JP2868981B2 - Apparatus and method for generating compressed signal and apparatus and method for reproducing - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention uses a writable / readable magnetic disk, optical disk, or CD-ROM as a recording medium, and generates / reproduces a compressed signal effective when the signal to be handled is an encoded compressed signal. / Synchronization and management device.

[0002]

2. Description of the Related Art A recording medium such as a magnetic disk or an optical disk has a smaller recordable data capacity than a magnetic tape, but can perform high-speed data access. Can be easily performed. Further, with the recent advance in the high-efficiency compression encoding technology for image data, it is considered that a large number of programs can be stored, and the use of the recording / reproducing apparatus using the disk as a recording medium is considered to increase. As such a moving image compression recording method, for example, ISO-11172
(MPEG).

Hereinafter, a conventional disk system will be described. The format of a disc usually includes a data area and a management area indicating what information is recorded in each part of the data area (FIG. 12A).

The data area is composed of clusters, and the size of one cluster is determined and fixed in the range of 1 to 64 sectors, and the size of one sector is 128 bytes to 8 bytes.
It is fixed within the range of 192 bytes.

[0005] The management area is provided on the inner circumference side of the disk, and a directory table (FIG. 12) is provided at the innermost circumference.
(B)), there is a file allocation table (FIG. 12C) on the outer periphery. The directory table stores a program name and a leading cluster. When the name of the program to be reproduced is searched in the directory table, the leading cluster can be found. When the head cluster is known, the head cluster number is searched in the file allocation table. When the first cluster number is searched, the next (2) cluster number is recorded there, so that the next (2)
The (second) cluster number is known. 2
When the second cluster number is known, the second cluster number is searched next, and the third is recorded in the file allocation table.
In this way, the cluster numbers to be reproduced are known one after another, and when the cluster number reaches the last cluster number, the end information forms a pair.

Therefore, when a disc is reproduced, a desired program name is searched in the directory table of the management area, and a leading cluster number is recognized.
Next, cluster numbers to be reproduced are read one after another in the file allocation table (FAT).

The capacity of the management area can be expressed as follows. Directory capacity = number of recorded programs x (program name + top cluster) [bytes] FAT capacity = number of programs x (1 program size / cluster size) x 1 cluster representation size on FAT [bytes] By the way, many programs are stored on disk. For recording, it is necessary to secure a large data area and reduce the management area. However, recording many programs requires a large capacity of the management area. Therefore, from the above relationship, the cluster size may be increased to reduce the capacity of the management area. However, if the cluster size is increased, the length of one cluster is fixed (a fixed number of sectors). Therefore, when the end of one program ends with a small number of sectors from the beginning of a certain cluster, the data area is There is waste.

[0008] Such waste is likely to occur particularly in a system employing the moving image compression technique described above. In the moving image compression technology, the physical signal length is uncertain because an inter-frame compression technology using data of temporally adjacent frames and a variable-length coding technology are used. Therefore, if the length of one cluster is fixed, there is a high possibility that the data area will be wasted. Further, when reproducing a signal using this moving image compression technique, there is no problem in reproducing sequentially from the beginning, but it is difficult to realize functions such as special reproduction and high-speed search.

[0009]

As described above, in the conventional disk recording / reproducing method, when recording a signal which has been encoded by a moving image compression technique or has been subjected to a variable length, it is easy for data areas to be wasted. There's a problem. Further, it is difficult to realize functions such as special reproduction of a signal recorded using a moving image compression technique and high-speed search. In the case of a video signal, an audio signal is accompanied. In the case of a signal using a moving image compression technique, it is necessary to synchronize video and audio.

In order to achieve the above object, the present invention provides a main video compression means for compressing and encoding main video data which can be separated in one frame unit, and a plurality of main video data which can be selectively superimposed on the main video. Sub-video data
A sub-picture compression means for compressing and encoding independently of the main picture compression means for compressing the main picture data; and a sub-picture compression means for forming the first packet from the encoded main picture data from the main picture compression means. A formatter for combining the encoded sub-picture data from the sub-picture compression means to form a second packet, and outputting the first and second packets to a recording or transmission system. It is provided.

[0011]

According to a first aspect of the present invention, video data is grouped so as to be separated into at least a fixed number of reproduction times or a predetermined number of video frames corresponding to the number of reproduction times. Video grouping and video compression means for compressing and encoding the obtained video data in group units, and audio grouping and audio compression means for grouping and compressing and encoding the corresponding audio data for each group unit of the video data. The encoded video data from the video compression unit is connected for a plurality of groups, and the encoded audio data from the audio compression unit is connected for a plurality of groups, and these groups are combined and output as a data unit. And a formatter for outputting to a transmission system.

According to the present invention, as a second means, a storage means for managing data of a program recorded on a recording medium is formed as a data allocation table, and each unit of the table has a unit on the disk. A section is provided for storing each information of a track number, a zone number belonging to a track, a sector number on the track, and a link pointer of a data unit to be reproduced next as one set.

According to a third aspect of the present invention, on the encoder side, a fixed number of video frames corresponding to a predetermined time length of the original video are encoded, and a plurality of encoded images are collected. A video grouping / compressing means for converting data into one video packet, and a voice group comprising encoded voice data corresponding to the packetized coded video data as a plurality of voice frames and forming one voice packet Generating and adding an audio frame number and an audio sample number in the audio packet relating to the original audio corresponding to the head of the specific video frame in the encoded video data as one video packet; Means for creating additional data as data, the additional data,
A formatter for creating such data units one after another as one data unit by connecting audio packets and video packets, and on the decoder side, coded video data, An output that decodes audio data and additional data, and operates when the output timing of the decoded specific video frame matches the audio frame number included in the encoded audio data and the audio frame number included in the additional data. And timing setting means.

[0014]

According to the first means, video frames are grouped by a predetermined number, encoded and compressed in each group, and a plurality of groups of compressed video data are included in the data unit. Not. For this reason, each data unit can be treated independently and a video signal can be decoded. Therefore, even if efficient disk recording is performed by compression, reproduction and decoding can be performed for each data unit. Second
According to the means, since the data unit has the track number, the zone number, and the sector number, even if the data length of the data unit recorded as described above is not fixed but variable, it can be easily managed. Specifically, the data recording efficiency of the disc can be improved. Further, according to the third means, the synchronization between the audio and the video is accurately obtained, and the synchronization state can be constantly monitored.

[0015]

Embodiments of the present invention will be described below with reference to the drawings. First, a moving image compression format according to the present invention will be described. When encoding video data, first, a plurality of groups of pictures (GOPs) are collectively packetized, and audio data (for about 1.0 second) and extension data corresponding to the packets are encoded and compressed. A data unit is added to the video data. The GOP is fixed in the same program, and a time code for audio synchronization is arranged in the first subcode in the extended data of the data unit.

FIG. 1 shows encoded data (see FIG. 1).
(A) and an example of an output image (FIG. 1 (B)) obtained by decoding the decoded image is schematically shown. In the figure, I is intra-frame coded video data, P is forward predictive coded image data, B is bidirectional predictive coded image data. In this mode, I, P, B, P , B are repeated. Therefore, the encoded data length of each frame is different. According to such a format, if only I is reproduced, the speed is 6 times, and if I and P are reproduced, 2 times.
Double-speed images can be obtained. The actual double speed is limited by the speed of reading data from the disk. This format is suitable for a high transfer rate, a large recording capacity, and quasi-random access. In this example, six frames are treated as one GOP as shown in FIG. Then, 5 GOPs become one packet. This packet corresponds to a reproduction time of 1.0 second. However, the actual recording signal length on the disk differs depending on the packet because it is encoded by the moving image compression technique.

Therefore, one packet is equivalent to 30 frames (=
5 GOP × 6 frames / GOP), and the audio data for each 30 frames is 48 KByte (= 4 ch × 1).
2 KByte / s). When the number of simultaneously used channels is 2, the required minimum memory capacity is 24 KBy.
te is fine.

The main data to be recorded on the disk for each data unit and each information rate are as follows. Extended data = 128 Kbit / s = 16 KByte / s Audio data = 384 Kbit / s = 48 KByte / s Video data = 4096 Kbit / s = 512 KByte
The / s extension data includes a subcode and sub-picture data. The sub-picture data can be used for subtitle information and the like used in a movie. The subcode is individual management information in the data unit, and includes audio and video synchronization information. The sub-picture data is updated on a GOP basis including the corresponding main picture, and the synchronization and synchronization between the picture and the audio are also performed on a GOP basis.

Regarding subtitle information, a plurality of channels may be prepared as sub-image data so that two types of sub-images can be selectively output, such as an English-language scenario in a Western movie and a Japanese subtitle. Sub-image data allocation rate is 64K
In the case of bit / s, the recording time of one packet is 1.
If it is 0 second, the buffer capacity for holding the sub-picture data is about 64 Kbit. However, when the sub-picture has two channels, the necessary buffer memory capacity may be 32 kbit.

The above-described encoding of video, audio, and extension data is all completed within a data unit, and is completely independent of other data units. Next, a management area described later is secured on the disk. Reading is performed for each data unit based on the information of the management area. To be processed independently for each data unit,
Editing and accessing each data unit is easy.

The relationship between the data area and the management associated therewith will be described. In an actual arrangement, byte alignment processing is performed for each GOP, and secura alignment processing is always performed for each data unit, so that data units can be easily separated. The reduction rate of the actual recording capacity due to the sector alignment is as follows. Data unit configuration is screen display frame rate 30 frames / sec, GOP
When the number of constituent pictures is 6 (frames) and the number of GOPs in one data unit is 5 GOPs, sector alignment occurs for each data equivalent to about 1.0 second. The capacity decreases. When the total recording capacity of the disk is 346752 sectors, the capacity reduction rate is 0.2% (1 KB).
/ Sector).

At the time of reproduction, the video starts from the decoding of the first frame (I picture) of the GOP. The audio starts with decoding of the audio frame specified in the audio-video synchronization. When the decoding of the specified audio frame and the decoding of the first frame of the video GOP are both completed, the output of the video and the specified audio sample is started simultaneously.

As audio data, encoded audio data for about 1.0 second is added in the data unit. However, in the audio encoding, a fixed number of samples is taken as one block, the ends of adjacent blocks are slightly folded, then encoded in units of this number of samples, and a header is added thereto to create one encoded audio frame.

The audio frame length is equal to or less than the length of 2048 samples of the original audio, and is 24 ms to 36
ms. The encoded data amount of the audio frame is 288
Bytes to 576 bytes. A frame ID is attached to the header of every audio frame for each audio channel. The frame ID is 24 bits, with 4 bits representing the audio channel and 20 bits representing the audio frame number. The encoded audio data for about 1.0 second usually has a length of several tens of audio frames, depending on the number of samples in one block and the sampling frequency. The audio-video synchronization added to the subcode includes the frame number of the coded audio to which the decoded audio sample to be output belongs in synchronization with the timing of starting the output of the first frame of the corresponding GOP, and the audio sample number in that frame. Is specified. The time code is 32 bits, and the audio frame number is represented by 20 bits, and the audio sample number is designated by the remaining 12 bits. As a result, the maximum error of audio and video synchronization in the entire system is equal to one half of the audio sampling period, and the maximum video and audio synchronization error is about 16 μs when fs = 32 KHz.

FIGS. 2 and 3 show other examples of the moving picture compression format. Next, a system for managing a plurality of programs encoded and recorded as described above will be described. First, a management area is provided on the disc, and a management table is recorded therein.

FIG. 4A shows a management table position in the management area and an example of a zone arrangement in the data area. The management table includes an innermost volume identity field (VID), an outermost program information field (PIF), and an outermost data unit allocation table (DA).
T). The VID is written from the first byte of the management table area, and uses 256 bytes to indicate specification information of the entire disk. For example, information on a general recording disk, a reproduction-only disk, and the like (FIG. 5A).

The program information field (PIF) records the specification information of each program. For example, 16 bytes are used for each program.
FIG. 5B shows an example of the contents of 16 bytes of the PIF.

ATMB is the absolute time of the current program start point in the volume. (In the case of a time code search, first, each ATMB data is checked in the order of program reproduction to detect a program number in which a desired time code exists. Next, each DAT (described later) in the corresponding program is checked, and a program time (PTMB : Later) and AT
A search can be performed by comparing the added MB with a desired time code value and detecting a DAT to which the corresponding time code belongs.) If the method is based on the absolute start time, the user can know the absolute start time from the desired program number. Therefore, by searching the ATMB corresponding to the absolute start time, a specific PI can be obtained.
F data can be detected.

PINF indicates a program attribute.
As the program attribute, the attribute is described in units of a program, and a copy prohibition flag (CPNH), a program type (PTYPE), and a write attribute (PWR)
T), number of GOPs constituting data unit (SGDU)
There is. If CPNH is "1", copy is prohibited, if "0", copy is permitted, and PTYPE uses 3 bits to indicate home video, movie, music, karaoke, computer graphics, interactive, games, computer data, programs, etc. Indicates the type. PW
If RT is "1", it indicates that writing is possible. SGDU indicates one of Mode 1, Mode 2, and Mode 3 described above using three bits.

The PIF also stores other parameters as shown in FIG. AINF is the identification of the audio encoding system, VINF is the identification of the video encoding system,
ATRT is a picture attribute, ie, aspect ratio, PA
HRES, information for identifying a system such as L, NTSC, etc.
Is the screen horizontal resolution, and VRES is the screen vertical resolution.

PNTB is a start pointer,
D where the data unit of the program start point is stored
This is a pointer value indicating the AT address (data unit number). DAT will be described next.
By knowing the address (data unit number), the position of the head sector of the program in the data area can be recognized.

The PGML indicates a program number to be reproduced following the current program when there is a related program or continuously. That is, the order in which the programs are reproduced does not always match the order of the program numbers. When the current program is the last program, there is no link destination, and all bits of PGML are "1".

FIG. 5C shows the structure of the DAT. In this table, parameters include a zone number (NZON), a sector number (NSCT), a track number (NTRC), a program time (PTMB), and a link pointer (PNTL).

NZON is the zone number to which the recording sector at the head of the data unit belongs. The zone number is continuously assigned to each track from the reference position. That is, as shown in the data area of FIG.
There is one, and the numbers are sequentially numbered from 0 from this position. One track is composed of many zones. NSCT is
When the zone is determined, the sector number in the zone is indicated. The sector number is not a serial number related to another track or zone, but a number completed within that zone. NTRC indicates the track number where the zone and the sector number (the head of the data unit) exist.
Further, PTMB is a flag indicating temporal position information of the video data (I picture) at the head of the data unit, and the content is relative elapsed time (second) from the program start point.
It is. This temporal position information is used when the time code search described above is performed. The time position information is taken into the reproducing apparatus when displaying the program time, the absolute time, the remaining amount, and the like, and is used as start reference data.

The next PNTL is a flag for indicating the next DAT unit number temporally continuous with the current DAT unit number. The unit corresponds to the data unit number, and when there is no link destination at the end point of the program or the like, all bits are “1” (= 0 × FFFF). Valid values for the link pointer are 0x0000 to 0xFFFF
It is.

Returning to FIG. FIG.
An example of an AT is shown. DAT unit number is 0 to N
Continuous at max. The first DAT unit number is determined by referring to the PNTB of the PIF. Now, DA
Assuming that the T unit number is 1, the next link pointer is 0. The link pointer of DAT unit number 0 is Nmax-1. And the DAT unit number Nmax-
The link pointer of 1 is 2. Where DAT
Looking at the zone number, track number, and sector number according to the transition of the unit number, the sector 3 of the zone 1 of the track 4, the sector 2 of the zone 0 of the track 7, and the track 10
, The reproduction order information of zone 3 and sector 30 can be obtained.

FIG. 6A shows an example of the address arrangement of the management table shown in FIG. 5A and an example of the DAT address arrangement. FIG. 6B shows another example of the arrangement of addresses in the management table, in which an unused area is set between VID, PIF, and DAT. In this case, VID
There is an address offset when searching for PIF data from the data of
MPU for drive control included in some data of
Is recognized when executing the address management program.

Next, the capacity of the above-mentioned management table is estimated and examined. The capacity for holding the management table depends on the number of programs and the number of data units recorded on the disk. When the total number of programs is 256 and the number of data units is 7200 (1 second / unit, equivalent to 2 hours),
The total data of the management table is 256+ (16 × 256)
+ (8 × 7200) = 61952 bytes.

That is, in a system in which one data unit corresponds to about one second, by allocating a memory of 63 KB to the data management table, it is possible to handle two hours of management data, and this capacity is sufficient for practical use. It is.

The physical location of the start sector in the management table is usually ZONE = 0, TRACK = 0, SECTOR
= 0, but may be overwritten in different areas as a spare from the viewpoint of data protection. Since the management table is often referred to, reading the data on the disk every time slows down the access operation. Therefore, the management table may be first mapped to the work RAM of the drive control MPU. However, if the table capacity is too large, the memory cost may not be compatible with the product cost.If the configuration of the table itself is not appropriate, a large amount of calculation may be required each time to convert to a desired parameter value. It is preferable to set the method appropriately according to the product cost and the table capacity.

FIG. 7 shows an encoder according to the present invention,
2 shows a block configuration of a decoder. Input terminal 100
Is supplied with the original signal. This original signal is supplied to the separating means 10.
1 and its audio data, video data, extended data such as subtitles, a synchronization signal, and the like are separated. The audio data is input to the audio grouping means 102, the video data is input to the video grouping means 103, and the extended data is input to the extended data grouping means 104. The synchronization signal is
It is input to the first system control means 110. For example, if the mode 1 is designated, the first system control means 110 controls the video grouping means 103 so that video data is grouped every six frames, and also groups the audio data in time units. The extended data grouping means 102 is controlled so that the extended data for the corresponding frame is grouped. The grouped video data is
The video data is input to the video compression unit 106, and is encoded and compressed as described with reference to FIG. The grouped audio data is also compressed by the audio compression unit 105, and the grouped extended data is also compressed by the extended data compression unit 107. The output of each compression means 105106, 107 is input to the formatter 108. Here, in the case of mode 1,
Five GOPs (encoded video data) are collected, and the corresponding encoded audio data, encoded extension data, and subcode (additional data) are added to the data unit as shown in FIG. Is output as When encoding is performed by each compression unit, the generated code amount is controlled so that the data amount is an integral multiple of the sector capacity.

The signal output from the formatter 108 is recorded on a recording medium or sent to a transmission system. The signal fetched from the recording medium or the transmission system is supplied to the separation unit 121.
In, processing is performed for each data unit, and coded audio data, coded video data, coded extended data, and subcode are extracted from the data unit. The encoded audio data is decoded by the audio decoder 122, the encoded video data is decoded by the video decoder 123, and the encoded extension data is decoded by the extension data decoder 124. The decoded extended data is combined with the decoded video data by the combining unit 125. This allows the original audio signal,
The video signal will be reproduced. Information included in the subcode is input to the second system control means 126,
It is used as a reference for generating a timing signal for each block, video and audio synchronization, or mode setting information.

In this system, a means for synchronizing audio and video is devised. Next, the data unit will be described again. As described above, for video data, one packet is equivalent to 30 frames (= 5 GOP × 6).
Frame / GOP), and the audio data of 30 frames / GOP corresponds to 48 KBytes (= 4 ch × 12 KBytes).
e / s) (mode 1). When the number of simultaneously used channels is 2, the required minimum memory capacity is 2
4 KBit may be used.

FIG. 8 shows encoded video data, encoded audio data, and additional data included in the data unit. The audio is encoded in a unit of a fixed number of samples as one block, and is encoded in units of the number of samples. The header includes a frame ID for frame identification.

Next, the subcode includes additional information. The additional information includes data indicating the correspondence between the coded video data and the coded audio data. That is,
Figure 8 (A) has the coded video data to the video frame number, also the audio frame number is also present in the encoded audio data (FIG. 8 (B)). Therefore, it is assumed that the first frame of each GOP (data subjected to intra-frame compression processing: this is a specific video) is now special 1, special 2,..., And the audio frame numbers of the encoded audio data are special. 1 for k-1
Corresponding to correspond to k + 6 in Japanese 2, assuming that corresponds to the k + n in Japanese 5, the relevant information is inserted in the additional data (FIG. 8 (C)). In addition, additional data includes
The sample numbers respectively corresponding to the features 2,... Are also added. In the example of the figure, the sample number is k-1 for the audio frame number of the encoded audio data corresponding to the special encoded video data, and the sample number of the frame is 6
It means that it is 15. Further, it means that the audio frame number of the encoded audio data corresponding to the special encoded audio data is k + 6, and the number of frame samples is 12.

Next, the means for creating the additional data will be described. FIG. 9 shows the additional data creating means.
The terminal 201 is supplied with an original video signal. This original video signal is quantized by the quantization means 202 and input to the frame memory 203. The original video signal from the frame memory 203 is input to the video encoding unit 204 in frame units, and is output as encoded video data in frame units. The encoded video data is in a format as shown in FIG. 1, for example, in a formatter provided at a subsequent stage. The input terminal 205 is supplied with a video frame pulse, and is used as a timing signal for writing and reading the frame memory 203 and a timing signal for the video encoding unit 204. The input terminal 206 receives a program start pulse. The 1/6 frequency divider 207 is cleared by the program start pulse, and the video frame pulse is counted, thereby generating a pulse having a specific video frame period shown in FIG. 8 , that is, a specific video frame pulse.

On the other hand, an audio sampling pulse is input to the input terminal 208, and an original audio signal is input to the input terminal 209. The original audio signal is sampled and quantized by the sampling and quantization means 210, and the output is
The data is input to the audio encoding unit 211 and encoded. As a result, encoded audio data is obtained, and a frame number is added to the header in the subsequent stage.

When the number of samples per audio frame is N, the audio sampling pulse at the input terminal 208 is 1 / N
The signal is input to the frequency divider 212, divided by 1 / N, and output as an audio frame pulse. The audio frame pulse is
The signal is input to the audio encoding unit 211. Thus, the audio encoding unit 211 performs audio encoding on a frame basis. The audio frame pulse is input to the audio sampling pulse counter 213 as a clock. The audio sampling pulse counter 213 is cleared by an audio frame pulse. Therefore, the output data represents the number of audio samples in one audio frame. This number of audio samples is input to the register 215. The audio frame number is also input to the register 215. The audio frame number is created by the audio frame pulse counter 214. In other words, the audio frame pulse counter 214 clears each program start pulse and counts the number of audio frame pulses, thereby creating an audio frame number. An audio frame number and the number of audio samples are input to the register 215, which are latched by a specific video frame pulse and output. The number of audio samples is cleared by the audio frame pulse and increases sequentially, but in the middle, it is latched by the specific video frame pulse.
This is the audio sample number corresponding to the specific video.

The additional data obtained from the register 215 is used in a subsequent formatter to create a data unit as shown in FIG. FIG. 10 shows a synchronization processing means for reproducing the above-mentioned additional data and synchronizing audio and video.

At the time of reproduction, encoded video data, encoded audio data, and additional data are reproduced for each data unit. The output timing of the decoded video data and audio data is defined by the additional data. For each data unit, the encoded audio data read from the recording medium is input to input terminal 3
01 is input to the audio buffer 302 via the input terminal 311 and the encoded video data is input to the video buffer 3 via the input terminal 311.
12 is input. Further, additional data is input to the input terminal 321 and is taken into the register 322.

The encoded voice data is also input to the frame number extracting means 305. The encoded audio data output from the audio buffer 302 is input to the audio decoding means 303 and decoded, and is input to the audio block buffer 304. The coded video data output from the video buffer 312 is input to the video decoding unit 313 to be decoded, and is input to the video frame buffer 314. Video data is decoded in units of one video frame. The audio data is decoded in units of one audio frame, and decoded sample data for one audio sample block is stored in the audio block buffer 304.

The audio frame number extracted by the frame number extracting means 305 is input to the comparing means 323. The comparing unit 323 compares the frame number extracted from the additional data with the audio frame number extracted from the header of the encoded audio data. When a coincidence pulse (YES) is obtained in the comparing means 323, a sample number included in the additional data is extracted in the gate means 324, and this sample number becomes a preset input of the address counter 325.

Thus, the audio block buffer 304
, The position of the audio sample data from which the decoded audio data is to be read is determined. In addition, the coincidence pulse starts the audio sampling pulse generating means 326 and the video frame pulse generating means 327, and the audio data is output in synchronization with the corresponding sample number whose relationship with the video data is specified by the additional data. Become.

In the comparing means 323, the mismatch pulse (N
If O) is obtained, the additional data in the shift register 332 is shifted, allowed to proceed to read the next synchronization information, for example, a mismatch pulse when processing should that become especially 1 = k-1 in FIG. 8 Once obtained, the next synchronization information, special 2 = k
The shift register 322 is shifted to +6, and the frame number k + 6 included in the additional data is set in the comparing means 323. Here, it is determined whether or not the frame number matches the frame number included in the encoded audio data string, and this processing is performed until the frame number matches. If the frame numbers match by this processing (for example, if the match is obtained with special 2 = k + 6), then the video data of the video decoding means 313 and the frame buffer 314 is processed until the special video data is decoded. You. This synchronization adjustment is performed by adjusting means 328. Therefore, in this case, the audio is output in synchronization with the second specific video output.

Until the coincidence pulse is obtained from the comparison means 323, both the video and audio need not be output, or only the video may be output. After the coincidence pulse is obtained, the audio and video in the video group are synchronized, and thereafter, the operation of the comparing means may be stopped. Then, the comparing means may be operated periodically in accordance with the position of the specific video signal.

In the operation of the adjusting means 323 when the mismatch pulse is obtained, if the structure of the audio frame is large, the process proceeds to the second and fourth specific video frames. The frame is at most 204
Since the sample length is about 8, synchronization is established by the third specific video frame.

In the above description, when synchronizing the output timing of the specific video frame and the designated audio sample, the data output time of the video frame buffer and the audio block buffer is controlled. Means for adjusting the storage time of the data buffer memory (not shown) or the storage time of the encoded data buffer memory (not shown) may be further introduced.

FIG. 11 shows another embodiment of the video / audio correcting means. Encoded video data is input to an input terminal 401, and the encoded video data is decoded in a video decoding and frame buffer 402. Input terminal 40
3 is supplied with an internal clock of the reproducing apparatus, and is divided into 1 / M by a 1 / M frequency divider 404 and output as a video frame pulse. This video frame pulse is supplied as a timing signal to the previous video decoding and frame buffer 402, and is input to the 1/6 frequency divider 405,
1/6 is frequency-divided, and output as the specific video frame pulse synchronized with the particular video signal shown in FIG.

The encoded audio data is input to the audio decoding means 407 via the input terminal 406 and is decoded, and the decoded audio data is input to the decoded audio block buffer 408. An input terminal 411 receives an internal clock, and the internal clock is divided by a 1 / N divider 412 into 1 / N
And output as an audio sampling pulse.
The audio sampling pulse is input to the audio frame pulse creating means 413 and also to the decoded audio sample address counter 414. The audio frame pulse creating means 413 creates an audio frame pulse corresponding to the audio frame period, and supplies it to the decoded audio sample address counter 414 and the audio decoding means 407 as a timing signal.

The decoded voice sample address counter 414
Is reset at the audio frame pulse and counts the audio sampling pulses, so its output will represent the audio sample number. The audio sample number is used as a read address of the decoded audio block buffer 408 and is input to the register 415. The register 415 latches the audio sample number at the timing of the specific video frame pulse, and inputs this to the comparing means 416. The comparing means 416 compares the sound sample number included in the additional data supplied from the input terminal 417 with the sound sample number. Here, if a coincidence pulse is obtained, it means that the video data and the audio data are synchronized in a predetermined relationship.

However, if a mismatch pulse is obtained, it means that the audio frame specified by the additional data does not correspond to the specific video signal. Therefore, synchronization adjustment is necessary. In this system, the comparison means 416
Divide frequency adjustment signal to the 1 / N frequency divider 412. Thus, the phase of the audio sampling pulse and the phase of the audio frame pulse are controlled. Actually, when the difference between the two audio sample numbers to be compared by the comparing means 416 is equal to or more than a certain value, for example, the frequency division number of the 1 / N frequency divider 412 is increased or decreased by about 1 or 2. After this processing, if the difference between the two audio sample numbers compared by the comparing means 416 falls within a certain range, the adjustment state is maintained.

In the above description, the frequency division number of the 1 / N frequency divider 412 is adjusted. However, the frequency division number of the 1 / M frequency divider 404 may be adjusted, or both may be adjusted. Is also good. In this way, even if there is a small difference between the internal clock frequencies on the encoding side and the decoding side, it is possible to continue correcting the synchronization between the video and audio at a stage where no large synchronization deviation occurs.

[0063]

As described above, according to the present invention,
Efficient data recording is possible, data management is easy, special reproduction of programs, high-speed search is possible, and synchronization between video and audio can be accurately and easily obtained.

[Brief description of the drawings]

FIG. 1 is a diagram showing a signal compression format according to an embodiment of the present invention.

FIG. 2 is a diagram showing a signal compression format according to another embodiment of the present invention.

FIG. 3 is a diagram showing a signal compression format according to still another embodiment of the present invention.

FIG. 4 is an explanatory diagram of a disk data array and a management table shown for explaining data management in one embodiment of the present invention.

FIG. 5 is a detailed explanatory diagram of a control chart table.

FIG. 6 is an explanatory diagram showing an example of a management chart table.

FIG. 7 is a diagram showing a configuration example of an entire system according to an embodiment of the present invention.

FIG. 8 is a diagram showing details of a data unit in one embodiment of the present invention.

FIG. 9 is a diagram illustrating an example of an encoder of the data unit processing system of FIG. 8;

FIG. 10 is a diagram showing an example of a decoder of the data unit processing system of FIG. 8;

FIG. 11 is a diagram showing another example of the decoder of the data unit processing system of FIG. 8;

FIG. 12 is an explanatory diagram shown to explain a conventional disk management system.

[Explanation of symbols]

101: separating means, 102: voice grouping means, 10
3 ... Video grouping means, 104 ... Extended data grouping means, 105 ... Audio compression means, 106 ... Video compression means, 107 ... Extended data compression means, 108 ... Formatter, 121 ... Separation means, 122 ... Audio decoder, 123
... video decoder, 124 ... extended data decoder, 125
... Synthesizing means, 202 ... Quantizing means, 203 ... Frame memory, 204 ... Video encoding means, 207 ... 1/6 frequency divider, 210 ... Sampling quantizing means, 211 ... Speech encoding means, 211 ... 1 / N Frequency divider 214 audio frame pulse counter 213 audio sample pulse counter 215 register 302 buffer 303 audio decoding means 304 audio block address 305
Frame number extracting means, 312 ... buffer, 313 ... video decoding means, 314 ... video frame buffer, 322 ...
Register, 323: comparison means, 324: gate means, 3
25 ... Address counter, 326 ... Audio sampling pulse generating means, 327 ... Video frame pulse generating means.

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int. Cl. ⁶ , DB name) H04N 5/781 H04N 5/91-5/956 G06F 17/30 G06T 9/00 H04N 7/24

Claims (4)

(57) [Claims] 1. Main video data separable in units of one frame
Main video compression means for compressing and encoding the main video,Select subtitles for subtitles in different languages for
Superimpose is possibleSub-picture data before
Independent of the main video compression means for compressing the main video data
Sub-picture compression means for compressing and encoding;Audio data code that encodes audio data of multiple channels
Encoding means; Sub-picture data and sound corresponding to the time unit of the picture data
The main video compression means and sub-video
Means for controlling image compression means and audio data encoding means;  The encoded main video from the main video compression means
Combining the data to form a first packet; and
The encoded sub-picture from the sub-picture compression means
Combining the data to form a second packet;Further
The encoded audio data from the audio data encoding means.
Combine to form a third packet,
Adds additional data for audio synchronization and adds
A data unit to record or transmit to the data unit.
OutputAnd a formatter.
Compressed signal generator. 2. Main video data separable in units of one frame.
Compress and encode the main videoSelect subtitles for subtitles in different languages for
Superimpose is possibleSub-picture data before
Independent of the main video compression means for compressing the main video data
Compress and encode,Encodes audio data of multiple channels, Sub-picture data and sound corresponding to the time unit of the picture data
Compression and encoding of the main video so that voice data is synchronized
Code, sub-picture compression and encoding, audio data code
Control the encryption,  A first packet of the encoded main video data
Combined to form and the encoded sub-picture
Combining the image data to form a second packet;Sa
The encoded audio data obtained by encoding the audio data
3 to form a packet, plus video and audio
Data unit with additional data for voice synchronization
To generateRecording or transmission system with the data unit
And generating a compressed signal. The 3. A main video data that can be separated by one frame and the main picture packets compressed and encoded main picture圧宿means, auxiliary for different languages subtitles for the main-video
Sub-picture data that allows the video to be selectively superimposed , sub-picture packets that are compressed and encoded independently of the main picture compression unit that compresses the main picture data;
An audio packet obtained by encoding audio data of a plurality of channels and received from a recording system or a transmission system for recording or transmitting additional data for synchronizing video and audio , the main video packet, the sub video packet,
A separating unit that separates an audio packet and additional data, a main video decoder that receives the main video packet from the separating unit and obtains decoded main video data, and receives the sub video packet from the separating unit. A sub-video decoder that obtains decoded sub-video data; an audio decoder that receives an audio packet from the separation unit and obtains decoded audio data; and an additional data from the separation unit. Signal, means for synchronizing video and audio, and superimposing the sub-video data obtained from the sub-video decoder on the main video data output from the main video decoder to generate subtitles of different languages. An apparatus for reproducing a compressed signal, comprising: synthesizing means selectively obtained. The main video data that can be separated by 4. 1 frame and main video packets compressed and encoded main picture圧宿means, auxiliary for different languages subtitles for the main-video
Sub-picture data that allows the video to be selectively superimposed , sub-picture packets that are compressed and encoded independently of the main picture compression unit that compresses the main picture data;
Receiving a plurality of channels of audio data and audio packets encoded, from the recording system or transmission system for transmitting for recording or transmitting the additional data for synchronization Film images and sound, the main video packets, sub-picture packet,
Audio packet and additional data, receiving the separated main video packet, obtaining decoded main video data, receiving the separated sub video packet, obtaining decoded sub video data, Receiving an audio packet, obtaining decoded audio data, receiving the separated additional data, obtaining a timing signal of each of the decoders, synchronizing video and audio, and performing decoding on the decoded main video data with respect to the decoded main video data. A compressed signal reproducing method characterized by superimposing video data and selectively obtaining subtitles in different languages.