JP3273601B2 - Encoding device and method, data recording / reproducing device and method, and recording medium - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an encoding apparatus and method, a data recording and reproducing apparatus and method, and a recording medium. The orthogonal transform coefficient is
DC components and AC components are arranged in order from low to high,
The present invention relates to an encoding apparatus and method for encoding fixed-length encoded data in a variable-length format, a data signal recording / reproducing apparatus and method, and a recording medium.

[0002]

2. Description of the Related Art In recent years, several types of formats for recording and transmitting digital video signals on a recording medium have been provided. Generally, a digital video signal has an extremely large amount of data. Therefore, when it is desired to record the digital video signal on a recording medium for a long time, it is necessary to compress and encode the video signal. A so-called MPEG system is known as a representative example of the compression encoding system. This MPEG (Moving Picture Image Coding Experts Gr
oup) method is ISO-IEC / JTC1 / SC2 / W
This is discussed in G11 and proposed as a standard, and includes motion compensated predictive coding and discrete cosine transform (DCT:
This is a hybrid system combining Discrete Cosine Transform (encoding) and encoding. In the MPEG system, the redundancy in the time axis direction is first reduced by taking the difference between frames of the video signal, and then the redundancy in the spatial axis direction is reduced by using the discrete cosine transform. The signal is coded efficiently.

A video tape recorder (VTR) using a tape-like medium such as a magnetic tape as a recording medium.
Video Tape Recorder) uses a rotating head,
Generally, recording is performed so as to form a track obliquely inclined with respect to the tape running direction, that is, a so-called helical track. When reproducing a tape-shaped recording medium on which such a helical track is recorded at a higher tape running speed such as double speed, triple speed, or search, the angle of the trajectory of the rotary head on the tape is changed. Since the angle differs from the inclination angle of the recording track, it is impossible to reproduce all the signals recorded on the helical track. That is, at the time of high-speed reproduction, reproduction is performed in which a part of each helical track is scanned (traced).

When the above-mentioned MPEG system is used as it is as a compression encoding system for a tape-shaped recording medium and high-speed reproduction such as a search as described above is performed, a part of each helical track is traced and reproduced. It is difficult to effectively use the data to obtain a high-quality reproduced image.

[0005] For this reason, it is preferable to use a compression encoding method that enables a somewhat effective image reproduction even at high speed reproduction, rather than using the MPEG method as it is for the compression encoding method of a tape-shaped recording medium.

[0006] In view of this point, the applicant of the present application firstly
For each macroblock, all D in the macroblock
By arranging each DC coefficient of a CT block and sequentially arranging low-order components to high-order components of each AC coefficient of all DCT blocks for each order and sequentially arranging them, a high-speed reproduction such as a search can be performed within a macro block. Proposed a recording format of a compression coding system in which all DC coefficients and low-order AC coefficients important for image reproduction can be picked up.

[0007]

Incidentally, such a format is specialized for, for example, a VTR for a broadcast service, and when data transmission with other devices is considered, the format conforms to the MPEG standard. It is preferable to use encoded data in a global standard format.

However, encoded data in a format optimized for recording on the VTR or the like is decoded to uncompressed original video data, and compression encoding in a standard format such as MPEG is performed. When applied, the circuit configuration and the processing amount become large. It cannot be realized without image quality deterioration.

Further, in the case of the MPEG system, the encoder is controlled by a feedback system so that the code amount becomes a target code amount. As a result, as shown in FIG. 1, for example, when the target code length is observed in GOP (Group Of Picture) units, the code length of each GOP becomes longer or shorter than the target code length. In the example of FIG. 1, although GOP1 and GOP2 are longer than the target code length, GOP3 and GOP4
Is shorter than the target code length. The next GOP5 and GOP6 become longer than the target code length again, and GOP7 becomes shorter than the target code length. GOP8 is also longer than the target code length.

As described above, the code length of each GOP is controlled so that its average value becomes the target code length.

However, the storage capacity of the helical track formed by the VTR is constant. If the target code length is set according to the capacity of the helical track, for example, the target code length of GOP1, GOP2, GOP5, GOP6, GOP8 is The long part of the data will be deleted and will not be recorded on the helical track.

Therefore, if the target code length is set to a value sufficiently shorter than the capacity of the helical track, even if an overshoot occurs with respect to the target code length, the G value of the target code length can be reduced.
It is possible to record all data of the OP on the helical track.

[0013] However, in this case, the utilization efficiency of the helical truck is reduced.

SUMMARY OF THE INVENTION The present invention has been made in view of such a situation, and it is possible to record video data encoded by the MPEG system on a helical track without lowering the utilization efficiency of the helical track. It is to be.

[0015]

According to a first aspect of the present invention, there is provided an encoding apparatus for orthogonally transforming input data and compressing and encoding the data, and for each orthogonal transform block, the orthogonal transform coefficient includes a DC component and an AC component. disposed in order from the low order higher components, comprising converting means for substantially converting the fixed length coded data of the format of a variable length standard, conversion means is inputted
Data is orthogonally transformed and compression-encoded, and multiple orthogonal transforms are performed.
Orthogonal transformation of each block aggregate
Coefficients are arranged in order from low order to high order of DC component and AC component.
Into fixed-length encoded data of the first format.
First converting means for converting, and
The fixed-length first format encoded data
For each orthogonal transform block, the orthogonal transform coefficient is DC
Variable, arranged from low to high order of component and AC component
To encoded data of the second format
And a second conversion unit for converting .

The encoding method of claim 8, the input data, and orthogonal transformation and compression coding, every orthogonal transform block, the orthogonal transform coefficient, a direct current component, high from low-order AC component A conversion step of converting the input data into substantially fixed-length encoded data in a format of a variable-length standard, the conversion step being arranged in the following order:
Are orthogonally transformed and compression-encoded, and multiple orthogonal transform
The orthogonal transform coefficient for each block aggregate
DC components and AC components are arranged in order from low to high,
Convert to fixed-length first format encoded data
The first conversion step and the processing of the first conversion step
Coded data of the fixed format first format
Data for each orthogonal transform block.
Are arranged in the order of low to high order of DC component and AC component.
In addition, the encoded data of the second format which is a variable length standard
And a second conversion step of converting the data into data.

The recording medium of claim 9, the input data, and compression coding by orthogonal transformation, for each orthogonal transform block, the orthogonal transform coefficients, the DC component, higher from lower order AC components Comprises a conversion step of converting the input data into substantially fixed-length encoded data in a format of a variable-length standard , the input step comprising:
Orthogonal transform and compression encoding, and multiple orthogonal transform blocks
The orthogonal transformation coefficient is DC for each
Fixed, arranged from low to high order of component and AC component
First to convert the encoded data into a long first format
In the conversion step and the processing in the first conversion step.
Converted fixed-length encoded data of the first format
For each of the orthogonal transform blocks,
DC components and AC components are arranged in order from low to high,
Coded data of the second format which is a variable length standard
And a second conversion step of converting the program into a program.

According to a tenth aspect of the present invention , in the data recording / reproducing apparatus, the input data is orthogonally transformed and compression-encoded,
Conversion means for converting orthogonal transform coefficients into substantially fixed-length coded data in a variable-length standard format in which the orthogonal transform coefficients are arranged in the order of low-order to high-order DC components and AC components for each orthogonal transform block Recording means for recording the encoded data converted by the converting means on a recording medium; reproducing means for reproducing the fixed-length encoded data recorded thereon from the recording medium; and reproducing means for reproducing the encoded data. Also, in the standard,
Output means for outputting coded data of a fixed length in a variable length format , wherein the conversion means outputs the input video data.
Data is orthogonally transformed and compression-encoded, and multiple orthogonal transform
For each block aggregate that locks together, its orthogonal transform
The numbers are arranged in the order of low to high order of DC component and AC component.
Converted to encoded data of fixed-length first format
A first converting unit that converts the
In addition, the fixed-length first format encoded data is
For each orthogonal transform block of
Variable length, arranged from low to high order of AC component
To encoded data of the second format that is the standard
And a second conversion unit that performs the conversion .

According to a twelfth aspect of the data recording / reproducing method, the input data is orthogonally transformed and compression-encoded,
For each orthogonal transform block, a transform step of transforming the orthogonal transform coefficients into substantially fixed-length coded data of a variable-length standard format in which the DC component and the AC component are arranged in the order of low to high order. A recording step of recording the encoded data converted by the processing of the conversion step on a recording medium, and a reproduction step of reproducing the fixed-length encoded data recorded on the recording medium by the processing of the recording step.
The conversion step includes converting the input video data directly.
Interchange transform and compression-encode and combine multiple orthogonal transform blocks.
The orthogonal transform coefficient of each block
Fixed length, arranged from low to high order of AC component
To convert to encoded data of a first format of
Conversion by the processing of the conversion step and the first conversion step
Fixed-length encoded data of the first format
For each of the orthogonal transform blocks,
DC components and AC components are arranged in order from low to high,
Coded data of the second format which is a variable length standard
And a second conversion step of converting into

The recording medium according to claim 13, the input data, and compression coding by orthogonal transformation, for each orthogonal transform block, the orthogonal transform coefficients, the DC component, higher from lower order AC components A conversion step of converting the data into a substantially fixed-length encoded data of a variable-length standard format, and a recording step of recording the encoded data converted by the processing of the conversion step on a recording medium. , in the process of recording step, recorded on the recording medium, and a reproduction step of reproducing the fixed length coded data, converts stearate
The input video data is orthogonally transformed and compressed.
A block that encodes and combines multiple orthogonal transform blocks
For each assembly, its orthogonal transform coefficients are
A first fixed-length format arranged from low to high order.
First conversion step of converting to encoded data of mat
Fixed by the processing of the first conversion step
Coded data of the first format with the orthogonal transform
For each transformation block, its orthogonal transformation coefficient is
A variable length standard that is arranged in order from low to high order of components
A second format for converting into encoded data of a certain second format
And a conversion step .

[0021] In the encoding device according to the first aspect, the encoding method according to the eighth aspect , and the recording medium according to the ninth aspect , the input data has a fixed length of a variable length format according to the standard. Is converted to encoded data.

A data recording / reproducing apparatus according to claim 10 ,
A data recording / reproducing method according to claim 12 , and claim 1.
In the recording medium described in No. 3 , input video data is converted into fixed-length encoded data in a variable-length format according to the standard, recorded on the recording medium, and reproduced and output.

[0023]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 2 is a block diagram showing a schematic configuration of the data conversion device according to the embodiment of the present invention.

The data conversion apparatus shown in FIG.
At the time of compression encoding of video data accompanied by orthogonal transform such as discrete cosine transform (discrete cosine transform), orthogonal transform is performed for each block aggregate (for example, macro block) in which a plurality of orthogonal transform blocks (for example, DCT blocks) are combined. Coefficients (for example, DCT coefficients) are coded data of a first format arranged in the order of low order to high order of a DC component and an AC component, and an orthogonal transform during compression encoding of video data accompanied by an orthogonal transform. For each block, data conversion is performed between the orthogonal transform coefficient and the coded data of the second format in which the DC component and the AC component are arranged in order from the lower order to the higher order. Specifically, as the first format, the coefficients of the direct current (DC) components of each DCT block are grouped in a macroblock, and the coefficients are grouped in order from low to high order of an alternating current (AC) component. (For example, SX format (trademark) proposed by the present applicant), and the second format is a so-called MPEG (Moving Picture Image Coding Expert).
s Group) standards, especially MPEG2 4: 2: 2P @ ML (4: 2:
Format according to the standard of 2 profiles (main level).

In the data converter shown in FIG.
An input terminal 10 is supplied with a data stream DS1 of the encoded data of the first format, which is partially decoded until obtaining the orthogonal transform coefficients (eg DCT coefficients). Variable length decoding circuit 11 as a decoding means for performing
Sent to The first coded data, which is an input data stream, has a DCT coefficient that has been previously subjected to variable length coding (VLC: Va).
riable Length Coding) and variable length decoding circuit 1
1 decodes the variable-length-encoded data portion to clarify the data delimiter, and sends an output signal S1 capable of rearranging the data to the conversion circuit 12.

The conversion circuit 12 corrects the arrangement of data, which is the difference between the first format and the second format, and when the control signal S5 from the stuffing circuit 15 is "1", the DCT coefficient Higher order (AC
The non-zero coefficient (of higher order) is replaced with zero and the output signal S2 is output.

The variable length coding circuit 13 receives the signal S2 from the conversion circuit 12 as an input signal, performs variable length coding again on data requiring variable length coding, and outputs an output signal S3. .

The header adding circuit 14 detects the timing of the input data in advance from the data stream DS1 of the input coded data of the first format, and
For example, various types of header information defined in MPEG2, for example, are prepared. Header addition circuit 14
Then, the prepared header information is added to the signal S3 from the variable length coding circuit 13 to output an output signal S4.

The stuffing circuit 15 receives the signal S4 to which the header information has been added by the header adding circuit 14 as an input signal, and receives a GOP (Group Of Picture).
icture) unit, a stuffing bit (“0”) is inserted if necessary, the length is smoothed in GOP units, and the second format (MPEG
The data stream DS2 of 2) is output from the output terminal 16. Here, when the data length in GOP units exceeds (exceeds) the set data length, the control signal S5 is set to “1” and sent to the conversion circuit 12 as appropriate. As a result, it is possible to increase the efficiency of variable-length coding by replacing higher-order non-zero coefficients of DCT coefficients with zeros, and to perform simple data rate reduction.

The control I / F (interface) 17
It communicates with an external CPU (not shown) via a terminal 18 by a system controller interface signal S6, and performs initial settings to the variable length decoding circuit 11 to the stuffing circuit 15 by respective control signals, Has a function of notifying the external CPU of the operation state of the external CPU.

Next, the MPEG standard as a specific example of the second format and a specific example of the first format will be described with reference to FIGS.

FIG. 3 shows an MPEG standard as a specific example of the second format, in particular, 4: 2: 2P of MPEG2.
It is a figure for explaining the hierarchical structure in the case of ML (4: 2: 2 profile at main level).

The sequence (Sequen) shown in FIG.
The ce) layer has a sequence header code (SHC: sequence).
_Header_code), header part, extension
ion), several GOPs (Group Of Pictures) are arranged, and if necessary, several sets from the SHC to the GOPs are arranged. Code (SEC: se
quence_end_code).

As shown in the GOP layer of FIG. 3B, the GOP has a group start code (GSC: group
_Start_code), a header part, an extension part, and a number of pictures (Pictures). The pictures include an intra-coded picture (I picture: Intra Picture), a forward predictive coded picture (P picture: Predictive Picture), and a bidirectional predictive coded picture (B picture: Bidirectionally Predictive P).
picture), and these I, P, and B pictures are arranged in a predetermined order to form a GOP.

The picture (Picture) is the picture shown in FIG.
As shown in the picture layer, the picture start code (PS
C: picture_start_code), a header section and an extension user data section, followed by a number of slices (Slices). The slices are arranged as shown in the Slice layer in FIG. , A header from a slice start code (SSC: slice_start_code) to several macro blocks (MB: macroblo
ck) is arranged.

One macro block MB is shown in FIG.
As shown in the MB (Macroblock) layer, the address (addr
ess), mode (mode), quantization scale code (qs
c: quantizer_scale_code), motion vector (mv:
motion_vectors) and code block pattern (c)
Following bp: coded_block_pattern), a predetermined number of blocks are arranged. This block is composed of 8 × 8 DCT coefficients obtained by performing a DCT (discrete cosine transform) on a DCT block of 8 × 8 pixels, and one macro block MB is 4: 2: 2P ＠ ML.
, Four luminance signal blocks Y as shown in FIG.
0, Y1, Y2, Y3, and two color difference signal blocks Cb0, Cb1, Cr0, Cr1, respectively, for a total of eight DCs
It is composed of T blocks.

Here, 8 × 8 in one DCT block
As for the DCT coefficients, a DC (direct current) component is arranged at the upper left, and as shown in FIG. 5 (A), from the lower left (low frequency) of the AC (alternating current) component from the upper left to the lower right, as shown in FIG. They are arranged in the following order (high frequency). DCT in this block
A method of extracting coefficients in the order shown in FIG. 5A and performing variable-length coding is called zigzag scanning. In MPEG2,
In addition to the zigzag scan, an alternate scan method as shown in FIG. 5B is permitted, and one of these zigzag scans and alternate scans can be switched and used on a picture basis. ing. The alternate scan method can efficiently pick up interlaced components and is suitable for encoding interlaced images.

FIGS. 6 (A) and 6 (B) show specific examples of DCT coefficients converted into a one-dimensional coefficient sequence by these scanning methods.
Shown in That is, FIG. 6A shows a case where 8 × 8 two-dimensional DCT coefficients in a DCT block are converted into a one-dimensional coefficient sequence by zigzag scanning, and FIG.
Shows DCT coefficients similarly converted into a one-dimensional coefficient sequence by alternate scan.

FIG. 3F is obtained by performing variable-length coding (also referred to as entropy coding) such as Huffman coding on a DCT coefficient sequence arranged in one dimension for each DCT block. It shows an encoded data sequence, and shows data obtained by encoding rAC-1 [y0] and the like. In the variable length coding, the code rAC ** is determined according to the zero run length (run) and the non-zero value (level). However, since the code rAC ** is not related to the embodiment of the present invention, the description is omitted. Omitted.

The above is the MPEG standard, especially MPEG2 4:
FIG. 7 shows an example of a coded data stream in the second format corresponding to the case of 2: 2P @ ML. As a first format for this, the format proposed earlier by the present applicant is, for example, as shown in FIG. It has a hierarchical structure.

In the specific example of the first format shown in FIG. 7, one GOP in the GOP sequence forming the sequence (Sequence) layer in FIG. 7A is added to the GOP layer in FIG. 7B. As shown in FIG.
Made of frame. This is a specific example of the first format when applied to a video tape recorder, and the first format itself includes I,
GOPs of up to 30 frames including P and B pictures are allowed. Further, when applied to a video tape recorder, for example, stuffing or the like is performed in order to record on a tape with a GOP unit as a fixed length, but the description is omitted because it has little relation to the embodiment of the present invention. I do.

The picture (Picture) shown in FIG.
The layer and the slice layer in FIG.
Together with the sequence layer of (A) and the GOP layer of FIG.
Although it substantially conforms to the MPEG standard of FIG. 3 described above, header information is not added, and when converting from the first format to the second format, various header information may be added. Needed.

The macro block MB (Macrob) shown in FIG.
In the (lock) layer, the arrangement order of the DCT coefficients is different from the case of the MPEG format in FIG. That is, as shown in the encoded data of FIG. 7F, only eight DC (direct current) components of the eight DCT blocks Y0 to Cr1 shown in FIG. In addition, from the low order (low frequency) to the high order (high frequency) of the AC (alternating current) component, eight DCT coefficients of the corresponding frequency of each DCT block are arranged collectively and sequentially. This means that DC components and AC low-frequency components having high importance during image reproduction are arranged close to each other in the macroblock MB.

Here, the advantage of setting the arrangement order of the DCT coefficients in this manner will be described with reference to FIGS.

When the encoded data in the first format or the second format as described above is recorded as an oblique helical track on a video tape using a rotary head, and when the data is reproduced at a high speed, the recording is performed. Since the inclination angle of the head trajectory is different from the inclination angle of the helical track, which is a track, only a part of the helical track can be reproduced, and even if error correction is performed, only a part of the macro block MB is effectively reproduced. And the rest fail.

FIG. 8 shows a case where a video tape recorded in the above-described MPEG2 format, which is a specific example of the second format, is reproduced at a high speed. In FIG. 8, a hatched portion indicates an uncorrectable error area. ing. FIG. 8 (A)
8B shows a one-dimensional array of encoded data of DCT coefficients in a macroblock. FIG. 8B shows DCT coefficients obtained by decoding the data, where the horizontal axis represents frequency components and the vertical axis represents two-dimensional DCT blocks. Are shown. This FIG.
In the example of (B), the DCT blocks Y0, Y of the luminance signal Y
1, Y2 and Y3 are valid, and the DCT block C of the color difference signal
b0, Cr0, Cb1, Cr1 are in error, and an image without color components is obtained. When the color difference signal is valid up to Cb0, a strange color is added, and the luminance signal Y
When the error 3 also occurs, some DCT blocks of the Y component in the macroblock are missing. Such a reproduced image is unsightly and is not preferable.

On the other hand, when a video tape recorded in the first format is reproduced at a high speed, FIG.
The hatched portion in the one-dimensional array shown in FIG.
As shown in FIG. 9B, all the Ds in the macroblock
For the CT block, DC components and A
The C low frequency component is effective, and although a slightly blurred image is obtained, a relatively good reproduced image is obtained. Note that, at the time of high-speed reproduction, the error occurrence position changes for each macroblock. However, if the DC component and the low-frequency component of AC cannot be obtained effectively in the macroblock, the data of the macroblock is deleted. It is sufficient to display the effective image of the past frame without using it, and the display image is such that it is partially updated in units of macroblocks, but there is no discomfort in appearance.

In the first format, the zigzag scan and the alternate scan shown in FIGS. 5 and 6 can be switched on a macroblock MB basis, whereby the image quality can be improved. ing.

That is, a frame DCT and a zigzag scan are used in combination for an image or a frame image (a progressive scan image) with a small motion, and a field DCT and an alternate scan are used for an interlace image with a large motion. Is advantageous. In the first format, by switching this in units of macroblocks, finer adjustments can be made and the image quality is improved.

On the other hand, in the second format (MPEG), the scan type must be unified within the picture as described above. That is,
The DCT type is allowed to switch between the field DCT and the frame DCT on a macroblock basis, but is not allowed to switch the scan type.

Therefore, in the conversion circuit 12 shown in FIG. 2, when the scan type is switched in macroblock units in the first format, the DCT coefficients are rearranged so that the pictures have the same scan type. Go and conform to the second format.

Regarding the specific examples of the first and second formats as described above, specific processing for converting encoded data of the first format to encoded data of the second format (MPEG) This will be described with reference to FIG.

FIG. 10A shows a one-dimensional array of coded data of DCT coefficients in one macroblock in the coded data stream of the first format, and the variable length decoding circuit 11 shown in FIG. Then, the head coded data DC [Y0] in this macroblock is detected to clarify the data division. This leading encoded data DC [Y0]
Thus, writing is performed in the memory in the variable length decoding circuit 11 in the order shown by the arrow in FIG. In addition,
In FIG. 10 (B), when expanding the coefficient data into a set of zero run length (run) and non-zero value (level) by variable length coding in a memory, for example, the coding AC
This shows that the coefficient data AC-2 and AC-3 may be expanded at once.

The coefficient data expanded in the memory in the variable length decoding circuit 11 and subjected to the variable length decoding processing is shown in FIG.
As shown by the arrow, the DC (direct current) component to the AC (alternating current) component are read in order from the low frequency to the high frequency for each DCT block, so that the coefficient arrangement order of the second format (MPEG) described above is obtained. Is converted to Also, at this time, if the scan type (zigzag scan and alternate scan) is switched for each macroblock in the first format, the DCT coefficients are rearranged so that they have the same scan type in the picture. , The second format (MPEG). This corresponds to the processing in the conversion circuit 12 in FIG. Next, the data is subjected to variable-length coding in the variable-length coding circuit 13 in FIG. 2 to become coded DCT coefficient data as shown in FIG. This one-dimensionally corresponds to the encoded data in FIG.

Referring to FIG. 11, a specific example in which the above-described data converter is applied to a helical scan type video tape recorder (VTR) which is a signal recording / reproducing apparatus for recording in the first format will be described. I will explain it.

In FIG. 11, a serial data signal SDI In in which a digital video signal and a digital audio signal are multiplexed is supplied to a serial data input circuit 112 at an input terminal 111. The serial data input circuit 112 converts the input signal SDI In from a serial signal to a parallel signal, separates a video signal and an audio signal from the converted parallel signal, and outputs the video signal Video In to the first format encoder 113. The audio signal Audio In is sent to each of the delay circuits 114. Further, the serial data input circuit 112 sends a signal Input Sync, which is a phase reference of the input signal SDI In, to the timing generation circuit 142.

The timing generation circuit 142 has an input terminal 1
A timing signal required by the video tape recorder in synchronization with either a reference synchronization signal (Reference Sync) extracted from the external reference signal REF In from the external reference signal 41 or a signal Input Sync from the serial data input circuit 112. Is generated as a timing pulse and output to each circuit.

The first format encoder 113 compresses data of the input video signal Video In by encoding, and converts the video signal VF1 of the above-described first format into a video signal VF1.
To the ECC (error correction code) encoder 115. In the video tape recorder of this specific example, since the audio signal is treated as uncompressed data, the delay circuit 114 converts the input audio signal Audio In into uncompressed data for the processing time of the first format encoder 113. As an audio signal AU delayed by only this, it is sent to the ECC encoder 115.

The ECC encoder 115 performs an error correction encoding process on the compressed video signal VF1 and the uncompressed audio signal AU as input signals, and performs recording data RE.
The data is sent to the equalizer 116 as C DATA. The equalizer 116 records the input recording data REC DATA in the recording RF
The signal is converted into a signal REC RF and sent to a rotating head (not shown) of the rotating drum 117. The rotating drum 117 (the recording head thereof) applies a recording RF signal REC RF to the tape 120.
Is recorded and formed on an oblique helical track.

At the time of reproduction, the rotating drum 117 (reproduction head) reproduces the reproduction RF signal PB RF from the tape 120 and sends it to the equalizer 116. The equalizer 116 performs a phase equalization process on the input reproduction RF signal PBRF, and sends reproduction data PBDATA to the ECC decoder 121.

The ECC decoder 121 performs an error correction decoding process on the input reproduction data PB DATA,
The compression-encoded playback video signal VF1PB of the first format and the uncompressed playback audio signal AUPB are output. The reproduced video signal VF1PB is supplied to the first format decoder 122, the serial data output circuit 12
6 and the data conversion circuit 131 corresponding to the data conversion device shown in FIG. 2, and the reproduced audio signal AUPB is sent to the delay circuit 123.

The first format decoder 122 obtains an original uncompressed video signal by decoding the input reproduced video signal VF1PB, and outputs an output video signal Video
It is sent to the serial data output circuit 124 as Out.

The delay circuit 123 delays the input reproduced audio signal AUPB by the amount of adjusting the timing with the video signal. Output signals AUDL1, AUDL2, AUDL delayed according to each video signal
3 to the serial data output circuits 124, 126, 132
To each other.

The serial data output circuit 124 outputs the video signal VideoOut and the audio signal
AUDL1 is converted from a parallel signal to a serial signal,
It is output from the output terminal 125 as a serial data signal SDI Out in accordance with a predetermined serial data transmission format.

The serial data output circuit 126 converts the audio signal AUDL2 whose timing has been adjusted and the reproduced video signal VF1PB of the first format from the ECC decoder 121 from a parallel signal to a serial signal. It is output from the output terminal 127 as a serial data signal in accordance with the transmission format.

A data converter as shown in FIG. 2 is used. The data conversion circuit 131 converts the input reproduced video signal VF1PB of the first format into
The video signal VF2 of the second format (MPEG)
And sends it to the serial data output circuit 132.

The serial data output circuit 132 outputs the audio signal AUDL3 whose timing has been adjusted and the video signal VF2 of the second format that has been subjected to the data conversion.
The signal is converted from a parallel signal to a serial signal, and is output from the output terminal 133 as a serial data signal in accordance with a predetermined serial data transmission format. The output signal from the output terminal 133 is a serial data signal of a standard MPEG stream and can be sent to an external MPEG device.

The system controller 144 performs optimal control of the digital VTR by communicating with the servo circuit 146 and the respective circuits using the SY_IO and SERVO_IO signals while cooperating with each other using the SY_SV signal.

By the way, if the data conversion circuit 131 shown in FIG. 11 is configured to simply convert a signal of the first format into a signal of the second format,
As described with reference to the above, when the code length of the GOP overshoots the target code length, since the capacity of the helical track is fixed, data is discarded as much as the overshoot. May not be able to record on helical tracks. Therefore, FIG.
1 can be configured, for example, as a fixed-length MPEG encoder 161 as shown in FIG.

This fixed-length MPEG encoder 161 has a pre-filter 162, and the input video data Video
o Remove unnecessary high frequency components of IN. This prefilter 1
62 is, for example, FIR (Finite Impulse Response) Dig
The ital Filter can be configured as a so-called low-pass filter. A video containing many high-frequency components has a large amount of information, and the bit rate is increased by that amount, which is disadvantageous in compression. Therefore, the image quality can be improved by attenuating the high frequency component in advance.

The pre-filter 162 can set a filter coefficient in response to a command from the CPU 166 instructed via the CPU bus, and includes a through mode in which the input video data Video In is output as it is. ,
A predetermined frequency characteristic can be set according to the bit rate.

The video data FL-Video whose wide area has been attenuated by the pre-filter 162 is
3 is input. The fixed length encoder 163 is configured as shown in FIG. 13, for example.

The pre-encoder 171 includes a pre-filter 1
The video data FL-Video input from 62 is encoded by eight preset quantization values, and a quantization value corresponding to a code amount exceeding the target code length and a code below the target code length are encoded. Among the quantized values corresponding to the amounts, the quantized values corresponding to the two code amounts closest to the target code length are selected, and the selection result is output to the bit allocation unit 172 as Result.

The bit allocation section 172 calculates the optimum quantization value for obtaining the code amount of the target code length for each macroblock based on the two quantization values input from the pre-encoder 171. Scale code (Q scale
code) to the encoder 174.

The delay unit 173 delays the video data FL-Video input from the pre-filter 162 by an amount corresponding to the delay time caused by the processing of the pre-encoder 171 and the bit allocation unit 172, and delays the video data FL-Video.
The video is output to the encoder 174.

The encoder 174 uses the quantization value input from the bit allocation unit 172 to
The video data Delayed Video input from 73 is encoded and output as a fixed-length first format video data stream SX stream.

Here, the fixed length is a code controlled so that encoding is performed efficiently (with a code amount close to the target code length) without exceeding the target code amount in each GOP. Means quantity.

Returning to FIG. 12, the fixed length encoder 163
, The bit rate and the GOP structure can be set from the CPU 166 via the CPU bus, and an optimal fixed length encoding is performed for each GOP.

The motion prediction processor 164 determines that the GOP is 1
This is unnecessary when the frame is composed of frames, but GOP is 1
This is required if the unit is composed of units larger than the frame.
In this case, the motion prediction processor 164 exchanges the signal Motion IF with the fixed-length encoder 163,
A motion vector is detected for each macroblock, and is notified to the fixed-length encoder 163. The operation of the motion prediction processor 164 is also controlled by the CPU 166.

The data conversion device 165 is configured to output the fixed-length video data SX Str input from the fixed-length encoder 163.
Convert eam to standard MPEG Stream. MPEG is
Although the format is a variable length format according to the standard, the input video data SX Stream has a fixed length, so that the MPEG Stream output from the data converter 165 also has a fixed length. This stream is MPEG2 4:
2: 2P @ ML. This data converter 1
65 is configured, for example, as shown in FIG.

The CPU 166 controls the data converter 165 via the CPU bus. Further, the CPU 166 communicates with an external CPU via a host interface Host IF, and executes a pre-filter 162 and a fixed-length encoder 16.
3. Motion prediction processor 164 and data converter 1
In addition to the initial setting of 65, control such as various settings is performed.

Next, the operation will be described. When receiving the input of the video data VF1PB input from the ECC decoder 121, the pre-filter 162 removes the unnecessary high frequency component, and outputs the video data FL-Video to the pre-encoder 171 and the delay unit 173 of the fixed-length encoder 163. Supply.

The pre-encoder 171 divides each macro block of the input video data into four DCT blocks as shown in FIG. 14, for example, and sets DCT coefficient values DCT1 to DCT4 of each DCT block in advance. Is divided by a value obtained by multiplying the value of the set Q table (Q table) by a Q scale (Q scale1) corresponding to one of eight types of Q scale codes, and a quotient A thereof is obtained. .

The Q scale code is represented by 5-bit data, and a predetermined one is assigned from a plurality of Q scales prepared in advance corresponding to the Q scale code. The pre-encoder 171 has Q scale1 through Q scale8
Has eight kinds of Q scales in total. In this case,
Regarding Q scale1 of these eight types of Q scales,
Processing is performed.

For example, the above operation for DCT1 is represented by the following equation.

DCT1 / (Q table × Q scale1) = A1 The same processing is performed on DCT2 to DCT4.
The quotients A2 to A4 are obtained.

The pre-encoder 171 adds the four quotients A1 to A4 obtained as described above, and calculates the sum S1
(= A1 + A2 + A3 + A4) is obtained.

A similar process is performed for the other seven types of Q scale2 to Q scale8. Then, the pre-encoder 171 calculates eight Q s obtained for each macroblock.
The code amounts S1 to S8 corresponding to cale1 to Qscale8 are compared with the target code amount R, and Q scalei corresponding to the code amount Si exceeding the target code amount closest to the target code amount R, and the code amount lower than the target code amount. Q scalej corresponding to Sj is selected and output to bit allocation section 172.

The bit allocation section 172 has two quantization values Q scale input from the pre-encoder 171.
Based on i and Q scalej, an optimum quantization value for the target is calculated for each macroblock, and output to the encoder 174 as a Q scale code.

The encoder 174 converts the video data Delayed Video delayed by the time necessary for the processing of the pre-encoder 171 and the processing of the bit allocation section 172 via the delay section 173 into the bit allocation section F.
Encoding is performed using the Q scale code input from 172.

As described above, the stream output from the encoder 174 is, as shown in FIG.
The code lengths of P1 to GOP8 are close to the target code length, and
It is controlled to a value that does not exceed the target code length, that is, a fixed length.

If the GOP exceeds one frame, the encoding process by the encoder 174 is performed using the motion prediction by the motion prediction processor 164.

The fixed-length SX Stream of the first format output from the fixed-length encoder 163 is input to the data converter 165 and converted into an MPEG Stream. Since the stream input to the data conversion device 165 is a fixed-length stream, the output stream is also a fixed-length stream. The data conversion device 165 is configured as shown in FIG. 2 and its operation is the same as in the above-described case, and the description thereof is omitted here.

FIG. 16 shows the fixed length MPEG shown in FIG.
14 shows another application example of the encoder. FIG. 16 shows an example of the editing system. In this editing system, video tape recorders (VTRs) 181 and 182 are connected to the MPEG disk system 200 as input devices, and a monitor 191 and a video tape recorder 192 are connected as output devices.
The video tape recorder 181 has an SDI (Serial Digital Interface) interface defined by SMPTE259M for outputting baseband video data.
SDTI (Seri) defined as PTE-305M
al Data Transport Interface). This SDTI is SDTI-CP (SDT
MPI on SDTI as I Contents Package)
G elementary stream (ES: Elementary Strea
m) can be output. The video tape recorder 182 is a fixed length encoder 1 shown in FIG.
A fixed length encoder 183 having the same configuration as that of the fixed length encoder 63 is provided. However, this fixed-length encoder 183
It also has a function of converting an EG stream into an SX stream.

The video tape recorder 192 has an SDTI-CP interface similarly to the video tape recorder 182, and incorporates a fixed length encoder 193 having the same function as the fixed length encoder 183.

The MPEG disk system 200 has an SDI receiving unit 201 functioning as an SDI interface for receiving baseband (uncompressed) video data from the video tape recorder 181. When receiving the video data SDI OUT 1 from the video tape recorder 181, the SDI receiving unit 201 converts the video data SDI OUT 1 from serial data into parallel data, further separates the data into video data and audio data, and converts the video data into a fixed length MPEG encoder 202. And supplies the audio data to the delay unit 204.

The fixed-length MPEG encoder 202 has the structure shown in FIG.
2, and generates a fixed-length MPEG stream from the video data input from the SDI receiving unit 201.
-A is output to the stream switch unit 203.

The delay unit 204 converts the audio data from the SDI receiving unit 201 into a fixed-length MPEG encoder 20.
2 is delayed by an amount corresponding to the processing time in the audio switch unit 2 as audio data Audio-A.
06.

[0099] The SDTI receiving unit 205 receives the compressed video data SDT from the video tape recorder 182.
The I-CP OUT1 is converted from serial data to parallel data, further separated into video data and audio data, and the video data is supplied to the stream switch unit 203 as MPEG-B, and the audio data is converted to audio data.
Output to the audio switch unit 206 as o-B.

The stream switch unit 203 has a fixed length M
One of the audio stream MPEG-A supplied from the PEG encoder 202 and the video stream MPEG-B input from the SDTI receiving unit 205 is selected according to the control from the system controller 208, and the stream REC-MPEG As disk 207
And record it. Similarly, audio switch unit 2
06 selects one of the audio stream Audio-A input from the delay unit 204 and the audio stream Audio-B input from the SDTI receiving unit 205 in accordance with the control of the system controller 208, and selects the audio stream REC- It is output to the disk 207 as Audio and recorded.

The disk 207 receives a control command from the system controller 208 via the Disk-Bus.
The stream switch unit 203 and the audio switch unit 206 receive commands from the system controller 208.
Receive via Syscon-Bus.

When the disk 207 receives a reproduction command from the system controller 208 via Disk-Bus,
Plays recorded video data, and uses PB-MPEG
SDTI transmission section 212 and MPEG decoder 209
To PB-Audi
Output as o to the SDTI transmission unit 212 and the delay unit 211. The MPEG decoder 209 decodes the video data PB-MPEG input from the disk 207 when it is compressed, and decodes the video data PB-MPEG as it is as the video data Dec-Video if it is not compressed.
Output to 0. The delay unit 211 delays the audio data PB-Audio input from the disk 207 by a time corresponding to the processing of the MPEG decoder 209, and supplies the audio data PB-Audio to the SDI transmission unit 210 as audio data DL-Audio.

The SDI transmission section 210 maps the video data input from the MPEG decoder 209 and the audio data input from the delay section 211 into an SDI format, converts parallel data into serial data, and converts the data into SDI-OUT2. Is output to the monitor 191.

The monitor 191 outputs the data SDI-OUT2
It is of course possible to connect devices other than the monitor here.

The SDTI transmission unit 212
After mapping the video data PB-MPEG and the reproduced audio data PB-Audio reproduced and output from the SDTI-CP format, the data is converted from parallel data to serial data, and the data SDTI-CPO is converted.
It outputs to the video tape recorder 192 as a UT.

Next, the operation will be described. SDI
The receiving unit 201 extracts video data and audio data from the baseband video data SDI OUT1 input from the video tape recorder 181 and outputs the video data to the fixed-length MPEG encoder 202.
After the audio data is delayed by a predetermined time by the delay unit 204, the audio data is output to the audio switch unit 206. The fixed-length MPEG encoder 202 generates a fixed-length MPEG stream from the input video data and outputs it to the stream switch unit 203, as in the case described with reference to FIG.

When the system controller 208 controls the stream switch unit 203 and selects the output MPEG-A of the fixed length MPEG encoder 202, the system controller 208 controls the audio switch unit 206 and controls the audio input from the delay unit 204. The data Audio-A is selected. As a result, the video stream MPEG-A selected by the stream switch unit 203 is supplied to the disk 207 as a video stream REC-MPEG, recorded, and the corresponding audio data Au.
dio-A is selected by the audio switch unit 206, supplied to the disk 207 as an audio stream REC-Audio, and recorded.

On the other hand, when receiving the supply of the data SDTI-CP OUT1 compressed by the fixed-length encoder 183 from the video tape recorder 182, the SDTI receiving unit 205 separates the video data and the audio data,
The video data is supplied to the stream switch unit 203 as MPEG-B, and the audio data is output to the audio switch unit 206 as Audio-B. The stream switch unit 203 and the audio switch unit 206 are controlled by the system controller 208, and the stream switch unit 203 controls the output MPE of the SDTI reception unit 205.
When GB is selected, the audio switch unit 206
Selects the output Audio-B of the SDTI receiving unit 205. Then, the selected video data is the video data RE.
The audio stream is supplied to the disk 207 as C-MPEG and recorded, and the selected audio stream is
The data is supplied to the disk 207 as C-Audio and recorded.

The audio stream Audio-A and the audio stream Audio-B input to the audio switch unit 206 are not compressed. This is because the quality of audio data is likely to degrade when compressed.
In order to stand up , audio data is not compressed and the quality as a broadcasting device is maintained.

As described above, when data is recorded on the disk 207, the user operates the system controller 2
08 is appropriately controlled to perform editing processing. At this time, since what is recorded is a disk, complicated editing can be performed as compared with a case where editing is performed by a video tape recorder.

When the user instructs reproduction, the system controller 208 controls the disk 207 to reproduce the data recorded therein. The reproduced video data PB-MPEG is transmitted to the MPEG decoder 209 and S
The audio data PB-Audio that is supplied to and reproduced from the DTI transmitting unit 212 is supplied to the delay unit 211 and the SDTI transmitting unit 212.

An MPEG decoder 209 decodes the input video data PB-MPEG and outputs the decoded data.
The Dec-Video is supplied to the SDI transmission unit 210. Delay unit 2
11 is input playback audio data PB-Audio
Is delayed by an amount corresponding to the processing time of the MPEG decoder 209, and S is converted to audio data DL-Audio.
Output to DI transmitting section 210. The SDI transmission unit 210
The input video data Dec-Video and audio data DL-Audio are mapped to an SDI format, and output as data SDI OUT2 to the monitor 191 for display.

Further, the SDTI transmission unit 212
Video data PB-MPEG and audio data PB-Audio reproduced from 00 are mapped to the SDTI format and output to the video tape recorder 192 as data SDTI-CP OUT2. The video tape recorder 192 converts the input video data into MP
Since the video has already been compressed by the EG method, the video is recorded on the video tape attached as it is without being further compressed by the fixed-length encoder 193.

In this manner, the video tape recorder 19
2 can be used for broadcasting in television broadcasting or the like.

In the video tape recorder 192, S
It is also possible to receive baseband data output from the DI transmission unit 210, compress the data by the fixed-length encoder 193, and record the data on a tape. However, in that case, MP
Since the data decoded by the EG decoder 209 is encoded again, the number of generations of the encoding and decoding increases, and the image quality deteriorates accordingly. Therefore, it is preferable to receive the data in the encoded state from the SDTI transmission unit 212 and record the data as it is on the tape.

As described above, since the output of the SDTI transmitting unit 212 is recorded as it is on the video tape, the fixed length M
If the PEG encoder 202 is replaced by a normal variable-length MPEG encoder, its output will include the amount of data that overshoots. All of the data is recorded on the disk 207.
When output from the TI transmission unit 212 and recorded on a tape by the video tape recorder 192, data for an overshoot exceeding the capacity of the helical track is deleted, and the image quality deteriorates. Therefore, by using this MPEG encoder as a fixed-length MPEG encoder, it is possible to prevent data overshooting from being deleted. Therefore, it is possible to perform recording and reproduction by performing lossless compression / decompression processing (without image quality deterioration).

Next, with reference to FIG. 17, a description will be given of a medium used to install a program for executing the above-described series of processing in a computer and to make the computer executable.

As shown in FIG. 17A, a program is stored in a hard disk 302 or a semiconductor memory 3 as a recording medium built in a personal computer 301.
03 in advance.

Alternatively, the program is executed as shown in FIG.
As shown in (B), a floppy (registered trademark) disk 311, a CD-ROM (Compact Disk-Read Only Disk) 31
2. MO (Magneto-Optical) disc 313, DVD (Digit
al Versatile Disk) 314, a magnetic disk 315, a semiconductor memory 316, or other such storage media, which can be temporarily or permanently stored and provided as package software.

Further, as shown in FIG. 17C, the program is wirelessly transferred from a download site 321 to a personal computer 323 via an artificial satellite 322 for digital satellite broadcasting, a local area network, and the Internet. Network 331
, And can be transferred to the personal computer 323 by wire, and stored in a built-in hard disk or the like in the personal computer 323.

The medium in the present specification means a broad concept including all these media.

In this specification, the step of describing a program provided by a medium is not limited to processing performed in chronological order according to the described order, but is not necessarily performed in chronological order. It also includes processes that are executed individually or individually.

In the present specification, the system is
It represents the entire device composed of a plurality of devices.

Next, effects obtained by the above-described embodiment of the present invention will be described.

The MP, which is a worldwide standard for digital video signals and is listed as the second format.
The EG standard does not consider a VTR (Video Tape Recorder) for broadcast business, for example, and therefore, when performing VTR recording in the same format, there are parts that are not necessarily optimized. In a VTR for a broadcasting business, an editing function is very important, and the quality of a search image for searching for an editing point is an important factor. In a VTR, since recorded data is recorded as a helical track pattern on a tape by a head on a rotating drum, all recorded data cannot be picked up at the time of high-speed reproduction, such as during a search. The arrangement of the DCT coefficients in the MPEG of the second format is independent for each of the eight DCT blocks (Y0 to Y3 and Cb0 to Cr1 in FIG. 4) in the macroblock. When recorded, DC (direct current) coefficients important for decoding a macroblock and low-order AC
(AC) coefficients cannot always be picked up for all eight DCT blocks, and high quality search images cannot be expected.

On the other hand, in the first format, each DC coefficient of eight DCT blocks is put together in units of a macro block, and each AC coefficient is put together in order from the lower order component into eight pieces and sequentially arranged. Is done. As a result, since eight DC coefficients and eight low-order AC coefficients in a macroblock can be collectively picked up at the time of high-speed reproduction such as a search, an image obtained by decoding the macroblock becomes easy to see. As described above, the first format is optimized for functions required in a broadcast business VTR or the like while using a compression algorithm equivalent to the world standard MPEG.

However, the first format is specialized for VTRs for broadcasting services and the like, and when data transmission with other devices is considered, the MPEG standard such as the second format is useful. It is. In this case, when transmitting the compressed and encoded data of the first format, it is conceivable to decode even the uncompressed original video data and perform compression and encoding in the second format. The circuit configuration and the processing amount are large. It cannot be realized without image quality deterioration.

Therefore, the encoded data of the first format is partially decoded up to variable length decoding to obtain D
The CT coefficients are obtained, the DCT coding is not decoded, the DCT coefficients are rearranged according to the second format, and only the variable-length coding is partially coded. MPEG encoded data stream of the format

As a result, it is possible to convert encoded data of the first format into encoded data of the second format with a simple circuit configuration and a small amount of processing.

Also, by incorporating this data conversion unit into a VTR, when data is transmitted between the VTR and an external MPEG device, decoding of encoded data of the first format and encoding of the second format are performed. An encoding process for data is not required, and an interface with no image quality deterioration is possible.

Further, the encoded data of the second format, which is variable in length according to the standard, is changed to fixed-length encoded data, so that video data compressed to a helical track can be recorded and reproduced without deteriorating image quality. Becomes possible.

The present invention is not limited only to the above-described embodiment. For example, the second format is not limited to the MPEG, and the first format is not limited to the above-described S format.
It is not limited to the X format. Also, it is possible to easily realize the conversion of the encoded data of the second format into the encoded data of the first format. In this case, the encoded data of the second format is converted into an orthogonal transform coefficient (for example, a DCT coefficient). ) Is obtained, and the orthogonal transform coefficients for the data in the second format obtained by decoding are rearranged, so that a DC component and an AC component are obtained for each block aggregate (for example, macroblock) in the second format. May be arranged in order from the lower order to the higher order, and the arrangement-transformed orthogonal transform coefficients may be encoded and encoded into encoded data of the first format.

Further, in FIG. 12, the fixed length encoder 163 and the data converter 165 are configured independently, but they may be configured integrally.

As described above, the encoding device according to claim 1, the encoding method according to claim 8, and the encoding device according to claim 9
According to the recording medium, the input data is orthogonally transformed and compression-encoded, and a block in which a plurality of orthogonally transformed blocks are put together.
The orthogonal transform coefficient for each
Fixed-length first files arranged in order from low to high
Format encoded data, and the converted
The fixed-length first format encoded data is directly
For each intersecting block, the orthogonal transform coefficient is a DC component,
Variable-length rules arranged in order from low to high AC components
Since the data is converted into encoded data of the second format, it is possible to record the input data on a recording medium having a limited capacity without deteriorating the quality. .

The data recording / reproducing apparatus according to claim 10 ,
A data recording / reproducing method according to claim 12 , and claim 1.
According to the recording medium described in No. 3 , the input video data is orthogonally transformed and compression-encoded, and a plurality of orthogonal transform blocks are input.
The orthogonal transform coefficient for each block aggregate
DC components and AC components are arranged in order from low to high,
Converting into encoded data of a fixed-length first format,
Encoding the transformed, fixed-length first format
The data is transferred to the orthogonal transform block for each orthogonal transform block.
The numbers are arranged in order from low to high DC components and AC components.
Of the second format, which is a variable length standard
Convert to data, record on recording medium, record on recording medium
Since the reproduced data is reproduced and output, even when the reproduced video data is recorded on a video tape or the like having a helical track, it is possible to prevent the image quality from deteriorating.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating control of a GOP code length by feedback.

FIG. 2 is a block diagram illustrating a schematic configuration of a data conversion device according to an embodiment of the present invention.

FIG. 3 is a diagram showing a hierarchical structure of a data stream of a first format.

FIG. 4 is a diagram showing a macroblock.

FIG. 5 is a diagram for explaining zigzag scanning and alternate scanning for DCT coefficients in a DCT block.

FIG. 6 is a diagram showing a DCT coefficient data string that has been subjected to zigzag scan and alternate scan.

FIG. 7 is a diagram illustrating a hierarchical structure of a data stream of a second format.

FIG. 8 is a diagram for explaining a relationship between a coefficient data string in a first format and an error.

FIG. 9 is a diagram for explaining a relationship between a coefficient data string in a second format and an error.

FIG. 10 is a diagram for explaining an operation of a main part of the data conversion device of FIG. 2;

FIG. 11 is a block diagram showing a schematic configuration of an embodiment of a data recording / reproducing apparatus according to the present invention.

FIG. 12 is a block diagram illustrating a configuration example of a fixed-length MPEG encoder.

13 is a block diagram illustrating a configuration example of a fixed-length encoder in FIG.

FIG. 14 is a diagram illustrating DCT coefficients of a macroblock.

FIG. 15 is a diagram illustrating control of a target code amount by feedforward.

FIG. 16 is a block diagram illustrating a configuration example of an MPEG disk system according to an embodiment of the present invention.

FIG. 17 is a diagram illustrating a medium.

[Explanation of symbols]

Reference Signs List 11 variable length decoding circuit, 12 conversion circuit, 13 variable length coding circuit, 14 header addition circuit, 15 stuffing circuit, 161 fixed length MPEG encoder, 162 prefilter, 163 fixed length encoder, 164 motion prediction processor, 165 data conversion device , 166 CPU, 171 pre-encoder, 172 bit allocation unit, 173 delay unit, 174 encoder, 202 fixed length MPEG
Encoder, 207 days

Claims (13)

(57) [Claims] 1. An encoding device for converting input data into encoded data of a format of a variable length standard, wherein the input data is orthogonally transformed and compression-encoded. Orthogonal transform coefficients are arranged in the order of low to high order of the DC component and the AC component, and include conversion means for converting into substantially fixed-length encoded data in the format of the variable-length standard , wherein the conversion means Converts the input data to orthogonal transform
A block that compresses and encodes multiple orthogonal transform blocks.
The orthogonal transform coefficient for each cluster is DC component and AC component
Fixed-length first format, arranged in order from low to high
First converting means for converting the encoded data into encoded data of a fixed format, and the fixed-length first data converted by the first converting means .
1 encoded data in the orthogonal transform block.
The orthogonal transform coefficient of the DC component and the AC component
The variable length standard, which is arranged in order from low to high.
A second format for converting into encoded data of a second format
An encoding device comprising: a conversion unit. 2. The first transforming means comprises : first encoding means for preliminarily encoding input data by using a plurality of quantized values; and first encoding means for encoding the input data by the first encoding means. Deciding means for deciding a quantization value for setting the code amount when the input data is coded to be equal to or less than the target code amount, based on the code amount of the input data, 2. The code according to claim 1 , further comprising: a second encoding unit that encodes input data based on the encoded value and outputs the encoded data as encoded data of the first format having a fixed length. Device. 3. The encoding apparatus according to claim 1 , further comprising a removing unit that removes a high frequency component of the input data and supplies the high frequency component to the first encoding unit. 4. The decoding device according to claim 1, wherein the second conversion unit decodes the coded data in the first format until the orthogonal transform coefficient is obtained, and the data in the first format obtained by the decoding unit. Arranging means for arranging the orthogonal transform coefficients for each of the orthogonal transform blocks in the second format in order from the lower order to the higher order of the DC component and the AC component; the transform coefficients is encoded, the encoding apparatus according to claim 1, characterized in that it comprises an encoding means for said second format encoded data. Wherein said decoding means is a variable length decoding unit, the encoding apparatus according to claim 4, wherein said coding means is a variable-length coding means. 6. The sign according to claim 4, further comprising a header adding means to the encoded data stream of the adds a header to the encoded data first format from said encoding means Device. 7. The orthogonal transform is a discrete cosine transform (DCT).
, And the said orthogonal transform block is a DCT block, the encoding apparatus according to claim 1, wherein the block assembly is macroblock. 8. A coding method for a coding apparatus for converting input data into coded data of a format of a variable length standard, comprising: Each time, the orthogonal transform coefficients are arranged in the order of low to high order of the DC component and the AC component, and include a conversion step of converting into substantially fixed length encoded data of the format of the variable length standard. The transforming step performs orthogonal transform on the input data.
Compression encoding, and a block in which a plurality of orthogonal
The lock assemblies each, the orthogonal transform coefficients are DC Ingredient, AC
A fixed length first, arranged in order from low to high order of the components
A first conversion step for converting into encoded data of a format
And converted by the processing of the first conversion step,
The fixed-length first format encoded data is directly
For each intersecting block, the orthogonal transform coefficient is a DC component,
The variable length, which is arranged in the order of low to high order of the AC component
To encoded data of the second format that is the standard
And a second transforming step . 9. A program for controlling an encoding device for converting input data into encoded data of a variable length standard format, comprising: orthogonally transforming the input data to perform compression encoding; For each transform block, a transform step of transforming the orthogonal transform coefficients into substantially fixed-length encoded data in the format of the variable-length standard, in which the DC component and the AC component are arranged in the order from low to high. wherein the said conversion step, the input data, and orthogonal transformation
Compression encoding, and a block in which a plurality of orthogonal
For each lock aggregate, its orthogonal transform coefficients are DC component, AC
A fixed length first, arranged in order from low to high order of the components
A first conversion step for converting into encoded data of a format
And converted by the processing of the first conversion step,
The fixed-length first format encoded data is directly
For each intersecting block, the orthogonal transform coefficient is a DC component,
The variable length, which is arranged in the order of low to high order of the AC component
To encoded data of the second format that is the standard
A second conversion step of performing
A recording medium on which a gram is recorded. 10. The input data, and orthogonal transformation and compression coding, every orthogonal transform block, the orthogonal transform coefficient, a direct current component, which is disposed from the lower order AC components in the order of higher order, the variable Conversion means for converting the data into substantially fixed-length encoded data in a format of a length standard; recording means for recording the fixed-length encoded data converted by the conversion means on a recording medium; and the recording medium. from recorded therein, comprising: a reproducing means for reproducing the coded data, reproduced by said reproducing means, on the standard, and output means for outputting the encoded data of the fixed length of the variable length format , Said transform means transforms the input video data into an orthogonal transform
Compression-encoded, and combined multiple orthogonal transform blocks
For each block aggregate, its orthogonal transform coefficient is
Fixed-length first, arranged from low to high order of flow components
First conversion method for converting into encoded data in the format
And a stage having the fixed length converted by the first converting means.
1 encoded data in the orthogonal transform block.
The orthogonal transform coefficient of the DC component and the AC component
The variable length standard, which is arranged in order from low to high.
A second format for converting into encoded data of a second format
A data recording / reproducing device, comprising:
Place. 11. The data recording / reproducing apparatus according to claim 10 , wherein said recording medium is a disk. 12. The input data, and orthogonal transformation and compression coding, every orthogonal transform block, the orthogonal transform coefficient, a direct current component, which is disposed from the lower order AC components in the order of higher order, the variable A conversion step of converting the encoded data into a substantially fixed-length encoded data in a format of a length standard; a recording step of recording the encoded data converted by the processing of the conversion step on a recording medium; and A reproducing step of reproducing the fixed-length encoded data recorded on the recording medium in the processing , wherein the converting step converts the input video data into orthogonal data.
Transform and compress and encode multiple orthogonal transform blocks
The orthogonal transform coefficient of each block set
Fixed length, arranged from low to high order of AC component
To convert to encoded data of a first format of
A conversion step, the conversion performed by the processing of the first conversion step,
The fixed-length first format encoded data is directly
For each intersecting block, the orthogonal transform coefficient is a DC component,
The variable length, which is arranged in the order of low to high order of the AC component
To encoded data of the second format that is the standard
A data conversion / reproduction method. 13. The variable data in which input data is orthogonally transformed and compression-coded, and for each orthogonal transform block, the orthogonal transform coefficients are arranged in the order of low-order to high-order DC component and AC component. A conversion step of converting the encoded data into a substantially fixed-length encoded data in a format of a length standard; a recording step of recording the encoded data converted by the processing of the conversion step on a recording medium; and A reproducing step of reproducing the fixed-length encoded data recorded on the recording medium in the processing , wherein the converting step converts the input video data into orthogonal data.
Transform and compress and encode multiple orthogonal transform blocks
The orthogonal transform coefficient of each block set
Fixed length, arranged from low to high order of AC component
To convert to encoded data of a first format of
A conversion step, the conversion performed by the processing of the first conversion step,
The fixed-length first format encoded data is directly
For each intersecting block, the orthogonal transform coefficient is a DC component,
The variable length, which is arranged in the order of low to high order of the AC component
To encoded data of the second format that is the standard
A second conversion step of performing
A recording medium on which a gram is recorded.