefore, it is necessary to set a limit so that the audio gap is opened over a certain interval.

In this example, the following restrictions are provided to avoid the above problem. Since the audio gap is generated by editing, the maximum number of audio gaps is one for each editing unit, that is, VOB. Next, the time length of the audio gap is limited to less than one audio frame. Next, the position (time) in the audio gap is limited. Since the audio gap is caused by editing, the editing boundary V
Based on the OB video display start time, the start time of the audio gap is limited to be less than one audio frame before and after. Next, the time length of the VOB is limited to 1.5 seconds or more.

By providing the above-mentioned restrictions, it is possible to minimize the generation interval of the audio gap during seamless reproduction to “1.5 seconds−1 audio frame reproduction time × 2”. Specifically, when numerical values are applied, if the audio is Dolby AC3, the playback time of one audio frame is 32 msec, and therefore, the minimum interval between audio gaps is 1436 msec.

(Buffer Control) The second is the continuity of buffer control between VOBs.

As described in the description of the multiplexing, the MPEG
In order to realize highly efficient compression, an input buffer is provided in a decoder so that a necessary code amount can be dynamically allocated to each frame. On the other hand, if the decoder buffer underflows or overflows, normal reproduction cannot be performed, and troubles such as interruption during reproduction may occur. Therefore, in the MPEG encoder, a simulation of the decoder buffer is performed by providing a virtual buffer inside the encoder so that the decoder buffer does not underflow or overflow.

The MPEG data encoded in such a series of flows is guaranteed not to cause overflow or underflow of the decoder buffer. However, when seamless playback is performed between separately encoded VOBs, the There is no guarantee that the decoder buffer will not underflow or overflow between them. Therefore, in order to perform seamless playback between VOBs, it is necessary to guarantee the continuity of buffer control between VOBs. Therefore, near the VOB boundary, local re-encoding is performed so that buffer control between VOBs is continued during editing.

(Re-encoding method) Next, local re-encoding near the VOB boundary will be described.

First, the AV data at the VOB boundary is read. The reading method and the editing operation described later will be described. The read AV data is separated into video data and audio data, and each is processed.

First, video GOP reconstruction is performed. Video reconstruction will be described with reference to FIG.

Since video data is encoded in units of one GOP, if the GOP is broken by editing, decoding of all frames cannot be performed normally. Therefore, when editing is requested in a frame in the middle of a GOP as shown in FIG. 35, it is necessary to newly create a GOP. As described above, the MPEG has three types of picture types: an I picture that does not use temporal correlation, a P picture that uses temporal correlation from the past, and a B picture that uses temporal correlation from the past and the future. There is MPEG B
For pictures, I and P pictures, the display order of frames (display order) and the compressed data order of frames (cod
ing order) has been replaced. In the GOP, the last frame must end with an I or P picture in the display order of the frames, and the first frame must be I in the compressed data order of the frames (coding order). Therefore, when reconfiguring the GOP, processing according to the following rules is required.

The first is the display order of frames in the last GOP of the preceding VOB. If the last picture is a B picture, this picture is replaced by a P picture (or I picture). Change the type (rule 1).

Second, in the first GOP of the rear VOB, if the first picture is a P picture in the compressed data order (coding order) of the frame, the picture is changed to an I picture (rule 2).

Third, in the first GOP of the rear VOB, the display order of the frames is as follows.
For a B picture displayed before a picture (including an I picture changed according to the rule 2), the compression method in the picture is changed so that the temporal correlation from the past is discontinued (rule 3).

Next, video buffer control is performed. The video buffer control will be described with reference to FIG.

As described in the description of the buffer control, the video buffer control must be continuous between two VOBs to be edited. Therefore, re-encoding near the VOB boundary is performed so that the buffer control between the two VOBs can be continued. FIG. 36 is a diagram illustrating buffer control at the time of re-encoding. First, the buffer occupancy Bv1 at the VOB boundary is determined. Bv
Although there is no limitation on how to determine 1, in this example, the buffer occupancy before re-encoding in the rear VOB is used as it is. front
In the VOB, the buffer occupancy at the end of the front VOB (VOB boundary) is B
The code amount is re-allocated in each picture so as to become v1, and each picture in the re-encoding section is re-encoded. For the rear VOB, the buffer occupancy at the beginning of the VOB is Bv1
The buffer control in the re-encoding section is performed so as to be connected to the buffer occupancy Bv2 (the same value as before the re-encoding) in the first picture after the re-encoding range.

With the above processing, the buffer control between the two VOBs can be continued. In addition, the range in which the video is re-encoded requires at least a frame whose picture type is changed, but is not limited to this. The re-encoding may be performed across a plurality of VOBUs. However, there is a demerit that the re-encoding time during editing becomes longer as the re-encoding range becomes longer.

Next, an audio gap is created.
A method for creating an audio gap will be described with reference to FIG.

As mentioned in the description of the audio gap,
Depending on the difference between video and audio frame rates,
Audio discontinuity (audio gap) occurs at the VOB boundary during editing. The audio gap start time is before and after the VOB video display start time VOB_V_S_PTM.
Must be less than one audio frame. Therefore, the audio gap is determined as follows. In FIG. 36, time t2 is VOB_V_S_PTM of the rear VOB,
Using this time t2 as a reference, the front VOB uses up to the audio frame including the time t2 as edited data (when the time t2 matches the audio frame boundary,
The frame up to the end of playback at time t2 is used), and the rear VOB edits from the first audio frame that starts playback immediately after (including the same time) the playback end time t3 of the last audio frame of the front VOB. Use as data.

The time from the reproduction end time t3 of the last audio frame of the front VOB to the reproduction start time t4 of the first audio frame of the rear VOB becomes the audio gap length A_GAP_LEN, and the reproduction of the first audio frame of the rear VOB starts. Audio gap length A_GAP_LEN from time t4
The time just before is the audio gap start time A_STP_PTM
become.

Next, multiplexing of video and audio is performed. The multiplexing of video and audio will be described with reference to FIG.

After the above-described reconstruction of the video GOP structure, buffer control, video re-encoding, and creation of an audio gap, video and audio multiplexing are performed. As described in the description of multiplexing, video data is multiplexed prior to audio data due to the nature of MPEG. The same applies to the VOB boundary. At the beginning of the rear VOB, the audio data in the front VOB is wrapped around the rear VOB by the amount of the video data multiplexed earlier. Conversely, the front VOB is multiplexed except for the audio data wrapping around the rear VOB. Specifically, based on time t1 (video data input start time at the rear VOB) in FIG. 36, the audio frame whose reproduction is started immediately after the time t1 (not including the time t1) is multiplexed as the rear VOB. Perform the conversion.

[0206] Regarding a method of continuity and multiplexing of buffer control between VOBs, refer to the patent "International Publication No.
13367 "and International Publication Number WO97 / 1336
3 ", and further description in this example is omitted.

[0207] By using the concept of seamless connection / playback, even MPEG data that cannot be edited without rewriting all time stamps can be easily edited without moving or rewriting a large amount of data. Becomes

(Non-Seamless Connection) Next, non-seamless connection will be described.

[0209] Multiple VOBs in an AV file by editing
What we can do is explained earlier. Here, a non-seamless connection between VOBs in an AV file will be described.

[0210] Although the seamless playback method between VOBs and the seamless stream creation method have been described above, it is not possible to convert all MPEG streams to be edited into a seamless stream.

For example, two audio streams to be connected by editing, one of which is an AC3 stream,
If the other stream is an MPEG stream, seamless playback cannot generally be performed by a decoder even if seamless stream processing is performed on the connection boundary. This is because, when the stream changes from AC3 to MPEG, the stream attributes are switched inside the decoder, so that decoding stops during this time. The same applies when the attribute of the audio stream changes.

If at least one of the following conditions is satisfied:
Seamless playback is no longer possible. 1) Different video frame rates (NTSC, PAL, ...) 2) Different audio encoding methods (AC-3, MPEG, LPCM) 3) Different audio stream bit rates Also intentionally non-seamless connection It is also possible to use VOB.

In the above case, editing is performed so that a non-seamless stream is formed between VOBs. Specifically, re-encoding is performed as follows. Read first VOB boundary, A
It is the same up to the separation of the V stream and the reconstruction of the video GOP. There is no need to control the video buffer between VOBs, it is only necessary to control the video buffer for each VOB alone.Even when extracting audio data, strict calculations are not required unlike the creation of audio gaps, and almost match the video You just need to cut out the audio data you want. Also, multiplexing of video and audio may be performed only by multiplexing video and audio in each VOB.

[0214] 5. Navigation data 5.1. AV file management information About logical format of AV file management information FIG.
This will be described with reference to FIG. FIG. 55 shows the whole AV file management information, and FIG. 57 shows Stream Information which is a part of the AV file management information.

(AV File Management Information) AV Fil
The e Management Information records an AV File ID indicating the identification name of the corresponding AV file, a Sector Size indicating the size of the AV file in sector units, a Playback Time indicating the playback time of the AV File, and a VOB Table. You.

[0216] The VOB Table contains the V included in the AV file.
The number of OBs and a pointer to an array of VOB Information are recorded. VOB Information contains Start address indicating the address of the beginning of the VOB with the relative sector address from the beginning of the AV file.
Sector, start of AV file with time code 00:00:00:
Start Time, T indicating the playback elapsed time at the beginning of the VOB as 00
Sector Offset for correcting Start Sector value recorded in ime Map, Time Offset for correcting Start Time value recorded in Time Map, Stream Information indicating data attribute of VOB, and the following description Time Map Tab
le and VOBU Map Table are recorded.

In the Time Map Table, the number of Time Maps, Time
Time Unit indicating the time interval between maps in seconds, Time Base for correcting the time indicated by each Time Map, VOBU of Time Map
Index Offset to correct Map Index and Time Map
Is recorded. Time Map contains Time
VOBU Map Index indicating the index of the VOBU in the VOBU Map array corresponding to the time indicated by the Map, Start Secto indicating the relative sector address of the beginning of the VOBU from the beginning of the VOB
r, Start Time indicating the relative time code of the video frame at the head of the VOB from the head of the VOB is recorded.

[0219] The VOBU Map Table contains the VOBUs included in the VOB.
And the pointer to the array of VOBU Map are recorded. VOBU
In the Map, Star indicating the relative sector address at the beginning of the VOBU
t Sector Offset, Start TimeOffset indicating the relative time code of the video frame at the beginning of the VOBU, End Sector of the reference picture used for special playback such as fast forward or rewind, indicating the relative end address from the beginning of the VOBU
Reference Picture is recorded.

[0219] The Time Map Table records information on the VOBUs existing at a certain time interval from the beginning of the VOB. VOBU Map
The Table records information on all the VOBUs present in the VOB in order from the VOBU Map # 1 in the order in which they are present in the VOB.

[0220] Time Map # i indicates a VOBU existing at the time point of ((Time Unit) * i + (Time Base)) elapsed playback time from the beginning of the VOB. When Time Unit is 1 and Time Base is 0, Time Map # 1, Time Map # 2, and Time Map # 3 are Vs existing 1 second, 2 seconds, and 3 seconds after the beginning of the VOB, respectively.
Refers to OBU. Time Map VOBU Map Index and Time Map Tabl
The value obtained by adding the Index Offset of e indicates the index of the VOBU Map. Time Map StartSector and Time Map Table Sec
The value obtained by adding tor Offset indicates the start address of the VOBU indicated by the Time Map as an offset in sector units from the start of the VOB. Start Time of Time Map and Time Of of Time Map Table
The value obtained by adding fset indicates the playback elapsed time at the beginning of the VOBU indicated by the Time Map, as a time code with the time code at the beginning of the VOB being 00: 00: 00: 00.

The Start Sector and Start Time of the VOBU Map are
Expressed as a relative value based on the VOBU or VOB pointed to by the Time Map. VOBU before VOBU indicated by Time Map # 1
Is based on the beginning of the VOB. As for other VOBUs, the VOBU pointed to by the Time Map is set to be the closest VOBU. When the Time Unit is set to several seconds, the Start Time can be represented by 2 bytes and the Start Time can be represented by 1 byte.

FIG. 56 shows VOBU Map Index and VOBU of Time Map.
The relationship of BU Map is shown. (Stream Information) Stre
The am Information records a Video Attribute indicating various attributes of the video stream, an Audio Map Table indicating audio stream mapping, and an Audio Attribute Table indicating audio stream attributes.

The Audio Map Table contains Aud for 8 streams.
Record io Map. The Audio Map has a validity flag indicating whether the stream is valid and an Au in the Audio Attribute Table.
Record the index that points to dioAttribute.

[0224] The Audio Attribute Table records various attributes of eight audio streams. Vi
For deo Format, MPEG1 can be set in addition to MPEG2. Vide
o System can be set to NTSC or PAL / SECAM.
In SC, Video Resolution can be set to 352 x 480 and 352 x 240. Video Aspect can be set to 4: 3 or 16: 9 wide. In addition to the standard display mode, Letter Box or Pan Scan can be set for wide. Video APS indicates a method of copy protection control for an analog video signal, and AGC prevents a copy to a VTR by changing the signal amplitude of a blank section of a video signal to impair the AGC circuit of the VTR. Color Str
You can also set ipe or a combination of both. Audio Forma
t indicates the audio encoding format, such as MPEG2 or Dolby
Digital, linear PCM, etc. can be set. Audio Samp
ling indicates the sampling frequency, and 44.1 KHz or the like can be set. Audio Bitrate indicates a value for a fixed bit rate, and is described as "VBR" for a variable bit rate. These Attributes are used for initial setting of the AV decoder 131 in FIG.

(Generation of AV File Management Information) A at Recording
The generation of the V file management information will be described.

The AV data input unit 111 shown in FIG. 24 transfers information on the number of sectors for each GOP and the number of sectors of the reference picture to the AV file management information generating unit 112 in real time during recording.

The AV file management information generation section 112 generates the AV file management information up to the present time in real time from the AV file management information generated so far and the passed GOP information, and stores it on the main storage. Then, at the end of the recording, the AV file management information stored on the main storage is recorded as a non-AV file in the AV file management information file.

The AV file management information generation unit may store only the GOP information in real time, and generate the AV file management information before recording the AV file management information on the disc.

The AV file management information generating section may record the AV file management information in the AV file management information file before the end of the recording, as long as the real-time recording of the AV data is not hindered.

[0230] Fig. 58 shows an example of GOP information passed by the AV file management information generation unit from the start to the end of recording. FIG. 59 shows AV file management information generated from the GOP information in FIG. In addition, FIG.
An example of Stream Information in AV file management information when a book audio stream exists is shown.

First, the AV file management information generated after recording will be described. FIG. 59 shows an example of AV file management information generated from the example of GOP information in FIG. The contents of the AV file management information shown in FIG. 59 will be described. The number of VOBs is one. VOB Information
Sector Offset, Time Offset, TimeB of Time Map Table
Both ase and Index Offset are 0, and the value recorded in the Time Map indicates a value based on the VOB head without correction.

Next, the Steam Information shown in FIG.
Will be described. Audio Map and Audio Attribu
The first three of te are Valid, and the others are Invalid. Audio Map's Audio Attribute Index contains Audio Ma
The same value as the index of p is recorded.

(Split of AV file management information)
Update of the AV file management information accompanying the plit will be described. The operation when the trick-play information editing unit in FIG. 24 receives a Split request will be described using an example. FIG.
Indicates the previous AV file management information. FIG. 61 shows this A
This shows two pieces of AV file management information after V file management information is split between VOB # 3 and VOB # 4.

First, the contents of the first half of the AV file management information will be described. Since the first half of the AV file takes over the original file, the AV File ID is the same as before the Split. As the file size, playback time, and number of VOBUs have become smaller, the AV File Management Information
ector size, playback time, VOBU of VOBU Map Table
The number and Time Map number in the Time Map Table become smaller.

Next, the contents of the second half of the AV management information file will be described. A new name is recorded in the AV File ID. As the file size, playback time, and number of VOBUs become smaller, AV File Management Information
Sector Size, Playback Time, VOBU of VOBU Map Table
The number, Time Map number of Time Map Table, becomes smaller,
Same as the first half. Because the front of the file is gone,
The values recorded in each VOBU Map and each Time Map must be corrected. Therefore, instead of rewriting the values recorded in each VOBU Map and each Time Map, VOB Information
Sector Offset and Time Offset and VOBU Time T
Set the value to be corrected in Time Base and Index Offset of able.

[0236] Regarding Stream Information, the value before Split is directly used for both Stream Informatio after Split.
is carried over to n.

(Shorten of AV file management information)
In the case of Shorten at the time of editing, Split
Since the operation is the same as that of generating only one of the AV management information, the description is omitted.

(Merge of AV File Management Information) M at Editing
Update of the AV file management information accompanying the erge will be described. The operation when the trick-play information editing unit in FIG. 24 receives a Merge request will be described using an example. FIG.
e Indicates previous two AV file management information. FIG.
This shows the AV file management information after two VOBUs in which the two AV file management information are re-encoded are added and merged. However, after the front AV file, 1
It is assumed that a VOBU composed of 40 sectors of 2 frames is added, and a VOBU composed of 56 sectors of 18 frames is added before the rear AV file. FIG. 67 shows an example of two pieces of stream information before Merge, and FIG. 68 shows an example of two pieces of stream information after Merge.

The contents of the merged AV file management information will be described. A after Merge in the front AV file
Since the V file is inherited, the name of the preceding AV file is recorded in the AV File ID. AV File Managemen
Sector Size and Playback Time of t Information are values obtained by summing up two AV files and two VOBUs to be added. Since the two AV files are each a VOB, the number of VOBs is two. As for VOB # 1, VOBU # 5 was added and VOB became larger, so VOBU Map T
The number of VOBUs in able and the number of Time Maps in the Time Map Table increase. As for VOB # 2, the number of VOBUs in the VOBU Map Table increases as VOBUs are added and VOBs increase. With the addition of VOB and VOBU in front, each VOBU
The values recorded in the map and each time map must be corrected. Instead of rewriting the values recorded in each VOBU Map and each Time Map, the VOB Information Sect
or Offset and Time Offset and VOBU Time Table
Set values for correction in Time Base and Index Offset.

[0240] Regarding the stream information, the index of the valid one in the Audio Map is the same, and the Audio Att corresponding to the Audio Map of the same index.
The stream information of the second half AV file is changed so that the ribute attribute becomes the same.

(Size of AV File Management Information) In the example of FIG. 59, the Start Sector of the VOBU Map is 2 bytes, the Start Ti
me is 1 byte, End Sector of the Reference Picture
Is 1 byte, Start Sector of Time Map is 4 bytes, Start Time is 4 bytes, and VOBU Map Index is 2 bytes. For example, the playback time is 7 hours and each VOBU is 0.5 seconds A
If V data is recorded, the size of VOBUMap is about 200 kilobytes, and if the Time Unit of the Time Map Table is 5 seconds, the size of the Time Map Table is about 50
Since the size is kilobytes, the size of the AV file management information is about 250 kilobytes. If you access the disk by 32 kilobytes, you can access the AV with about 8 accesses.
File management information can be written to disk.

When performing special reproduction, it is possible to perform special reproduction by referring only to the Time Map Table without referring to the VOBU Map Table. However, the number of types of special playback speeds is reduced.

According to such a trick play method, trick play can be executed without storing the VOBU Map Table in the main memory even in a playback device having a small main memory capacity.

5.2. AV clip management information (logical format of AV clip management information)
The logical format of the clip management information will be described with reference to FIG.

First, the AV Clip part in the AV clip management information created for each AV file will be described.

[0246] The AV Clip part includes an AV File ID indicating the identification name of the corresponding AV file, and a Clip defined for each VOB.
The number of Information and a pointer to Clip Information are recorded. Clip Information is the VOB ID that identifies the VOB
, The video display start frame Start Time of the VOB indicating the relative value from the beginning of the AV file by the time code, the video display start time VOB_V_S_PTM of the VOB indicated by the time stamp in the stream, and the time stamp in the stream VOB video display end time VOB_V_E_PTM and VOB
FIRST_SCR, which is the input time to the buffer of the first pack
And LAS which is the input time to the buffer of the last pack of the VOB
Seam indicating the seamless connection information between T_SCR and the previous VOB
Less Information, Mark Table, and Clip Table are recorded. Seamless Information includes a Seamless Flag to determine whether a connection is seamless or non-seamless.
In addition, A_STP_PTM indicating the start time of the audio gap by the time stamp in the stream and A_GAP_LEN indicating the time length of the audio gap are recorded. Mark Tabl
In e, the number of Marks and a pointer to the mark are recorded. Each mark has a time code (Time_Code) indicated by a relative value from the beginning of the VOB, and a V to which the frame indicated by the mark belongs.
Relative time code in OBU VOB (VOBU_Time_Code)
And the relative address of this VOBU within the VOB (VOBU_Addres
s), the relative address (Frame_Address) from the beginning of this VOBU to the first data of this frame, and Delete_Permission (deletion is possible by default) indicating whether the mark can be deleted or not. The Clip Table records the number of Clips and pointers to the clips.
, End Mark, Played flag indicating whether this clip has been played one or more times (not played at the time of recording), Delete_Permission indicating whether the clip can be deleted (default: deleteable), and free space on the DVD-RAM When space is low,
An Auto_Delete flag (automatic deletion is impossible by default) indicating that deletion may be performed automatically is recorded.

Next, the AV Clip Sequence, which is the AV clip sequence information in the AV clip management information, will be described.

[0248] In the AV Clip Sequence, the number of Clip Sequences and a pointer to the Clip Sequence are recorded. Clip
Sequence contains the number of Entry Clips indicating the clips set in the clip sequence, a pointer to the Entry Clip, and Delete_permiss indicating whether the clip sequence can be deleted.
ion (default can be deleted) is recorded. Entry Clip
Specifies whether an AV File ID for identifying an AV file having actual AV data of a clip, a VOB ID for identifying a VOB, a Clip ID for identifying a Clip, and whether a connection with a previous Entry Clip is seamlessly reproduced. The Seamless Flag for discrimination is recorded.

(Creation of AV Clip Management Information) The AV clip management information created at the time of recording will be described.

The AV clip management information generation unit shown in FIG.
At the end of recording, the AV clip management information shown in FIG. 38 is created and written to a file. Next, A generated at the time of recording
The contents of the V clip management information will be described. Basically, in one recording, one VOB, two marks, one clip, and one clip sequence are generated. The two marks are the VOB video display start and end frames, and are relative values from the VOB head. One clip is one Clip with these two marks as Start Mark and End Mark, and one clip sequence is a Clip Sequence composed of this one clip. Each field is filled with a value as described in the format of the AV clip management information.

As described in the description of the AV data recording unit, when a pause is made, a mark is placed at the place where the pause was made. In addition, VOBs are temporarily stopped
Is divided, a mark, a clip, and a clip are created in VOB units as described above. However, only one clip file is created in one AV file as a series of recording units.

Next, a method of adding a mark, a clip, and a clip sequence will be described. Processing when the recording / editing / reproduction control unit of FIG. 24 receives a mark addition request will be described using an example. FIG. 38 is a diagram immediately after recording and before adding AV clip management information, and FIG. 39 is a diagram after adding AV clip management information.

First, a method of adding a mark will be described. FIG. 39 shows AV clip management information when a new mark 00: 00: 01: 00 is added. Mark to add is Ma
Given Mark ID # 3 not used in rk Table, Mar
k Added to Table.

Next, a method of adding a clip will be described. FIG. 39 shows a mark added at the beginning (Mark # 3 in FIG. 39) as a start mark and a mark at the end of the AV file (Mark in FIG. 39).
Shows AV clip management information when a new clip having Mark # 2) as an end mark is added. The clip to be added is given ID Clip # 2 not used in the Clip Table, and is added to the Clip Table.

Next, a method of adding a clip sequence will be described. FIG. 39 shows the clip added earlier (FIG. 39).
3 shows AV clip management information when a new clip sequence having Clip # 2) as one playback sequence is added. The clip sequence to be added is AV Clip Sequen
Given ID Clip_Sequence # 2 not used in ce,
Added to AV Clip Sequence. The added clip sequence is the clip added earlier (Clip # 2 in FIG. 39).
AV File I to which this clip belongs
D, VOB ID, Clip ID, and Seamless Flag are managed and recorded as Entry Clip.

Next, a method of deleting a mark, a clip, and a clip sequence will be described. The processing when the recording / editing / reproduction control unit of FIG. 24 receives a mark deletion request will be described using an example. Deletion is just the opposite of adding marks, clips, and clip sequences as described above. That is, FIG. 39 used in the aforementioned additional example
FIG. 38 is a diagram before the clip management information is deleted, and FIG. 38 is a diagram after the AV clip management information is deleted.

First, a method of deleting a clip sequence will be described. Clip Seq to delete
uence) for each entry clip (EntryClip) information
Delete from AV Clip Sequence. At the same time, the number of managed clip sequences (Num of Clip Sequences) is reduced by the number of deleted clip sequences.

Next, a method of deleting a clip will be described. Delete the clip (Clip) from the clip table (Cl
ip Table), and the number of managed clips (Num
of Clips) is reduced by the number of deleted clips. If there is a clip sequence including the deleted clip, the clip sequence is deleted.

Next, a method of deleting a mark will be described. Enter the mark to be deleted (Mark) in the mark table (Mark T
number) and the number of marks managed (Num of Mar
ks) is reduced by the number of deleted marks. If there is a clip including the deleted mark, the clip is deleted, and the clip sequence including the deleted clip is also deleted.

S when editing (Split of AV clip management information)
Update of the AV clip management information accompanying the plit will be described.

The trick-play information editing unit 123 in FIG.
The operation when a request is received will be described using an example.

FIG. 40 shows AV clip management information before Split, and FIG. 41 shows two pieces of AV clip management information after this AV clip management information is split at frame 00: 00: 02: 00.

Note that it does not matter to which AV clip the frame 00: 00: 02: 00 to be split belongs to the essence of the present invention. Therefore, in this example, the following description will be made assuming that the frame belongs to the second half AV clip. However, the handling of frames to be split is not limited.

First, the contents of the first half of the AV clip management information will be described. Since the first half of the AV file inherits the contents of the original AV file, AV_File_ID is
t Same as before. The mark of the section lost compared to the original AV file (Mark # 2 in FIG. 40) and the clip including this mark (Clip # 1 in FIG. 40) are deleted. Instead, a new mark (Mark # 2 in FIG. 41) indicating the video display end frame of the new AV file is created, and a mark indicating the video display start frame of the AV file (Mark in FIG. 41 (a)).
# 1) is the Start Mark, and the new mark (Mark # 2 in FIG. 41A) indicating the video display end frame of the AV file is End Ma
A new clip (Clip # 1 in FIG. 41A) is created as rk. In addition, since the latter half of the VOB is lost, the video display end time VOB_V_E_PTM of the VOB is changed to a new mark (FIG. 41A).
Is changed to "270270" which is the display end time of Mark # 2),
The input time LAST_SCR to the buffer of the last pack of the VOB is also changed to the value in the actual stream (in this example, "1"
60270 ").

Next, the contents of the second half of the AV clip management information will be described. AV_File_ of the second half of the AV file
A new name is recorded in the ID. The mark of the section lost compared to the original AV file (Mark # 1 in FIG. 40) and the clip including this mark (Clip # 1 in FIG. 40) are deleted. Instead, a new mark (Mark # 1 in FIG. 41B) is created in the video display start frame of the new AV file.
Also, the remaining marks (Mark # in FIG. 40)
2) is recorded at an earlier time by 00: 00: 02: 00 lost. The mark of the video display start frame of the AV file (Mark # 1 in FIG. 41B) is a Start Mark, and the mark of the video display end frame of the AV file (FIG. 41).
A new AV clip (Clip # 1 in FIG. 41 (b)) with the Mark # 2 of (b) as the End Mark is created. Also, since the first half of the VOB is lost, the VOB video display start time VOB_V_S_PTM
Is changed to “270270” which is the display start time of the new mark (Mark # 1 in FIG. 41B), and FIRST_SCR indicating the input time to the buffer of the VOB first pack is also changed to the value in the actual stream. (“170270” in this example).

Next, the AV which is the information of the clip sequence
_Clip_Sequence will be described. A clip sequence including a clip lost by Split (Clip # 1 in FIG. 40) is deleted. Instead, a new clip sequence for playing back a newly created clip (Clip # 1 in FIG. 40A for the first half AV file and Clip # 1 in FIG. 41B for the second half AV file) (FIG. 41 (c)
Clip_Sequence # 1 and Clip_Sequence # 2) are created.

In the case of Shorten at the time of editing, Split and
Since this operation is equivalent to the operation of deleting one piece of AV clip management information that can be performed by Split, the description in this example is omitted.

(Merge of AV Clip Management Information) M at Editing
Update of the AV clip management information accompanying the erge will be described.

The operation when the trick-play information editing unit in FIG. 24 receives a Merge request will be described using an example.

[0270] Fig. 42 shows two pieces of AV clip management information before Merge. FIG. 43 shows that two re-encoded VOBUs are added between the two AV clip management information to merge
This shows the AV clip management information after the clipping. A VOBU consisting of 12 frames is added after the front AV file, and a VOBU consisting of 18 frames is added before the rear AV file. Further, it is assumed that the two VOBUs are data encoded successively to the time stamp of the VOB (AV file) connected to each.

The contents of the merged AV clip management information will be described. A after merging the front AV file
Since the V file takes over, the AV File ID is
The name of the file is recorded. In the front AV file
Since a VOBU of 12 frames is added, a mark indicating the video display end frame of the AV file (Ma in FIG. 42A)
rk # 2) increases by 12 frames. Similarly, since a VOBU of 18 frames is added to the subsequent AV file, a mark indicating the video display end frame of the AV file (FIG. 42)
Mark # 2 in (b) increases by 18 frames.

Although the two AV files become one AV file by Merge, there is no correlation between the time attributes of the two AV files, so that a time stamp discontinuity occurs between the two AV files. Therefore, even if Merge is performed, each is managed as a different VOB. VOB ID of the AV file ahead
1. The VOB ID of the rear AV file is recorded as 2 in the AV clip management information.

Also, as described in the audio gap described above, an audio gap occurs between VOBs due to a difference in video and audio frame periods. As described in the audio gap, when the audio gap start time and the gap time are obtained based on the video display start time at the beginning of the VOB, the audio gap start time A_
STP_PTM = 216810, audio gap time length A_GAP_LEN =
It becomes 378.

Since the mark and the clip are completed in the VOB, only the VOBU added by Merge described above is changed. However, the front VOB has the rear 12 frames, and the rear VOB has the front VOB. VOBs increase by 18 frames each.
Video display end time VOB_V_E_PTM increases by 12 frames, and video display start time VOB_V_S_PTM
Reduced by 18 frames.

Also, since the packs for the increased frames are increased respectively, the LAST_SCR of the front VOB is changed to the last SCR of the increased pack, and the FIRST_SCR of the rear VOB is changed.
Is changed to the SCR at the beginning of the increased pack. Also,
If seamless playback (refer to presentation data) is possible between the two merged VOBs, Sea V
Change mless Flag to "YES".

[0276] Next, AV_Clip_Sequence, which is information on an AV clip sequence, will be described.

Entry constituting AV clip sequence
Clip is basically unchanged, but the AV file is Merge
The attribute of each Entry Clip is changed by this. En
Since try Clip # 1 is included in the preceding AV file, the attribute information is inherited as it is. Since Entry Clip # 2 is included in the subsequent AV file, the AV File ID as attribute information is changed to "System_2" and the VOB ID is changed to "2". Also, when seamless playback between two VOBs is enabled by Merge, the Seamless Flag in the Entry Clip indicating the backward VOB to be seamlessly played back is changed to "YES". The determination of the Seamless Flag is performed as follows.

A procedure for determining seamless playback between Entry Clips will be described with reference to FIG.

First, it is checked whether the AV File IDs of the Entry Clips match. If they match, the process proceeds to step 2, and if they do not match, the Seamless Flag is set to "NONE" (step 1).

Next, it is checked whether the VOB IDs of the Entry Clips match. If they match, the process proceeds to Step 3, and if they do not match, the process proceeds to Step 4 (Step 2).

Next, in step 2, the VOB ID is set between Entry Clips.
Will be described. En in the AV file
Check if try Clips are continuous, and if continuous, Seam
Set lessFlag to "YES", and if not continuous, Seamless Flag
To "NONE" (step 3).

Next, in step 2, the VOB ID is set between Entry Clips.
Will be described. In the AV file
It is checked whether the VOBs are continuous. If the VOBs are continuous, the process proceeds to step 5, and if not, the Seamless Flag is set to "NONE" (step 4).

Next, it is checked whether the preceding Entry Clip is the clip at the end of the VOB and the subsequent Entry Clip is the clip at the leading end of the VOB.
If this condition is satisfied, the Seamless Flag is set to "YES"; otherwise, the Seamless Flag is set to "NONE" (step 5).

Also, adding one VOBU to two AV files in Merge is the same as the above-mentioned example in which one VOBU has a size of 0, and merging two AV files directly In the above example, the size of both VOBUs is the same as 0, and one AV file contains one VOBU.
Is the same as that in the above-described example where one of the files does not exist and the size of the VOBU to be added to this empty file is 0. In the above example, one VOBU is added to one file. However, the number of VOBUs is not essentially limited to one.

The description of the logical format of the file recorded on the disc has been completed. Hereinafter, the operation of the DVD recorder will be described based on the configuration diagram of FIG.

[0286] 6. Operation of DVD Recorder: Each operation will be described together with a user interface and file operation.

6.1. Initial state When a new disk is loaded in the DVD recorder, the disk reading unit 101 waits for the rotation of the disk to stabilize in order to check the state of the disk. And that the stable reading state has been set.

The file system unit 102 sends volume information and file information to the disk reading unit 101,
Instructs reading of AV clip management information. The read volume information, file information, and AV clip management information are stored in the memory of the file system unit 102,
The data in this memory is referred to for subsequent file operations. These pieces of information are held in the memory in the file system unit 102 until an instruction to remove the disc is issued.

The recording / editing / playback control unit 105 graphics the state of the AV file recorded in the disc from the read AV clip management information to the user IF unit 106, and the result is shown in FIG. Request display as initial screen. If nothing is recorded on the disc, only the cursor arrow is displayed.

In the initial screen shown in FIG.
By operating the buttons on the remote control shown in FIG. 5, the state shifts to the respective states of reproduction, editing, and recording.

6.2. Recording When the user presses the recording button on the remote controller shown in FIG. 45 on the initial screen in FIG. 46, the screen shifts to the screen 201 in FIG. In this embodiment, the recording conditions that can be set are two points: recording time and recording quality.

First, when "1" and "select" are pressed, the focus is on the designation of the recording time, the rectangle is moved to "unlimited" or "specified time" by the cursor button of the remote control button in FIG. The condition is set by pressing the button.

When "specified time" is selected, the screen is switched to a screen for inputting the time by using the ten-key buttons. When the setting is completed, the screen returns to the screen 201 in FIG. The “recording quality” of the recording condition controls the bit rate and resolution of AV data, and controls the bit rate of MPEG data generated by the AV data input unit 111 in FIG.

FIG. 65 shows a control method for high image quality, standard, and time priority. It is assumed that, after shifting to 202 in FIG. 64, the “record” button on the remote control is pressed.

Before starting the recording process, the recording
The edit / playback control unit 105 requests “CREATE” from the commands in FIG. 1 to the file system unit 102,
If the file can be created, the file system returns a file identifier. At this time, the maximum file size of the disc is taken as the file size according to the recording conditions. Recording
The editing / playback control unit 105 passes the file identifier and the time-priority parameters shown in FIG. 65, which are recording conditions, to the AV data recording unit 110, and instructs to start recording.

The AV data recording unit 110 clears the contents of the track buffer 1.5 MB, which is an internal buffer memory, so that AV data can be input.

The AV data input unit 111
The encoding of the signal is started. The bit sequence resulting from the encoding is constantly recorded in the track buffer. In the AV file management information generation unit 112, the AV data input unit detects the GOP boundary identifier to be inserted into the AV data and the boundary of the I-picture at the head of the GOP, and converts the bit amount counted until the detection into the number of sectors. And store it in the internal memory.

[0298] In the AV clip management information generation unit 114,
At the start of the recording of the AV data, a new data area of the AV clip management information is created in the memory, the start point of the video data is marked and stored in the memory. At this time, the other areas remain at the initial values (or undefined).

When the track buffer is occupied to a predetermined capacity, the overflow countermeasure unit 113
Reads the contents of the track buffer and requests the AV file system unit 103 to write data to the disk using the “AV-WRITE” command of FIG. Since the bit rate recorded on the disc through the file system is higher than the bit rate generated by the AV data input unit 111, the data occupancy in the track buffer is reduced as shown in the buffering model of FIG. In parallel with the processing in the AV data recording unit, the AV data requested to be recorded through the AV file system unit 103 is recorded on the disk by the disk recording unit 100. The recording method at this time will be described with reference to FIG.

FIG. 7 shows the AV format explained in FIG.
This shows the current state of the block management table.
AV block management table address (upper) 0, (lower)
0, 1, and 2 indicate that they are used in advance for non-AV files (information) such as volume information, file management information, AV file management information file, and AV clip management information file. Such information has not yet been recorded on the actual disc. On the other hand, the AV data for which recording has been started starts recording from the AV block of address (upper) 1 and (lower) 0. This is a block indicated by hatching in FIG. Since the recording starts from the initial state of the disc, after the recording is continuously performed in the successive AV block,
The zone is shifted to zone 2 at the boundary between zone 1 and zone 2. Thereafter, recording is continuously performed in zone 2 and zone 3 and the zone, and the contents of the corresponding AV block management table also change.
Here, since the AV block management table itself is also stored in the memory of the file system unit 102 in FIG. 24, only the AV data is written on the disk, so that there is no useless seek or rotation wait.

FIG. 8 shows the state of the file information in the file system information. In the DVD-RAM file system, a file is managed as a link list of a plurality of extents. One Extent corresponds to a continuous recording area on the disc. As described above, since the sector buffer exists at the zone boundary, continuous recording cannot be performed at the zone boundary, and a new extent is generated.

At the zone boundary, a delay of several hundred ms occurs.
However, since all sectors in the AV block perform continuous recording, the track buffer in the AV data recording unit 110 shown in FIG. 24 does not overflow, and continuous data input is performed.

An operation in the case where the recording of AV data as described above makes it impossible to record more data than is already buffered on the disc will be described. In the AV data recording unit in FIG. 24, the AV data input unit 111 stops the encoding process. Since valid data boundaries in MPEG must be GOP boundaries,
The GOP boundary immediately before the interruption is the final point of the data.

After recording starts, the AV file management information generation unit 112 detects all GOP boundaries and the I-picture boundary at the head of the GOP, and shows the number of sectors of each GOP and the number of sectors of the I-picture at the head of the GOP in FIG. It is held in the form of the array shown. These pieces of information are transferred to the control data management unit 107 and used for generating an AV file management information file. Further, the overflow countermeasure unit 113 transfers the AV data up to the last GOP boundary as valid data to the AV file system unit 103, and all the recording processes in the AV data recording unit 110 are completed.

At the end of recording, the AV clip management information generation section 114 marks the end of the AV data,
It is written to the memory as V clip management information. This A
The V clip management information is transferred to the control data management unit 107, and the AV clip management information file AV Clip part, A
Used for generating and updating V Clip Sequence.

The control data management unit 107 converts the AV file management information file and the AV clip management information file based on the GOP information shown in FIG. 58 and the AV clip management information shown in FIG.
4 on the disk. Here, from the start point to the end point of the recorded AV data recorded in the AV clip management information is defined as one clip (the mark position is the start point and the end point), and one AV clip string is automatically generated. If the AV clip management information AV ClipSequence file is a new disc, a new clip is created.
Update the clip management information to the AV Clip Sequence. At this time, the new clip includes the Seamless Informati
Non Seamless is set as on.

Prior to the disk recording, volume information stored in the file system unit 102 and file information such as Extent shown in FIG. 8 are also recorded on the disk. Even after the recording of such information is completed, the volume information and the file information remain in the memory in the file system and are used for the next recording and editing.

When recording is performed in a state other than the initial state, when the recording button on the remote controller in FIG. 45 is pressed, an initial screen for recording appears, and recording can be performed in the same procedure as described above.

6.3. Temporary Editing Next, the editing procedure will be described.

[0310] The editing in this embodiment is hierarchical editing composed of temporary editing using AV clips and actual editing for actually processing (re-editing) an MPEG stream. Here, the temporary editing will be described.

(Mark Setting) First, a procedure for setting a mark in an AV file will be described.

One clip sequence is selected from the initial state shown in FIG. 46 and reproduced. The details of the reproduction will be described later.

FIG. 47 is a diagram showing the flow of mark setting. The user presses the mark button on the remote controller on the favorite screen while watching the playback screen. With this operation, a mark request is sent from the remote controller to the recording / editing / playback control unit through the user IF unit. Recording / editing / playback control unit is A
The AV data reproducing unit requests the V data reproducing unit for an AV File ID which is an identifier of the AV file being reproduced and a time stamp (a relative value in the AV file).
Return the AVFile ID and time stamp to the playback control unit.
The recording / editing / playback control unit sends a mark request to the AV data editing unit with the AV file ID being played back and a time stamp. The AV clip sequence editing unit of the AV data editing unit specifies the VOB being reproduced from the received AV file ID and time stamp using the information of the AV Clip part in the AV clip management information, and starts displaying the video of the VOB. The time VOB_V_S_PTM is compared with the received time stamp, and the relative time code in the VOB of the frame with the specified mark is specified. The AV clip sequence editing unit adds the requested mark information to the AV clip management information according to the mark addition procedure of “Creating AV clip management information”. This AV clip management information is sent to the control data management unit after the reproduction of the AV data is completed, and is recorded on the disc through the non-AV file system unit.

Next, a mark setting procedure at the time of recording will be described. Recording is performed from the initial state shown in FIG. 46 according to the above-described recording procedure.

FIG. 48 is a diagram showing a flow of mark setting at the time of recording. The user presses a mark button on the remote control on a favorite screen while watching the recording screen. With this operation, a mark request is sent from the remote controller to the recording / editing / playback control unit through the user IF unit. The recording / editing / playback control unit requests the AV data recording unit for an AV File ID, which is an identifier of the AV file being recorded, and a time stamp (relative value in the AV file). -Return the AVFile ID and time stamp to the editing / playback control unit. The recording / editing / playback control unit creates a temporary mark table including an AV File ID, a temporary mark ID, and a time stamp, and registers the temporary mark and the time stamp in the temporary mark table. Thereafter, every time there is a mark request, the same processing is performed, and registration in the temporary mark table is performed. The recording / editing / reproduction control unit sends a mark request to the AV data editing unit with an AV file ID and a time stamp for each provisional mark registered in the provisional mark table simultaneously with the end of recording. The subsequent processing is the same as the above-described mark setting method at the time of reproduction.

Next, an automatic mark setting method at the time of recording will be described. When a scene change is detected by the AV data recording unit, the AV data recording unit automatically sends an AV File ID and a time stamp to the recording / editing / playback control unit.
The recording / editing / reproduction control unit automatically registers the AV File ID and the time stamp sent from the AV data recording unit in the above-described temporary mark table. Thereafter, processing similar to the above-described mark setting procedure at the time of recording is performed to set a mark.

Next, an automatic mark setting method at the time of reproduction will be described. When a scene change is detected by the AV data reproducing unit, the AV data reproducing unit automatically sends an AV File ID and a time stamp to the recording / editing / reproduction control unit.
The recording / editing / playback control unit sends the AV File ID and the time stamp sent from the AV data playback unit to the AV data editing unit together with the mark request. Thereafter, processing similar to the above-described mark setting procedure at the time of reproduction is performed to set a mark.

(Clip Setting) Next, a procedure for setting a clip in an AV file will be described.

FIG. 49 is a diagram showing the flow of clip setting. When the temporary edit button on the remote controller is pressed in the initial state shown in FIG. 46, the recording / editing / playback control unit reads the AV clip management information from the control data management unit, and
Section to display the clip setting screen. More specifically, as shown in FIG. 49, a recording date, recording time, recorded channel of AV data which is a recording of the recording, a simple graph showing the recorded data on a time axis, and a registration in the AV data. Mark and clip, a clip sequence, and a cursor for selecting one mark or clip are displayed. The user presses up / down / left / right keys on the remote control to move the mark or clip to be focused. The user uses the select key to select two marks with the remote control. By this operation, a clip request is sent to the recording / editing / playback control unit through the user IF unit so that the section between the two marks becomes one clip. The recording / editing / playback control unit is provided with an AV data editing unit,
A clip request is sent together with the selected two marks and the AV file ID and VOB ID to which the two marks belong. A
The AV clip sequence editing unit of the V data editing unit adds the requested clip information to the AV clip management information according to the clip adding procedure of “Creating AV clip management information”. This AV clip management information is sent to the control data management unit, and is recorded on the disc through the non-AV file system unit.

[0320] Further, according to the movement of the cursor, it is possible to display the frame of the mark pointed by the cursor or the image after the mark in the background. The details are the same as the reproduction procedure described in the reproduction of the clip sequence.

(Clip Sequence Setting) Next, a procedure for setting a clip sequence will be described.

FIG. 50 is a diagram showing the flow of clip sequence setting. When the temporary edit button on the remote controller is pressed in the initial state shown in FIG. 46 similarly to the clip setting, the recording / editing / playback control unit reads the AV clip management information from the control data management unit and sends the clip sequence through the user IF unit. Display the setting screen. Specifically, FIG.
The recording date, recording time, recorded channel, registered marks, clips, a cursor for selecting one mark or clip, and a clip sequence are displayed as shown in FIG. Is done. The user presses up / down / left / right keys on the remote control to move the mark or clip to be focused. The user
Use the select key to select multiple clips using the remote control according to the playback order of the clips. Through this series of operations, a clip sequence request is sent to the recording / editing / playback control unit via the user IF unit so that the selected clip sequence becomes one clip sequence. The recording / editing / playback control unit sends a clip sequence request to the AV data editing unit together with the selected clip sequence, the AV file ID to which each clip belongs, and the VOB ID. The AV clip sequence editing unit of the AV data editing unit adds the requested clip sequence information to the AV clip management information according to the clip sequence addition procedure of “Creating AV clip management information”. This AV clip management information is sent to the control data management unit, and is recorded on the disc through the non-AV file system unit.

In addition, it is also possible to display a clip indicated by the cursor in the background according to the movement of the cursor. The details are the same as the reproduction procedure described in the reproduction of the clip sequence.

(Clip sequence deletion) Next, mark,
A procedure for deleting a clip and a clip sequence will be described.

FIG. 51 is a diagram showing the flow of mark, clip, and clip sequence deletion.

When the temporary edit button on the remote control is pressed in the initial state shown in FIG. 51, the recording / editing / playback control unit reads out the AV clip management information from the control data management unit, similarly to the setting of the clip and the clip sequence. A screen for deleting marks, clips, and clip sequences is displayed through the user IF unit. Specifically, as shown in FIG.
A recording date, recording time, recorded channel, registered mark, clip, clip sequence, and a cursor for selecting a mark or a clip or a clip sequence are displayed. The user presses the up / down / left / right keys of the remote control, moves the cursor to the mark or clip or clip sequence to be deleted, and presses the delete key. This operation allows the user to record / record through the user interface to delete the selected mark or clip or clip sequence.
A deletion request is sent to the edit / playback control unit. Recording / editing /
The playback control unit sends a deletion request to the AV data editing unit together with the selected mark or clip or clip sequence, and in the case of a mark or clip, the AV file ID and VOB ID to which each mark or clip belongs. A
The AV clip sequence editing unit of the V data editing unit deletes the requested deletion information from the AV clip management information according to the procedure of “deletion of AV clip management information”. This AV clip management information is sent to the control data section and recorded on the disc through the non-AV file system section.

6.4. AV File Editing (Main Editing) Next, the AV file editing (main editing) will be described.
In the AV file editing (main editing), one clip sequence is selected from a plurality of clip sequences created in the above-described temporary editing, and the MPEG stream is actually processed (edited).

(Setting of AV File Editing) Next, the procedure of setting the actual editing of the AV data logically provisionally edited by the clip sequence will be described.

FIG. 52 is a diagram showing the flow of editing settings. When the main edit button on the remote controller is pressed in the initial state shown in FIG. 46, the recording / editing / playback control unit reads AV clip management information from the control data management unit, and performs a clip sequence setting screen for editing through the user IF. Is displayed. Specifically, as shown in FIG. 52, a cursor for selecting one clip sequence, a recording date, a recording time, a recorded channel, a mark, a clip, a clip sequence, and a clip sequence of the AV data which is the recording of the recording. Is displayed. The user presses the up, down, left, and right keys of the remote control to move the clip sequence to be focused. The user selects one favorite clip sequence from a plurality of clip sequences using the selection key.
By this operation, an editing request is sent to the recording / editing / playback control unit via the user IF unit so that the selected clip sequence is actually edited. The recording / editing / playback control unit sends an editing request to the AV data editing unit with an ID indicating the selected clip sequence. The AV clip editing unit of the AV data editing unit edits the requested editing information according to an editing procedure of an AV file, AV file management information, and AV clip management information described later. The edited AV file is recorded on the disc through the AV file system unit, the AV file management information and the AV clip management information are sent to the control data management unit, and are recorded on the disc through the non-AV file system unit.

(Operation of Editing AV File) Next, an editing procedure of an AV file, AV file management information and AV clip management information will be described.

FIG. 53 is a diagram showing the flow of editing.
FIG. 53 is a diagram showing an example in which clip sequence # 3 is selected from the clip sequences shown in FIG. 52 and edited. It should be noted that the AV file editing of the present invention is realized by efficiently combining clip extraction and clip connection.

First, each clip used for editing is extracted from the AV file. Since the AV file "System_1" has two clips to be used, an AV file system is prepared so that each of them can be extracted.
Using the Split function, the AV file "System_1" is split into two AV files. Similarly, AV file management information, A
The V clip management information is also divided into two AV files using the prepared Split function (step 1).

Next, the AV data including the start mark and the end mark at both ends of each clip is read.
As mentioned in the description of multiplexing, video and audio data are multiplexed at a maximum of 1 second apart due to the nature of MPEG, so to read both video and audio data at both ends of each clip, the video indicated by the mark One or more VOBUs containing one second of video data before and after the VOBU including the frame (hereinafter referred to as "VOBU group")
Need to be read. First, the AV file to which the specified clip belongs is opened. Next, the time code of the start mark and the VOBU in the AV file management information
Compares with the Start Time in the Map, and includes V for 1 second before and after the video frame indicated by the start mark.
Specify OBU group. Next, the start sector and the end sector of the VOBU group to be read are specified using Start Sector information in the VOBU Map, SEEK of the AV file is performed until the VOBU group start sector, and the AV file is read to the VOBU group end sector.
The same processing is performed for the end mark, and the VOBU group is read. For the VOBU group read as described above,
Analysis of the read VOBUs determines whether seamless playback between B is possible. According to the procedure described in the description of the re-encoding method, video re-encoding and creation of an audio gap (only during seamless playback) or audio data Cut out (only during non-seamless playback) and multiplex video and audio data. VO to be read
The method of determining the BU group is not limited to the above, and A
The V data may be analyzed to read only the minimum necessary AV data, or vice versa (for example, fixed to 10 VOBUs) (step 2).

Next, unnecessary portions are removed from each AV file, and an AV file containing only clips is created. Step 2
Since the VOBU group including the clip cutting plane is read and re-encoded and multiplexed, the VOBU group including the read VOBU group is discarded as unnecessary data. To delete data from the AV file, use the Shorten function provided in the AV file system. Similarly, AV
The management information is changed using the Shorten function, which also provides the file management information and the AV clip management information (step 3).

[0335] Finally, the clips are connected in the order of the playback sequence, and merged into one AV file. First, the first clip (clip (3) in FIG. 53) and the second clip (clip (4) in FIG. 53) are combined into one AV.
Merge into a file. At this time, the VOBU group <32 '> read in step 2 and re-encoded (step 2
VOBUs <32> re-encoded from VOBUs <32> and VOBUs <41 '> (VOBUs <41> read in step 2)
VOBU group). Two clips and two VOBU groups are merged into one AV file by using the Merge function prepared in the AV file system. Similarly, the AV file management information and the AV clip management information are also stored in one AV file management information, A
It is merged with the V clip management information. Similarly, the merged AV file, the third clip (clip (5) in FIG. 53), and the VOBU group <42 '> (read in step 2)
VOBU group <42> re-encoded from VOBU group <42>) and VOBU group <51 '
> (The VOBUs re-encoded from the VOBUs <51> read in step 2) are merged. Finally, VOBU group <3 at the beginning
1 '> (a VOBU group re-encoded from the VOBU group <31> read in step 2) and finally a VOBU group <52'> (a VOBU group re-encoded from the VOBU group <52> read in step 2) Are connected by the Merge function (step 4).

By the above processing, the clip sequence logically provisionally edited can be edited into one AV file with the actual situation. In this editing, M
By not changing the time stamp of the PEG stream and minimizing the movement of the MPEG stream recorded on the optical disc, a short time (several seconds ) Can be edited.

Next, the hierarchical editing method in this embodiment will be described with reference to FIG. The hierarchical editing method according to this embodiment includes the following three steps. 1) Temporary editing step: Editing by mark, clip and clip sequence 2) Confirmation step: Confirmation of temporary editing result 3) Main editing step: Processing (editing) of MPEG stream First, the "temporary editing step" will be described. Temporary editing is roughly divided into two steps. The first half is a mark setting step, and the second half is a clip and clip sequence setting step. First of all, mark setting is MPEG
It comprises a mark requesting step for receiving a mark request from a user during reproduction of a stream, a mark setting step for setting a mark when there is a mark request, and a reproduction end determining step for determining reproduction end. The latter half includes a clip request step for receiving a clip request from the user, a clip setting step for setting a clip when there is a clip request, a clip sequence request step for receiving a clip sequence request from the user, and a clip sequence request. In this case, the sequence includes a clip sequence setting step of setting a clip sequence and a clip sequence determining step of determining the end of clip editing.

Next, the “confirmation step” will be described. It comprises a clip sequence selection step for receiving a selection of a clip sequence to be confirmed (reproduced) from the user, and a reproduction step for reproducing the clip sequence selected by the user.

Next, the “main editing step” will be described. The sequence includes a clip sequence selection step of receiving a selection of a clip sequence to be subjected to main editing from a user, and an MPEG stream processing (editing) step of processing an MPEG stream so that the selected clip sequence can be seamlessly reproduced. .

The present embodiment is characterized in that the DVD-RAM is not merely a recording medium for AV data, but is a recording medium for AV data and provisionally edited data for realizing an unprecedented hierarchical editing. Another feature is that all the steps in editing are realized on one DVD-RAM.

6.5. Playback (Playback of Clip Sequence) Next, the operation for playing back the AV data file edited as described above will be described. The user IF unit 106 in FIG. 24 displays all clip sequences existing on the disc on the screen shown in FIG. 46 based on the AV clip management information obtained from the recording / editing / playback control unit 105. In that screen, the cursor points to one clip sequence. When the user operates the “up arrow” button or the “down arrow” button on the remote control, the user IF unit 106 moves the cursor pointing to the clip sequence. When the user presses the "play" button on the remote control, the playback operation starts. To stop playback, press the "STOP" button on the remote control.

When the user presses the "play" button on the remote controller, the user IF unit 106 requests the recording / editing / playback control unit 105 to pass a clip sequence number as an argument to perform a playback process. When the recording / editing / playback control unit 105 is instructed to play back, the recording / editing / playback control unit 105 transfers the clip sequence passed as an argument to the AV data playback unit 130 and requests playback.

[0343] The operation when the AV data reproducing unit 130 reproduces the clip sequence passed as an argument will be described.

First, the AV clip sequence reproducing unit 13 shown in FIG.
2 receives the AV clip management information from the control data management unit 107 and searches for a clip sequence of the corresponding number. Next, using the AV clip management information, information on the start and end marks of each clip in the clip sequence is created, and the information is passed as an argument to request the AV file reading unit 133 to reproduce. At the same time, seamless information between each clip is passed as an argument, and reproduction is requested from the AV decoder unit.

[0345] First, the AV file reading unit 133
Requests the transfer of AV file management information to the control data management unit 107, refers to the obtained AV file management information, and refers to the first sector of the VOBU corresponding to the leading mark of each clip passed as an argument. The address and the end sector address of the VOBU corresponding to the end mark are obtained.

Here, the operation for obtaining the start or end sector address of the VOBU from the designated time of the mark will be described. Referring to Start Time of VOB Information, search for a VOB that includes the specified time. Time map table of obtained VOB
See ((Time Unit) * (Index of Time Map) + (Tim
e Base)) indicates the time pointed by each Time Map, so the Time Map index that points before and after the specified time of the mark is obtained. Suppose that index is i. Time Map # i
VOBU pointed to by VOBU1 and VOBU pointed by Time Map # (i + 1) to VOBU2
Assume that The desired VOBU exists between VOBU1 and VOBU2. By correcting the Start Time of the Time Map, the head times of VOBU1 and VOBU2 can be obtained. Ti
By referring to the VOBU Map Index and VOBU Map Table of me Map, the VOBU existing between VOBU1 and VOBU2 can be obtained, and by correcting the Start Time of VOBU Map,
The start time of those VOBUs can be obtained. By comparing the time of the beginning of each VOBU obtained above with the time of the mark, a VOBU including the time of the mark can be obtained. Start Sector of VOBU Time Map or VOBUMap
By correcting the Start Sector of the VOBU, the AV
The sector address from the beginning of the file can be obtained. In the VOBU Map Table, the VOBU next to the obtained VOBU
In the same manner, the end address of the required VOBU can be obtained.

Next, the AV file reading unit 133
Requests the file system unit 102 to “OPEN” the AV file indicated by the AV File ID of the first clip, and then sends the VO corresponding to the mark at the tip of the clip.
Request "SEEK" up to the first sector address of the BU. Then, in order to read data of about 32 KB at a time from the AV file into the track buffer in the AV data reproducing unit 130, a "READ" is requested to the file system unit 102. Until the last sector of the VOBU corresponding to the mark at the end of the clip is read,
D "repeatedly. However, when the track buffer becomes full, a "READ" request is not made, and the system waits for a free space in the track buffer, and resumes the "READ" request when there is a free space in the track buffer. And
After the reading of the data corresponding to the clip is completed, an end mark indicating the break of the AV data is written to the track buffer.

The AV file reading unit 133 repeats the same operation for the second and subsequent clips in order, and continues to write data corresponding to all clips to the track buffer.

When the reproduction is requested, the AV decoder unit 131 waits while monitoring the track buffer until the track buffer becomes full. The track buffer is
When it becomes full, start the decoding operation. If the end mark exists in the track buffer and the connection to the next clip is a non-seamless connection, decoding is temporarily stopped up to the end mark data. Then, when the track buffer becomes full again, the decoding operation is started. When the end mark exists in the track buffer and the connection to the next clip is a seamless connection, the seamless reproduction operation described above is performed.

Next, the operation of the file system and the disk reading section in the continuous reading of the file will be described. It is assumed that an AV file to be read is recorded in an AV block as shown in FIG. AV_B
5.5MB for all AV blocks except LK # i and AV_BLK # l
It is assumed that the above data is recorded.

When the file system unit 102 receives the “READ” request, it searches the file management information resident in the memory for Extent information of the file based on the file identifier. The address information of the head AV block is extracted from Extent # 1 of the AV file, and data is read from the corresponding AV block. The AV decoding unit 131 and the A after receiving the AV data from the file system unit 102
The operation of V data reproducing section 130 is as described above. Assume that data of 5.5 MBB or more has been recorded in AV_BLK # i. In this case, decoding is started while reading AV_BLK # i. The file system unit 102 compares the sector from which data is being read with the last sector of AV_BLK # i, and if they are equal, jumps to the next Extent # 2. At this time, in the track buffer, data of a size capable of continuously supplying data to the AV decoding unit 131 during a jump is accumulated. This is AV_B
AV data is continuously recorded in the sector in LK # i, and data can be fully accumulated in the track buffer while the data is continuously read. When the data is full in the track buffer, the data can be supplied to the AV decoder unit 131 for about 1.5 seconds. The value of 1.5 seconds is the time required for jumping from the outermost circumference to the innermost circumference in a consumer device, and can guarantee the worst case. Even when the zones are different between Extents, the time required to read data by performing a jump across the zone boundary is several hundred ms, so that the continuity of data is sufficiently ensured.

(Fast-forward playback) An operation when fast-forward is instructed during playback of a clip sequence will be described. When the “fast forward” button is pressed on the remote controller in FIG. 45 during the reproduction of the clip sequence, the user IF unit 10 in FIG.
6 requests the recording / editing / playback control unit 105 to perform fast-forward playback processing, and the recording / editing / playback control unit 105 requests the AV clip sequence playback unit 132 to perform fast-forward playback processing.

[0353] The AV clip string reproducing section 132 requests the AV decoder section 131 to stop decoding. The AV decoder unit returns information on the clip and the time at the time of the stop to the AV clip sequence reproducing unit 132. Next, after clearing the track buffer, the AV clip sequence reproducing unit 132 passes information regarding the clip and time at the stop as arguments to the AV file reading unit 133, and
Request fast forward playback processing. At the same time, the AV decoder unit 1
A request for a fast-forward playback process is made to 31.

The operation when the AV file reading unit 133 is requested to perform the fast forward reproduction process will be described. First, a VOBU including a time five seconds ahead of the time at the stoppage passed as an argument is obtained. The operation for obtaining the VOBU is the same as the operation for obtaining the VOBU including the time of the mark described at the time of reproduction. The AV file reading unit 133 obtains the obtained sector address in the head AV file of the VOBU and the relative end address of the reference picture from the VOBU head. It is determined whether or not the data of the reference picture straddles the discontinuous boundary of the AV block by requesting “IN_AV_BLK_BOUND” to the file system unit 102. When the VOBU straddles the discontinuous boundary of the AV block, the next adjacent VOBU is obtained, and the sector address in the head AV file of the VOBU and the relative end address of the reference picture from the VOBU head are obtained. Next, the AV file reading unit 133 performs the same operation as reading the data of the clip during reproduction,
The VOBU reference picture data is read from the AV file and written to the track buffer. After the data writing is completed, the AV file reading unit 133 writes an end mark indicating a break in the AV data in the track buffer. Next, the same operation is sequentially performed every 5 seconds, that is, for VOBUs 10 seconds ahead, 15 seconds ahead,... If the specified time is beyond the end of the clip, the same operation is sequentially performed on the first VOBU of the next clip every five seconds from the VOBU of the next clip.

Note that the size of the VOBU is 8 Mb for the bit rate.
Even if ps is 1.2 seconds, it is 1.2 Mbytes, which is smaller than the size of the AV block. Therefore, neither two adjacent VOBUs cross the AV block boundary.

In the above description, the VOBU 5 seconds ahead is obtained, but by changing this number of seconds, the speed of fast-forward playback can be changed.

[0357] The operation when the AV decoder 133 is requested to perform the fast-forward reproduction process will be described. When the track buffer becomes full or the end mark is written out to the track buffer, the decoding operation is started. Then, decoding is stopped at the end mark. This operation is repeated.

(Fast rewind playback) During playback of an AV file,
The operation when the fast-forward playback is instructed is the same as the fast-forward playback described above, except that the operation for finding the previous VOBU every 5 seconds is changed to the operation for finding the VOBU returned every 5 seconds. Do the same thing.

(Stop during playback) If a stop is performed during playback, the clip being played is divided into two at the stop point, and a flag indicating that the divided front clip has been played (ie, Played).

(Simple Reproduction Apparatus) A logical format of the AV clip management information shown in FIG. 37 is shown.

As compared with the AV clip management information in the first embodiment, information indicating the head address of the VOBU and the time code is added to Mark. Therefore, clip playback is possible even without the VOBU information of the AV file management information, so that even a playback device with a small main storage capacity can play back the clip without storing the AV file management information in the main storage.

(Embodiment 2) Next, Embodiment 2 of the present invention
DVD that displays menus using thumbnails
The recorder will be described.

Since the structure, operation, and disc format of the DVD recorder of the second embodiment are basically the same as those of the DVD recorder of the first embodiment, only the parts different from the first embodiment will be described.

[0364] 1. System Configuration The system configuration of the DVD recorder will be described with reference to FIG. The system configuration of the DVD recorder is described in the first embodiment.
Therefore, functions that are not included in the first embodiment will be described.

(User IF Unit) The user IF unit 106
In addition to the function described in the first embodiment, a thumbnail registered for each clip is displayed on the screen. The handling of the thumbnail images will be described in detail in presentation data and navigation data described later.

(Recording / Editing / Playback Control Unit)
The reproduction control unit 105 controls the reproduction of the thumbnail in addition to the functions described in the first embodiment.

(AV Data Recording Unit) AV Data Recording Unit 1
Numeral 10 creates a thumbnail in addition to the functions described in the first embodiment. The AV data input unit 111 stores, in addition to the functions described in the first embodiment, the first frame at the start of recording (recording start / pause release) in a built-in frame memory and stops recording (stop / pause). 2), the image data stored in the frame memory is reduced to the image size of the thumbnail, MPEG video compression and conversion into an MPEG program stream are performed, and are recorded in the thumbnail file through the AV file system unit 103. Further, the AV clip management information generation unit 114 sends the created thumbnail data to the control data management unit 107 and records the thumbnail data in the AV clip management information.

(AV data editing unit) AV data editing unit 1
Reference numeral 20 creates a thumbnail in addition to the functions described in the first embodiment. The AV clip editing unit 122 reads a desired MPEG program stream through the AV file system unit 103, extracts the MPEG video data, reduces the uncompressed image subjected to video decoding to the thumbnail image size, and performs MPEG video compression of the reduced image. Conversion to an MPEG program stream is performed and recorded in a thumbnail file through the AV file system unit 103. Further, the AV clip editing unit 122 sends the created thumbnail data to the control data management unit 107 and records the thumbnail data in the AV clip management information.

(AV data reproducing unit) AV data reproducing unit 1
Numeral 30 decodes thumbnails in addition to the functions described in the first embodiment. The AV decoder unit 131 extracts an MPEG program stream for thumbnails through the AV file system unit 103, and extracts MPEG video data and performs video decoding. The decoded uncompressed thumbnail image data is stored in the recording / editing / playback control unit 1
05 is displayed on the screen by the user IF unit 106.

[0370] 2. Presentation Data Here, thumbnail data (image data) will be described.

(Directory Structure) FIG. 69 shows a structure of directories and files required in the DVD recorder.

Basically, the configuration is the same as that of the first embodiment. However, in the second embodiment, a file containing thumbnail image data (thumbnail file) is added.

As described in the first embodiment, the correspondence between the thumbnail file and other AV files, AV file management information, and AV clip management information is determined by the body name excluding the extension. Since the extension of the thumbnail file is “.sml”, when searching for the corresponding thumbnail file from the AV clip management information, the file having the extension “.sml” may be searched for in the own body name.

(Image Data) The image data format of the thumbnail will be described.

FIG. 70 is a diagram showing a thumbnail image data format. As shown in the figure, the thumbnail image data is composed of an MPEG program stream in which I pictures of an MPEG video reduced to a predetermined image size are packed (packets). The predetermined size to be reduced is 176 × 112 (thumbnail 1) in the case of NTSC.
And 96x64 (thumbnail 2). In the case of PAL,
176 x 144 (thumbnail 1) and 96 x 80 (thumbnail 2). Images of these two sizes are created and stored in the thumbnail file in order.

Also, the size of one thumbnail in the completed MPEG program stream is determined by the ECC block (32
KB). In this embodiment, 224 KB, which is the maximum size of an I picture of MPEG video, is packed (= 2048
B) The smallest common multiple of 228KB and 32KB, which is 256KB
And

[0377] Unless a special encoding process is performed, the size of image data compressed by MPEG generally does not become a fixed value. Therefore, when the data size is less than the above 256 KB, the data amount is adjusted by padding (stuffing).

An advantage of fixing the data size is that unnecessary thumbnails can be erased by overwriting other thumbnails. This eliminates the need to delete unwanted thumbnails each time,
Unnecessary file operations can be eliminated. The merit of setting the data size to an integral multiple of the ECC block is that data reading / writing becomes faster. Note that the thumbnail file is recorded as an AV file, and the thumbnail file is recorded from the head of the ECC block.

[0379] 3. The management data for managing the navigation data thumbnail will be described.

FIG. 71 shows the data structure of the AV clip management information. All the fields except the hatched portions in the figure are the same as the data structure of the AV clip management information in the first embodiment, and thus the description is omitted.

The portion shown by the hatched portion in FIG. 71 is the management information for the thumbnail, and indicates “Thumbnail1 Address / T” indicating the address of the thumbnail (sector size from the head of the file).
humbnail2 Address ”and“ Thumbnail1 Size / Thumbnail2 Size ”that indicates the data size of the thumbnail.
This is the value after MPEG video encoding and packing (packeting). In the explanation of the presentation data, the size of the thumbnail image was fixed, so "Thum
The “bnail Size” is a fixed 256KB.

[0382] 4. Operation of DVD Recorder The operation of the DVD recorder is the same as that of the first embodiment except for the thumbnail processing. In the second embodiment, description will be made mainly on handling of thumbnails.

4.1. Initial State The initial state of the DVD recorder is the same as that of the first embodiment except for the thumbnail display.

(Thumbnail Display) FIG. 72 is a diagram showing an initial state. The recording / editing / playback control unit 105 passes the thumbnail image data to the user IF unit 106,
As shown in FIG. 72, a thumbnail corresponding to each clip is displayed on the screen.

The display of a thumbnail image will be described with reference to FIG. The recording / editing / reproduction control unit 105 first extracts a thumbnail image. To extract a thumbnail image, a thumbnail file is first derived from the body name of the AV file to which the clip belongs.

Next, the recording / editing / playback control unit 105
The thumbnail address “Thumbnail1 Address / Thumbnail2 Address” and the size “Thumbnail1 Size / Thumbnail2 Size” corresponding to the clip are read from the AV file management information recorded in the control data management unit 107. Although either the thumbnail 1 or the thumbnail 2 may be used, in the present embodiment, the description will be made using the thumbnail 1.

The recording / editing / playback control unit 105 issues a request to read thumbnail data to the AV file system unit 103, using the thumbnail file and the address and size of the thumbnail 1 to be read as arguments. The AV file system unit 103 reads the MPEG program stream containing the thumbnails through the disc reading unit 101, and converts the read MPEG program stream into an AV file.
The data is sent to the data reproducing unit 130

In the AV data reproducing section 130, the AV decoder section 131 extracts an MPEG video stream from the MPEG program stream, performs video decoding, and sends an uncompressed thumbnail image to the recording / editing / reproduction control section 105.

[0389] The recording / editing / playback control unit 105 sends the uncompressed thumbnail image received from the AV data playback unit 130 to the user IF unit 106 and displays it on the screen.

[0390] 4.2. Recording The recording operation of the DVD recorder is the same as that of the first embodiment except for handling of thumbnail images.

As described in the description of the AV data recording unit,
The V data input unit 111 has a frame memory for one still image data, and stores the first video data (1
Frame). During the recording by the DVD recorder, the still image data is continuously buffered, and after the recording is stopped (stop or pause), the process of creating the thumbnail data is started.

FIG. 74 is a diagram showing a thumbnail creation procedure after the recording is stopped (stop or pause).

[0393] The AV data input unit 111 extracts the buffered still image data after the end of recording, performs image reduction and MPEG video encoding (one still image), and then performs system encoding into an MPEG program stream. I do. At this time, since no audio data exists, the completed MPEG program stream is an MPEG program stream of only still images without audio. Also, thumbnail images generate two types of data having different sizes.

The AV data recording unit 110 records the MPEG program stream containing the thumbnail data in A
An MPEG program stream containing a request to the V file system unit 103 and containing the thumbnail data is recorded on a disk through the disk recording unit 100.

[0395] The AV clip management information generation unit 114 creates thumbnail management information of the clip.

The AV clip management information generation unit 114 extracts the address (= A) and size (= S) of the immediately preceding thumbnail from the control data management unit 107. Using the extracted value, thumbnail management information in the AV clip management information (AV Clip part) is obtained as follows.

Thumbnail1 Address = A + S Thumbnail1 Size = 256 KB Thumbnail2 Address = A + S + 256 KB Thumbnail2 Size = 256 KB However, in the case of the first clip of the AV file, the previous thumbnail does not exist. In this case, the calculation is performed with A = 0 and S = 0.

The AV clip management information generation unit 114 uses the thumbnail management information obtained as described above in the first embodiment.
To the control data management unit 107 in the same manner as

4.3. Temporary Editing The temporary editing operation of the DVD recorder is the same as that of the first embodiment except for the handling of thumbnail images.

As described in the data structure of the thumbnail,
Since a thumbnail is data associated with a clip, thumbnail creation during temporary editing is performed simultaneously with clip creation.

The procedure for creating a thumbnail will be described with reference to FIG. The AV clip editing unit 122 extracts mark information (VOBU address, time code, VOBU time code, etc.) at the beginning of the clip and thumbnail information at the end of the thumbnail file (thumbnail address, thumbnail size) from the control data management unit 107.

[0402] The AV clip editing unit 122 obtains a frame indicated by the mark from the time code and VOBU time code of the mark information, and passes the VOBU data (from the VOBU address of the mark information) through the AV file system unit 103 until the frame data is obtained. MPEG program stream) and continue decoding.

[0403] Next, the AV clip editing unit 122 reduces the read frame data in accordance with the thumbnail size (thumbnail size 1 and thumbnail size 2). The data is recorded in the thumbnail file through the system unit 103.

[0404] Finally, the AV clip editing unit 122 obtains thumbnail management information and sends it to the control data management unit 107. The method of obtaining the thumbnail management information is the same as the method described for recording.

[0405] It should be noted that the thumbnails during recording were created by providing the AV data input unit 111 with a frame memory to buffer video frames at the start of recording.
Similar effects can be obtained by creating thumbnail data from recorded moving image data (MPEG program stream) in the same way as creating thumbnails during temporary editing.

Also, as shown in FIG. 76, the AV data input section 111 has a thumbnail library of data collection for thumbnails, and the user selects one from this data collection at the time of recording, and the same effect is obtained as a thumbnail. can get.

Also, as shown in FIG. 77, the AV clip sequence editing unit 121 has a thumbnail library of data collection for thumbnails, and the user selects one of the data collections at the time of temporary editing, and the same applies to thumbnails. The effect is obtained.

[0408] The same effect can be obtained even if the existing thumbnail data is shared and used instead of always creating the thumbnail data at the time of provisional editing.

[0409] Also, it has been described that the recording of the thumbnail image is additionally recorded in the file. However, the same effect can be obtained by performing overwriting on the thumbnail data that has become unnecessary due to the deletion of the clip.

[0410] Although the data size of the thumbnail is recorded as the thumbnail information, the same effect can be obtained even if the data size information is omitted because the data size of the thumbnail is fixed.

Although the start address of the thumbnail data is recorded as the thumbnail information, the same effect can be obtained by recording not the start address but the number of the sequentially attached thumbnail data because the data size of the thumbnail is fixed. Is obtained.

[0412] Although the data size of the thumbnail is set to a fixed value, the same effect can be obtained even if the length is variable since the data size information and the address information of the thumbnail are provided.

Also, the image size (resolution) of the thumbnail
Is fixed to two types, but the image size can be changed to a variable length,
The same effect can be obtained even if one or more image sizes are recorded as thumbnails.

[0414]

According to the present invention, in a rewritable optical disk medium in which moving image data and management information of the moving image data are recorded, for each partial reproduction section of the moving image data recorded on the optical disk medium, a partial reproduction section is provided. Thumbnail data indicating the contents and management information having pointer information to the thumbnails are recorded. In the reproducing apparatus for reproducing the optical disk medium, by reading the thumbnail data from the optical disk medium and displaying the thumbnail data on the screen, it is easy to select and reproduce the moving image data desired by the user from the large-capacity moving image data. Is obtained.

[0415] Further, by setting the size of the thumbnail data to a fixed value, an effect is obtained that new thumbnail data can be overwritten on unnecessary thumbnail data.

[0416] Also, the size of the thumbnail data is read from and written to the optical disc medium as an ECC unit.
By setting the integral multiple of the block, the effect of speeding up the writing and reading of the thumbnail data can be obtained.

[0417] In the recording apparatus for recording moving image data and management information on the optical disk medium, a frame memory for temporarily storing still image data related to moving image data to be recorded is provided, so that moving image data can be recorded. At the start, the first one frame is buffered in the frame memory as still image data, and after the recording of the moving image data is completed, the still image data stored in the frame memory is taken out and the thumbnail data converted into the MPEG program stream is recorded. This makes it possible to create a thumbnail image at the time of recording.

The recording apparatus for recording moving image data and management information on the optical disk medium has a still image library in which a plurality of still image data are stored in advance, so that the contents of the partial creation section of the moving image data are indicated. When a thumbnail is created, one piece of still image data is selected from the still images stored in the still image library and set as a thumbnail, so that a thumbnail indicating the content of the partial playback section of the moving image data is created. The effect that it can be easily created is obtained.

[0419] Also, in the recording method for recording moving image data and management information on the optical disk medium, a step of extracting only one frame from the moving image data recorded on the optical disk medium as still image data; , The step of recording the thumbnail data on the optical disk medium, the step of creating pointer information to the thumbnail data on the optical disk medium, and the step of recording the pointer information. As a result, it is possible to create thumbnail data indicating the contents of the partial playback section of the moving image data and create management information.

[0420] In the recording method for recording moving image data and management information on the optical disk medium, when a new partial playback section is defined from the moving image data, a playback section that overlaps the partial playback section already exists. In this case, the step of reading pointer information to the thumbnail data of the overlapping partial playback section from the management information of the overlapping partial playback section and the step of recording the pointer information as pointer information to the thumbnail of this partial playback section With this configuration, an effect equivalent to that of creating thumbnail data can be obtained without actually creating thumbnail data indicating the content of a partial playback section of moving image data created at the time of editing.

[Brief description of the drawings]

FIG. 1 is a diagram showing a list of functions of a file system unit.

FIG. 2 is a configuration diagram of the entire disk.

FIG. 3 is a configuration diagram of an AV block.

FIG. 4 is a configuration diagram of an AV block management table.

FIG. 5 is a diagram showing a buffering model when writing AV data;

FIG. 6 is a diagram showing a buffering model when reading AV data;

FIG. 7 is a diagram showing a state of an AV block management table immediately after starting recording from an initial state;

FIG. 8 is a diagram showing a state of file information when recording is started from an initial state;

FIG. 9 is a diagram showing allocation of AV data to AV blocks and sequential reading.

FIG. 10 is a configuration diagram of a sector bit map.

FIG. 11 is a configuration diagram of an AV block management table.

FIG. 12 is a diagram showing a sector bitmap process based on an AV block attribute change.

FIG. 13 is a diagram showing processing of an AV file “DELETE” command;

FIG. 14 is a diagram showing processing of a “DELETE” command for a non-AV file.

FIG. 15 is a diagram showing processing of a “SHORTEN” command;

FIG. 16 is a diagram showing handling of an AV block in a “SPLIT” command;

FIG. 17 is a diagram showing handling of an AV block in a “MERGE” command;

FIG. 18 is a diagram showing handling of an AV block in a “MERGE” command;

FIG. 19 is a diagram showing processing of an “IN_AV_BLK_BOUND” command;

FIG. 20 is an explanatory diagram of an AV block discontinuous boundary.

FIG. 21 is a diagram showing handling of an AV block in a “MERGE” command;

FIG. 22 is a diagram showing the handling of AV blocks in the “MERGE” command

FIG. 23 is a diagram showing the handling of AV blocks in the “MERGE” command.

FIG. 24 is a configuration diagram of a DVD recorder.

FIG. 25 is a configuration diagram of an AV data input unit.

FIG. 26 is a diagram showing an example of an AV clip and an AV clip sequence.

FIG. 27 is a diagram showing a relationship between an AV clip and AV data.

FIG. 28 is a configuration diagram of an AV decoder model.

FIG. 29 is a diagram showing an example of a file configuration for a DVD recorder.

FIG. 30 is an explanatory diagram of a video picture type.

FIG. 31 is an explanatory diagram of multiplexing.

FIG. 32 shows an example of an AV file composed of three VOBs.

FIG. 33 is a configuration diagram of an extended STD model.

FIG. 34 is a diagram showing an example of an audio gap.

FIG. 35 is an explanatory diagram of a video picture structure.

FIG. 36 is an explanatory diagram of buffer control, audio gap and audio multiplexing boundary.

FIG. 37 is a configuration diagram of AV clip management information.

FIG. 38 shows an example of AV clip management information after recording.

FIG. 39 is a diagram showing an example of adding AV clip management information.

FIG. 40 is a diagram showing an example of AV clip management information before Split.

FIG. 41 is a diagram showing an example of AV clip management information after Split

FIG. 42 is a diagram showing an example of AV clip management information before Merge.

FIG. 43 is a diagram showing an example of AV clip management information after Merge

FIG. 44 is a flowchart of seamless flag determination.

FIG. 45 illustrates an example of a remote controller.

FIG. 46 is a diagram showing an example of an initial screen.

FIG. 47 is a diagram showing an example of mark setting during reproduction.

FIG. 48 is a diagram showing an example of mark setting during recording.

FIG. 49 shows an example of clip setting.

FIG. 50 is a diagram showing an example of clip sequence setting.

FIG. 51 is a diagram showing an example of deleting a mark, a clip, and a clip sequence.

FIG. 52 is a diagram showing an example of main editing

FIG. 53 is a diagram showing an example of operation steps in main editing

FIG. 54 is a flowchart of a hierarchical editing method.

FIG. 55 is a configuration diagram of AV file management information.

FIG. 56 is a diagram showing the relationship between Time Map and VOBU Map

FIG. 57 is a diagram of stream information.

FIG. 58 is a diagram showing GOP information generated during recording;

FIG. 59 is a diagram showing an example of AV file management information after recording

FIG. 60 is a diagram showing an example of AV file management information before Split.

FIG. 61 is a diagram showing an example of AV file management information after Split.

FIG. 62 is a diagram showing an example of AV file management information before Merge.

FIG. 63 is a diagram showing an example of AV file management information after Merge.

FIG. 64 is a diagram showing an example of a display at the time of initial recording and recording;

FIG. 65 is a diagram showing an example of correspondence between recording conditions and image quality selection;

FIG. 66 is a diagram showing an example of stream information.

FIG. 67 is a diagram showing an example of stream information before Merge.

FIG. 68 is a diagram showing an example of stream information after Merge.

FIG. 69 is a diagram showing an example of a file configuration for a DVD recorder including a thumbnail file.

FIG. 70 is a diagram showing an example of creating thumbnail data.

FIG. 71 is a configuration diagram of AV clip management information including thumbnail management information.

FIG. 72 is a diagram showing an example of an initial screen using thumbnails;

FIG. 73 shows an example of thumbnail display.

FIG. 74 is a diagram showing an example of thumbnail creation at the time of recording;

FIG. 75 is a diagram showing an example of thumbnail creation at the time of editing

FIG. 76 is a diagram showing an example of thumbnail creation at the time of recording using a thumbnail library.

FIG. 77 is a diagram showing an example of thumbnail creation at the time of editing using a thumbnail library;

[Explanation of symbols]

 REFERENCE SIGNS LIST 100 disc recording section 101 disc reading section 102 file system section 103 AV file system section 104 non-AV file system section 105 recording / editing / playback control section 106 user IF section 107 control data management section 110 AV data recording section 111 AV data input section 112 AV file management information generation unit 113 Overflow prevention unit 114 AV clip management information generation unit 120 AV data editing unit 121 AV clip sequence editing unit 122 AV clip editing unit 123 AV file management information editing unit 130 AV data reproduction unit 131 AV decoder unit 132 AV clip string reproducing unit 133 AV file reading unit 134 Read error countermeasure unit 150 AV data input 151 DeMUX 152 Video decoder 153 Audio decoder 15 Seamless connection processing unit 155 at video output 156 audio output 200 recording selection screen 1 201 during recording selection screen 2 202 during recording selection screen 3

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Kazuhiro Tsuga 1006 Kadoma Kadoma, Osaka Prefecture Matsushita Electric Industrial Co., Ltd.

Claims (7)

[Claims] 1. A rewritable optical disk medium on which at least moving image data and management information of the moving image data are recorded, wherein each partial reproduction section of the moving image data recorded on the optical disk medium has a partial reproduction section. An optical disk medium, wherein a thumbnail indicating the contents of the thumbnail and management information having pointer information to the thumbnail are recorded. 2. The optical disk medium according to claim 1, wherein the data capacity of the thumbnail is a fixed value. 3. The optical disk medium according to claim 2, wherein the data capacity of the thumbnail is an integral multiple of an ECC block which is a unit of reading and writing to the optical disk medium. 4. A recording apparatus for recording moving image data and management information on an optical disk medium according to claim 1, wherein said apparatus temporarily stores still image data related to moving image data to be recorded. An optical disk recording device having a frame memory. 5. A recording apparatus for recording moving image data and management information on an optical disk medium according to claim 1, wherein a plurality of thumbnails indicating the contents of a partial playback section of the moving image data are provided. An optical disk medium recording device having a still image library storing still image data. 6. A recording method for recording moving image data and management information on an optical disk medium according to claim 1, wherein only one frame from the moving image data recorded on the optical disk medium is stopped. Extracting as still image data; converting the still image data into thumbnail data; recording the thumbnail data on the optical disc medium; and creating pointer information to the thumbnail data on the optical disc medium. Recording the pointer information. 7. A recording method for recording moving image data and management information on an optical disk medium according to claim 1, wherein, when a new partial playback section is defined from moving image data, Reading out pointer information to the thumbnail data of the overlapping partial playback section from the management information of the overlapping partial playback section when a partial playback section overlapping the playback section already exists; Recording the information as pointer information to a thumbnail of a partial playback section.