JPH08140041A - Digital video signal recorder and reproducing device - Google Patents

Abstract

PURPOSE: To perform encoding for reducing a minimum code amount to be accessed at the time of special reproduction and to efficiently arrange minimum data to be accessed at the time of special reproduction on a disk by arraying respective output data outputted from first and second encoding means at prescribed positions in respective picture group data for the respective picture group data. CONSTITUTION: The code processing of I-pictures, P-pictures and B-pictures is performed for both low resolution components and resolution residual components. In such a manner, sequences in which the low resolution components and the resolution residual components are alternately arranged can be constituted. When the data are rearranged to collect and arrange GOPs at a part convenient for the access of an optical head by a data reconstituting device 25 and a first half which is an area occupied by L components is read, the low resolution components can be reproduced. For the resolution residual components, a data amount is less than non resolution residual components and efficient hierarchization can be performed. That is, the first and second encoding means are provided and the efficient hierarchization is performed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital video signal recording apparatus and a reproducing apparatus for recording or reproducing a digital video signal, and more particularly to recording and reproducing a digital video signal coded based on motion compensation prediction and orthogonal transformation. The present invention relates to a digital video signal recording device and a reproducing device.

[0002]

2. Description of the Related Art A conventional digital video signal recording / reproducing apparatus is disclosed in, for example, JP-A-6-98314 and JP-A-6-7.
8289 and the like, an example of which is shown in FIG. In the figure, 75 is a video signal generator such as a camera or VTR, 76 is an audio signal generator such as a microphone or VTR, 62 is a video signal encoder, 63 is an audio signal encoder, 77 is a system layer bit stream generator, 78 Is an error correction encoder, 79 is a digital modulator, 80 is an optical disk, 56 is a reproduction amplifier, 86 is a wave detector, 81 is a digital demodulator, 58 is an error corrector, 59 is a system stream processor, and 82 is a video signal. Decoder, 83 is an audio signal decoder, 84 is a monitor, and 85 is a speaker.

At present, the most common type of optical disc is 1.
It has a diameter of 20 mm.
It is possible to record data of over 0 Mbytes. Most recently, there has been a product capable of recording a video signal and an audio signal for 74 minutes at a data rate of about 1.2 Mbps on this optical disk. At the time of recording, the video signal is generated from the video signal generator 75. The audio signal generator 76 inputs the audio signal to the audio signal encoding means 63 and encodes the audio signal. The system layer bit stream generator 77 performs a process of multiplexing a header or the like on these two encoded signals, further adds an error correction code by an error correction code adder 78, and then performs a digital modulation by a digital modulator 79. Then, a bitstream for recording is generated. A mother disk is created by a recording means (not shown) of this bit stream, and the contents of this mother disk are duplicated on the disk 80 to create a commercial video soft disk.

In the reproducing apparatus of the user, the signal obtained by the optical head from the video soft disk is amplified by the reproducing amplifier 56 and the reproduced signal is inputted to the detector 86. This reproduction signal is detected by the detector 86,
The digital demodulator 81 performs digital demodulation, and the error corrector 58 performs error correction. Thereafter, a video signal area is extracted from the error-corrected signal, the extracted data is decoded by a video signal decoder 82, and the result is monitored together with the result of decoding the audio signal by the audio signal decoder. It outputs to 84 and the speaker 85.

A typical method of encoding this video signal is the MPEG (Mobi) which is an international standard encoding method.
ng Picture Experts Group)
There are MPEG1 and MPEG2 called schemes. A specific example of the encoding method will be described below by taking MPEG2 as an example. MPEG has been introduced in "MPEG Technology" (supervised by Jun Kome, Trikeps Co., Ltd.).

FIG. 30 is a block diagram of a video signal coding unit in a conventional digital signal recording / reproducing apparatus for explaining the MPEG2 coding system, and FIG. 32 is a conventional digital signal recording for explaining the decoding system. -A block diagram of a video signal decoding unit in the reproducing apparatus is shown. 31 is a diagram showing a concept of moving image processing of video signal encoding in a conventional digital signal recording / reproducing apparatus for explaining grouping of moving images by the MPEG2 encoding method. In FIG. 31, IBBPBBP ... , I is an I picture, B is a B picture, and P is a P picture, respectively (hereinafter, I in the drawings is I unless otherwise specified).
Picture, B shows B picture, and P shows P picture, respectively. ). For example, in (a) of FIG. 31, moving images from I to the appearance of the next I are grouped by a certain number. The number of pictures constituting this group is usually 15 in many cases, but is not particularly limited.

The group of images constituting this group is a group of images that can be reproduced independently, or a group of n frames, or a group of consecutive images, or a group of pictures (Group of P).
image. Hereinafter referred to as an image group or GOP). This GOP has at least one intra-frame coded screen (hereinafter referred to as an I picture) that can be decoded with only one image. In addition, motion compensation is performed based on the I picture by unidirectional prediction of the time series. An inter-frame predictive coding screen (hereinafter referred to as a P picture) for predicting and coding,
A bidirectional predictive coding screen (hereinafter referred to as B, based on the I picture and the P picture, which is coded by bidirectional prediction of the time series.
It is called a picture). (Incidentally, the arrow in FIG. 31 represents the prediction relationship)

That is, the B picture can be coded and decoded only after the I and P pictures have been prepared, and the first P picture in the GOP can be coded and decoded by the preceding I picture. Become. The second and subsequent P-pictures in the GOP can be coded and decoded together with the P-picture immediately before the second P-picture. Therefore, if there is no I picture, P, B
Both pictures cannot be coded or decoded.

In FIG. 30, reference numeral 87 is an image rearranger, and 8
8 is a scan converter, 89 is an encoder buffer, 90 is a mode determiner, 2 is a motion vector detector, 6 is a subtractor, 8
Is a DCT calculator, which has a field memory, a frame memory, and a DCT calculator. 10 is a quantizer, 14 is an inverse quantizer, 16 is an inverse DCT calculator (IDC
T is a notation), 18 is an adder, 20 is an image memory, 22 is a rate controller, and 26 is a variable length encoder.

In FIG. 32, 33 is a variable length decoder (VL
D), 34 is an inverse DCT, 37 is an image memory, 38
Is an adder and 39 is an inverse scan converter. The motion vector detector 2 and the mode determiner 90 will be collectively referred to as a motion vector detector in the future.

Next, the operation of a digital video signal recording / reproducing signal using a conventional optical disk or the like will be described with reference to FIGS. In FIG. 30, the image rearrangement unit 87 rearranges the images for encoding in the order shown in FIG. 31, and the scan converter 88 converts the raster scan into the block scan. Hereinafter, the image rearrangement process and the raster scan to block scan conversion process are collectively referred to as a pre-process, and the image rearranger 87 and the scan converter 88 are collectively referred to as a pre-processor (see FIG. 30). (The part marked PR). The input image data is block-scanned in the encoding order, and if it is an I picture, the subtractor 6 is passed through and P
For pictures and B pictures, the reference image and subtractor 6
Is subtracted by.

At this time, by the motion vector detector 2 of the image (the input to the vector detector 2 may be the image after the image rearrangement or the image after the block scan as the original image, the latter is the circuit. Smaller,
Further, the reference image must also be input from the image memory 20, but the reference arrow in the figure is omitted.) The direction and amount of movement are obtained, and the signal of the area in consideration of the amount is read from the image memory 20. At this time, the mode determiner 90 determines whether to use bidirectional prediction or unidirectional prediction.

The subtraction with the reference screen in consideration of the motion vector is performed by the subtractor 6 to form an image with low power, and the coding efficiency is improved. The output of the subtracter 6 is collected by the DCT calculator 8 in field units or frame units and subjected to discrete cosine transform (DCT) to be converted into frequency component data. This data is input to the quantizer 10 having different weights for each frequency, and the variable-length encoder 26 outputs 2 over the low-frequency component and the high-frequency component.
It is dimensionally zigzag scanned and subjected to run-length coding and Huffman coding.

The run length coded and Huffman coded data is controlled to scale the quantization table using the rate controller 22 so as to match the target code amount, and the variable length code is controlled. Are subject to change. This variable length coded data is usually output via an encoder buffer. The quantized data is returned to the original by the inverse quantizer 14, and the inverse DCT calculator 1
6 returns the original image space data and the data referred to by the subtracter 6 is added by the adder 18 to obtain the same data as when decoded.

FIG. 32 shows a schematic block configuration of a decoding apparatus. The variable length decoder 33 decodes image data including header information such as motion vectors, coding modes, picture modes, etc. After the decoded data is quantized, an inverse DCT operation is performed by the inverse DCT operation unit 36 (note that the inverse quantizer in the preceding stage of the inverse DCT operation unit 36 is omitted in FIG. 32) to consider the motion vector. Then, the image data is referenced from the image memory 37, and the referenced image data is added to the data after the inverse DCT by the adder 38 to decode the motion compensation prediction.
The inverse scan converter 39 converts this data into a raster scan to obtain an interlaced image and outputs it.

In addition, according to the variable transfer rate disk system introduced in "Variable transfer rate disk system and its code amount control method" in the presentation by Sugiyama et al. A video signal encoding method has been proposed. A method is proposed in which the coding rate of GOP is fixed by one program (for example, one program), and each GOP therein is coded by setting the rate according to the difficulty of the picture. FIG. 33 shows a block diagram of a video signal encoding unit in a conventional digital signal recording / reproducing apparatus according to this system. In FIG. 33, 91 is a motion compensation predictor, 92 is a code amount memory, 93 is a GOP rate setter,
Reference numeral 94 is a code amount assigner, 95 is a subtractor, 96 is a code amount counter, and 97 is a switch. The GOP rate setting device shown in FIG. 33 changes the setting of the quantized value depending on the complexity of the picture. That is, while the switch 97 is connected to the temporary encoding side, the output of the variable length encoder 26 is input to the code amount counter 96, the code amount is counted by the code amount counter 96 and stored in the code amount memory 92. To be

The GOP rate setter 93 obtains a provisional code amount for all one programs from the value of the code amount stored in the code amount memory 92, and calculates and sets the optimum coding rate of each GOP. The code allocation at this time is calculated by the code amount allocation unit 94, and the preparation for the main encoding is completed. Switch 9
When 7 is connected to the main encoding side, the code amount allocation amount output from the code amount allocation unit 94 is compared with the value of the code amount counter 96, and the quantizer is controlled based on the actual code amount. To work. By doing so, a small amount of code is assigned to an easy pattern and a large amount of code is assigned to a complicated pattern so as to absorb the difficulty of encoding which changes slowly in the program. As a result, it is reported that the image quality recorded by this system at a rate of 3 Mbps was almost the same as that obtained by encoding by the conventional system at a rate of 6 Mbps.

Considering the possibility of skip search in the digital video signal recording / reproducing apparatus using the optical disc as described above, even if the head of the GOP can be accessed at high speed, for example, I-picture and P-picture are provided for fast-forwarding. If you want to play
Since it is written in an appropriate position in the inside, it becomes necessary to operate the optical head while searching for data on the bit stream. However, such control cannot be done in time due to the time constant of the servo of the actuator or the like. Certain GO
In order to find the beginning of P, an image of 15 frames is usually included per 1 GOP, and although there is a possibility of searching because there is a margin of 0.5 seconds in the NTSC scanning method, for example, I picture and P picture during skip search. However, if the head movement time is 200 ms, the bit stream for reading an image that is one-third of the frame rate requires 2.5 times the normal time. The above reading speed is required. This exceeds the response limit of the actuator, and skip search is practically impossible by a proper reproduction method.

[0019]

Since the conventional digital signal recording / reproducing apparatus is encoded in this way, if a skip search (seeing in fast-forward) is attempted like a video tape recorder, the B When reproducing data such as a picture that cannot completely restore the original image from one image data, a proper reproduced image cannot be obtained, and especially in skip search, jerkiness related to output processing in frame units. (Unnatural movement)
Occurs, or when variable rate coding with good playback quality is performed, the position of the head address of the GOP changes, which makes it difficult to access the head of the GOP itself, and the GOP unit is not aligned, and the disk area is not aligned. There is a problem that there is a blank in.

The present invention has been made in view of the above problems, and skips in a digital video signal recording / reproducing apparatus for recording / reproducing a digital video signal encoded by using motion compensation prediction and orthogonal transformation. (EN) A digital video signal recording / reproducing apparatus capable of searching and improving accessibility to a GOP and effectively utilizing an empty area on a storage medium on the assumption that variable rate coding is adopted. The purpose is to

[0021]

According to a first aspect of the present invention, there is provided a digital video signal recording apparatus for recording a digital video signal encoded by using motion compensation prediction and orthogonal transformation on a recording medium. A first encoding means for encoding a video signal composed of a coded image including at least an intra-frame coded image of the digital video signals coded using the motion compensation prediction and orthogonal transformation according to a predetermined condition. A second coding means for coding a residual component of the video signal by the coding using the first coding means, and the first and second coding means. Data array means for arranging each output data to be output at a predetermined position in each image group data for each image group data.

According to a second aspect of the present invention, there is provided a digital video signal recording apparatus having the structure of the first aspect,
The first encoding means is configured to encode the video information thinned out at a predetermined interval with respect to the video signal composed of the encoded image including at least the intra-frame encoded image.

According to a third aspect of the present invention, there is provided a digital video signal recording apparatus having the configuration according to the first aspect of the invention.
The first encoding means is configured to encode only the low frequency region that has been orthogonally transformed.

According to a fourth aspect of the present invention, there is provided a digital video signal recording apparatus having the structure of the first aspect,
The first coding means is configured to roughly quantize and code at the quantization level.

According to a fifth aspect of the present invention, there is provided a digital video signal recording apparatus for recording a digital video signal encoded by using motion compensation prediction and orthogonal transformation on a recording medium. And low-frequency region data is extracted from a data string in which a video signal composed of a coded image including at least an intra-frame coded image of a digital video signal coded using orthogonal transformation is divided into predetermined bits. And a low frequency region extracting means.

In the digital video signal reproducing apparatus according to the invention of claim 6, data is encoded by using motion compensation prediction and orthogonal transformation, and the low frequency region data and the high frequency region data are divided into regions. In a digital video signal reproducing device for reproducing a digital video signal from a recording medium, a data rearranging means for rearranging the low frequency region data and the high frequency region data in a predetermined order, and the rearranged data. And a mode switching means for selecting one of a decoding mode and a mode for selectively decoding the data in the low frequency region.

According to a seventh aspect of the present invention, there is provided a digital video signal recording apparatus having the structure of the sixth aspect,
Furthermore, when decoding in the mode that only decodes data in the low frequency region, only the decodable data is decoded, undecodable data near the boundary of the specified number of bits is discarded, and the obtained high frequency region The data has a data manipulating means for substituting fixed values for inverse orthogonal transformation.

According to the eighth aspect of the present invention, there is provided a digital video signal recording apparatus for recording a digital video signal encoded by using motion compensation prediction and orthogonal transformation on a recording medium. A predetermined number of bits as low frequency region data for the coded data of each block of the video signal consisting of the coded image including at least the intra-frame coded image of the digital video signal coded using When it reaches, the block end code is added, and the above-mentioned coded data exceeding the predetermined number of bits is coded as high frequency area data.

In the digital video signal recording apparatus according to the invention of claim 9, the data in the low frequency region and the data in the high frequency region encoded based on the motion compensation prediction and the orthogonal transformation are partitioned by the block terminating code. In a digital video signal reproducing apparatus for reading encoded data from a recording medium, a data reconstructing means for reconstructing data based on each of low-frequency region data, high-frequency region data and block end code is reconstructed. Mode switching means for selecting one of a data decoding mode and a mode for selectively decoding only low-frequency region data, and encoded data reconstructed based on the output of the mode switching means Decoding means for decoding, and a data operator that performs inverse orthogonal transformation by replacing fixed values in the high frequency region. And it has a door.

According to a tenth aspect of the present invention, there is provided a digital video signal recording apparatus for recording on a recording medium a digital video signal encoded by using motion compensation prediction and orthogonal transformation. And encoding means for encoding low-resolution component data obtained by thinning pixels of a video signal formed of a coded image including at least an intra-frame coded image of a digital video signal coded using orthogonal transformation And a difference component coding means for coding the difference component between the output data of the coding means and the image before pixel thinning, and the output of the coding means and the difference component coding means. The output is divided into a predetermined area to form data, and an information adding means for adding an audio signal, a header and the like and adding an error correction code is provided.

According to the eleventh aspect of the present invention, there is provided a digital video signal recording apparatus, wherein low resolution data which is coded on the basis of motion compensation prediction and orthogonal transformation and which has been thinned out of pixels, an image before thinning out of pixels and thinning out of pixels. In the digital video signal reproducing device for reading out from the recording medium a digital video signal having data corresponding to the differential component of the image after being interpolated, the data of the low resolution component and the data of the differential component are combined. It has a decoding means for decoding.

The digital video signal recording apparatus according to the twelfth aspect of the present invention has the configuration according to the eleventh aspect of the present invention, and further has a mode in which low-resolution component data and differential component data are combined and decoded. It has a mode switching means for switching between a mode for decoding only the resolution component.

The digital video signal recording apparatus according to the thirteenth aspect of the present invention has the configuration according to the eleventh aspect of the present invention, and further, when decoding a low resolution image, generates an interpolated video generating means for generating an interpolated video after decoding. Is to have.

In the digital video signal recording apparatus according to the fourteenth aspect of the present invention, when the digital video signal encoded using the motion compensation prediction and the orthogonal transformation is recorded on the recording medium, the motion compensation prediction and the orthogonal transformation are performed. Pattern determining means for determining the degree of image deterioration at the time of encoding and decoding based on the above, an adaptive encoder for adaptively changing and encoding the data rate based on the determination output from the pattern determining means, and an audio signal , A digital video signal recording device having additional information such as a header and information adding means for adding an error correction code, having a data rate setting means for setting a discrete value to a data rate to be adaptively changed Is.

In the digital video signal recording apparatus according to the fifteenth aspect of the present invention, when the digital video signal encoded using the motion compensation prediction and the orthogonal transformation is recorded on the recording medium, the motion compensation prediction and the orthogonal transformation are performed. Pattern determining means for determining the degree of image deterioration at the time of encoding and decoding based on the above, an adaptive encoder for adaptively changing and encoding the data rate based on the determination output from the pattern determining means, and an audio signal In a digital video signal recording apparatus having additional information such as a header and information adding means for adding an error correction code, the data rate information is multiplexed on the header or written in a predetermined area on the recording medium. It is a thing.

In the digital video signal recording apparatus according to the sixteenth aspect of the present invention, when the digital video signal encoded using the motion compensation prediction and the orthogonal transformation is recorded on the recording medium, the motion compensation prediction and the orthogonal transformation are performed. At least in a digital video signal recording device having a picture pattern judging means for judging the degree of image deterioration at the time of encoding and decoding based on the above, and an information adding means for adding additional information such as an audio signal and a header and an error correction code. First coding means for coding a video signal thinned out at a predetermined interval with respect to a video signal composed of a coded image including an intra-frame coded image, and the first coding of the video signal A second encoding means for encoding the residual component by means of encoding using a means, and based on the determination output from the pattern determining means. Serial and the data rate adaptively changed in at least one of the coding means of the first and second encoding means is obtained by a structure for encoding.

In the digital video signal recording apparatus according to the seventeenth aspect of the present invention, when the digital video signal encoded by using the motion compensation prediction and the orthogonal transformation is recorded on the recording medium, the motion compensation prediction and the orthogonal transformation are performed. At least in a digital video signal recording device having a picture pattern judging means for judging the degree of image deterioration at the time of encoding and decoding based on the above, and an information adding means for adding additional information such as an audio signal and a header and an error correction code. A first encoding unit that encodes only a low-frequency region that has been orthogonally transformed with respect to a video signal that includes an encoded image that includes an intra-frame encoded image; and the first encoding unit that is one of the video signals. A second coding means for coding the residual component by the used coding, and based on the judgment output from the pattern judgment means, It is obtained by a structure for coding adaptively changing the data rate in at least one of the encoding means of the first and second encoding means.

In the digital video signal recording apparatus according to the eighteenth aspect of the present invention, when the digital video signal encoded using the motion compensation prediction and the orthogonal transformation is recorded on the recording medium, the motion compensation prediction and the orthogonal transformation are performed. At least in a digital video signal recording device having a picture pattern judging means for judging the degree of image deterioration at the time of encoding and decoding based on the above, and an information adding means for adding additional information such as an audio signal and a header and an error correction code. A first coding means for roughly quantizing and coding a video signal composed of a coded image including an intra-frame coded image at a quantization level; and the first coding means of the video signal Second encoding means for performing encoding on the residual component by the above-mentioned encoding, and based on the determination output from the above-mentioned pattern determining means, It is obtained by a structure for coding adaptively changing the data rate in at least one of the encoding means of the first and second coding means.

In the digital video signal recording apparatus according to the nineteenth aspect of the present invention, data in which the data rate of the encoded data encoded based on the motion compensation prediction and the orthogonal transformation is adaptively variable according to the picture is read. In a digital video signal reproducing apparatus, a mode switching means for switching a reproduction mode between a normal reproduction mode and a special reproduction mode, a data rate information extracting means for extracting data rate information, and the data rate information extraction in the special reproduction mode. Position information calculation means for calculating the position on the recording medium at which the data for special reproduction exists based on the data rate information output from the means.

The digital video signal recording apparatus according to the twentieth aspect of the invention has the configuration according to the nineteenth aspect of the invention, and further, the head is located at a position on the recording medium according to the output from the position information calculating means and the special reproduction speed. It has a head position control means for controlling the position.

According to the digital video signal recording apparatus of the present invention, in the digital video signal recording apparatus for recording a digital video signal encoded by using motion compensation prediction and orthogonal transformation on a recording medium, the motion compensation prediction is performed. And encoding means for controlling and encoding the amount of code corresponding to an area assigned to one image group formed by a digital video signal encoded based on And a data supplementing means for embedding extra data in a blank area of an image group having a blank area based on the output from the code quantity comparing means.

In the digital video signal recording device according to the twenty-second aspect of the present invention, another image is encoded in the blank area of the image group which is encoded based on the motion compensation prediction and the orthogonal transformation and is formed based on this encoded data. In a digital video signal reproducing apparatus for reading digital video signal data in which group data is embedded from a recording medium, data reconstruction means for reconstructing the embedded video signal coded data into an original image group, and the data reconstruction means. Data decoding means for decoding the reconstructed data from the configuring means.

In the digital video signal recording apparatus according to the twenty-third aspect of the present invention, the encoding is performed based on the motion compensation prediction and the orthogonal transformation, and the first and second encoding is performed at a predetermined position of each image group data. In the digital video signal reproducing device for generating first and second decoded data corresponding to the first and second encoded data from the arranged data according to a predetermined condition, the first and second digital video signal reproducing devices. Decoding means for decoding the above data to obtain a reproduced image, and reproducing corresponding to the low frequency region of the intra-frame coded image or the number of thinned pixels or coarse quantization Second decoding means for obtaining an image and at least the low frequency region of the intra-frame coded image or the inter-frame predictive coded image by decoding the first data or the number of thinned pixels or Among the third decoding means for obtaining a reproduced image corresponding to a certain quantization, at least one decoding means is provided, and which decoding means is used for special reproduction is switched based on the special reproduction speed. It has a mode switching means for.

In the digital video signal recording device according to the twenty-fourth aspect of the present invention, the digital video signal reproducing device for reproducing the video information coded based on the motion compensation prediction and the orthogonal transformation from the recording medium, Video data extraction means for extracting data corresponding to the video signal,
Video data decoding / reproducing means for decoding and reproducing the video data output from the video data extracting means, and a normal reproduction mode when reproducing the video data, or reproducing either one of an odd field and an even field It has a mode switching means for switching between a display mode and a mode in which the field structure of the odd field or the even field is reversed to perform the reproduction display.

In the digital video signal reproducing apparatus according to the twenty-fifth aspect of the present invention, the encoding is performed based on the motion compensation prediction and the orthogonal transformation, and the first and second encoding is performed at a predetermined position of each image group data. An array of data is made,
A digital video for generating first and second decoded data corresponding to the first and second encoded data from data whose encoded data rate is adaptively variable according to a picture according to a predetermined condition. In the signal reproducing device, a first decoding means for decoding the first and second data to obtain a reproduced image, and a low frequency region or thinning of the intra-frame encoded image by decoding the first data. Second decoding means for obtaining a reproduced image corresponding to the number of pixels or coarse quantization, and at least the intra-frame coded image or the inter-frame predictive coded image in the low frequency region or thinning out by decoding the first data. At least one of the third decoding means for obtaining a reproduced image corresponding to the number of pixels or rough quantization is used, and any of the above decoding means is used for special reproduction. Or the one having a mode switching means for switching based on the trick play speed.

[0046]

According to the digital video signal recording apparatus of the first aspect of the present invention, the first encoding means encodes a basic portion of a moving image as compared with encoding all video signals. By doing so, it works to reduce the minimum access area. The second coding means acts so as to code the video information that has not been coded by the first coding means, and code all the video information with the data of the two coding means. Further, the data arranging means acts to rearrange the data obtained by the two encoding means so that the data can be conveniently accessed by the head.

According to the digital video signal recording apparatus of the second aspect, the first coding means functions to code the spatially thinned video data and reduce the minimum access area.

According to the digital video signal recording apparatus of the third aspect, the first coding means functions to code the video data that is partial in terms of frequency and reduce the minimum access area.

According to the digital video signal recording apparatus of the fourth aspect, the first encoding means coarsely quantizes the data to encode the upper bit data which has a great influence on the video, so that the minimum access is possible. It acts to reduce the area to be used.

According to the digital video signal recording apparatus of the fifth aspect, the data extraction means functions to partition the low frequency region of the video signal by partitioning each block into predetermined bits.

According to the digital video signal reproducing apparatus of the sixth aspect, the normal reproducing data rearranging means operates so as to connect the coded data divided into two parts to obtain a complete decoded image, At the time of special reproduction, it acts so that only the data in the low frequency region of the encoded data divided into two is decoded, and so that it is decoded according to the operating condition of the device.

According to the digital video signal reproducing apparatus of the present invention, in the special reproduction, when the data operating means decodes the low frequency region of the coded data divided into two, the decodable data Only the bits are decoded, the bits that cannot be decoded are discarded, and the remaining high-frequency regions of the block are replaced with fixed values for decoding.

According to the digital video signal recording apparatus of the eighth aspect, both the low frequency region and the high frequency region of the block are coded by the block end code (EOB) so that the block is apparently terminated.

According to the digital video signal reproducing apparatus of the ninth aspect, the normal reproducing data reconstructing means operates so as to obtain a complete decoded image from the encoded data delimited by the block end code (EOB). On the other hand, during special reproduction, it acts so as to decode only the data in the low frequency region of the encoded data, and acts so as to decode it according to the operating condition of the device. Further, when the data operating means decodes the low frequency region of the coded data, it operates so as to replace the remaining high frequency region of the block with a fixed value for decoding. E in both low and high frequency regions of the block
The OB acts to encode the block as if it had ended.

According to the digital video signal recording apparatus of the tenth aspect, the low resolution coding means codes the spatially thinned video data, and the differential component coding means,
The decoded data in the low resolution encoding means is interpolated and compared with the image before the low resolution conversion to obtain the difference component, and the high resolution video data not obtained by the low resolution encoding means is obtained. It acts to encode.

According to the digital video signal reproducing apparatus of the eleventh aspect, during normal reproduction, the video signal decoding means includes the low resolution encoded data, the data before thinning out to the low resolution and the low resolution data. It acts so as to synthesize and decode the encoded data of the high resolution component of the difference component of.

According to the digital video signal reproducing apparatus of the twelfth aspect, the mode switching means, during normal reproduction,
The low-resolution coded data divided into two and the high-resolution coded data of the difference component between the data before thinning out to the low resolution and the low-resolution data are combined and decoded, and during special reproduction, It acts to switch the decoding means so that only the low resolution encoded data is decoded.

According to the digital video signal reproducing apparatus of the thirteenth aspect, when the video generating means decodes only the low resolution encoded data during the special reproduction, the video data of the low resolution component is interpolated and the original image is reproduced. It acts to return to size.

According to the digital video signal recording apparatus of the fourteenth aspect, the data rate setting means operates so that the rate of the variable rate video signal encoding means is limited to a very limited value.

According to the digital video signal recording apparatus of the fifteenth aspect, the data rate setting information when encoded at a variable rate is recorded on the recording medium separately from the video data.

According to the digital video signal recording apparatus of the sixteenth aspect, the encoding means operates so as to realize high-quality encoding by the variable rate, and also in the GOP in which the rate is greatly increased by the variable rate. , It encodes spatially thinned video data and acts to reduce the minimum access area.

According to the digital video signal recording apparatus of the seventeenth aspect, the encoding means operates so as to realize the high quality encoding by the variable rate, and also in the GOP in which the rate is greatly increased by the variable rate. , It encodes partial video data in terms of frequency and acts to reduce the minimum access area.

According to the digital video signal recording apparatus of the eighteenth aspect, the encoding means operates so as to realize the high-quality encoding by the variable rate, and also in the GOP in which the rate is greatly increased by the variable rate. By coarsely quantizing, high-order bit data, which has a great influence on the video, is coded, and acts to reduce the minimum access area.

According to the digital video signal reproducing apparatus of the nineteenth aspect, the data rate information extracting means is provided for each GO.
When the data rate information of P is extracted to reproduce GOPs having different data rates, both of the coded data divided into two are combined and decoded at the time of normal reproduction, and the GOP of the access target is made at the time of special reproduction. Calculate the position on the recording medium, play back the minimum accessed data, and
Access OP. At this time, the position information calculating means operates from the obtained data rate information to calculate the position information on the recording medium having the GOP to be accessed.

According to the twentieth aspect of the digital video signal reproducing apparatus, the position information on the recording medium having the GOP to be the access target is calculated according to the special reproduction speed, and the optical head position converting means sets the position information. It acts to control the position of the optical head.

According to the digital video signal recording apparatus of the twenty-first aspect, even when the data is encoded and recorded at a variable rate, the GOP is arranged at a position where the head can easily access, and a blank portion such as a blank portion is filled as much as possible. Acts like.

According to the digital video signal reproducing apparatus of the twenty-second aspect, the coded data in which the data of another GOP is embedded in the blank area functions to reconstruct the coded data of the original GOP.

According to the digital video signal reproducing apparatus of the twenty-third aspect, it operates so as to switch the mode for decoding and displaying only the I picture and the mode for displaying the I and P pictures according to the speed of special reproduction. .

According to the digital signal reproducing apparatus of the twenty-fourth aspect, the field structure is optimized according to the mode during the special reproduction.

According to the digital video signal reproducing apparatus of the twenty-fifth aspect, the data rate information extracting means is provided for each GO.
When the data rate information of P is extracted to reproduce GOPs having different data rates, both of the coded data divided into two are combined and decoded at the time of normal reproduction, and the GOP of the access target is made at the time of special reproduction. Calculate the position on the recording medium, play back the minimum accessed data, and
Access OP. At this time, the position information calculating means operates from the obtained data rate information to calculate the position information on the recording medium having the GOP to be accessed. Also, it operates so as to switch the mode for decoding and displaying only the I picture and the mode for displaying the I and P pictures according to the speed of special reproduction.

[0071]

【Example】

Example 1. Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a digital video signal encoding processing unit in a digital video recording / reproducing apparatus. In the figure, 1 and 4 are preprocessors (PR),
2, 5 are motion vector detectors (ME), 3 are resolution converters, 6, 7 are subtractors, 8 and 9 are DCT calculators, 10, 1
1 is a quantizer (Q), 12 and 13 are variable length encoders (V
LC), 14, 15 are inverse quantizers (IQ), 16, 17
Is an inverse DCT calculator (IDCT), 18 and 19 are adders,
20 and 21 are image memories, 22 and 23 are rate controllers,
Reference numeral 24 is a resolution inverse converter, and 25 is a data reconstructor as data arranging means. Further, FIG. 1 shows the first encoding means and the second encoding means as an example. In particular, the subtractor 6 here outputs the difference component in both encoding processes by the first encoding means and the second encoding means. In addition,
The same or corresponding parts in the drawings are designated by the same reference numerals and the description thereof will be omitted.

Next, the operation will be described. Video data is input to the resolution converter 3 in the order of interlace raster scan. The input video data is filtered and thinned by the resolution converter 3 in order to prevent aliasing noise in the high frequency region. (FIG. 2 is an explanatory diagram showing the concept of the resolution conversion described above on an image.) For example, taking data of horizontal 704 pixels and vertical 480 pixels as an example, after filtering, the horizontal and vertical ½ times each. Is thinned out to an image of horizontal 352 pixels and vertical 240 pixels and converted into low-resolution screen data.

The low-resolution screen data is input to the preprocessor 4 to be converted from raster scan to block scan. (Note that the block scan here means sending data in the order of DCT blocks.) Since an I-picture is intra-frame encoded, it is encoded without performing inter-frame arithmetic using the output of the frame memory. Turn into.

In the case of an I picture, since the image memory 21 which is the input of the subtractor 7 does not output anything, the video signal passes through the subtractor 7. This data is the DCT calculator 9
Are orthogonally transformed into frequency components, and the orthogonally transformed data are input to the quantizer 11 and quantized in a zigzag scanning order from the low frequency region. Further, the quantized image data is entropy coded through the variable length encoder 13 and output to the data reconstructor 25.

On the other hand, the image data quantized by the quantizer 11 is inversely quantized by the inverse quantizer 15 and inverse DC
The T device 17 inversely converts the frequency component data into spatial component data. The I picture is decoded without performing the interframe calculation using the output of the frame memory that is intraframe coded. Therefore, in the case of I picture, adder 1
Since there is no input from the image memory 21 of 9, the data is passed through. The output of the adder 19 is used as data stored in the image memory 21. It is necessary to store at least I picture, or I picture and P picture data in the image memory. Because usually MPEG
1, in MPEG2 I for decoding B picture
This is because the picture and P picture are necessary as reference data.

The image memory 20 receives the output of the decoded data adder 18 and the result of the pixel number returned to the original value by interpolating the pixel by the resolution inverse converter 24, and averages them with a certain weighting. Store image decoding data. For simplicity of weighting, the case where the output of the resolution inverse converter 24 is 1 and the output of the adder 18 is 0 will be described for simplicity.

Further, the input video data is buffered by the preprocessor 1 to perform scan conversion from raster scan to block scan, and from the image memory 20 which stores the signal subjected to the above low resolution processing. The data is subtracted by the subtracter 6 (this is called a resolution residual component). The resolution residual component is orthogonally transformed into the frequency domain by the DCT calculator 8, converted into a scan from the low frequency region, and appropriately quantized by the quantizer 10. This data is entropy coded through the variable length encoder 12 and output to the data reconstructor 25.

On the other hand, the data quantized by the quantizer 10 is inversely quantized by the inverse quantizer 14 and inversely transformed into the data in the spatial domain by the inverse DCT unit 16. In the adder 18, the input from the image memory 20 is low-resolution processed, but the inversely transformed data and the inverse DCT calculator 16 are added to each other to obtain low-resolution data in terms of resolution and data other than low-resolution data. , And the decoding result of the data in two layers by the data of the residual component is obtained and stored in the image memory 20. This layer is determined by the number of resolution conversions,
It is also possible to perform resolution conversion twice to form three layers, and data of an arbitrary layer can be created by a similar approach.

Regarding encoding such as normal MPEG, I
In order to perform bidirectional prediction from the picture and the P picture to encode the B picture, the I picture and the P picture are decoded and stored as decoded data.
In this way, the I picture and the P picture are encoded, and then the B picture is processed.

I picture, P picture, B as described above
The picture encoding process is performed on both the low resolution component and the resolution residual component. In this way, a sequence in which low resolution components R (hereinafter referred to as R components) and resolution residual components S are alternately arranged can be configured. The data reconstructor 25 transfers these data to a location convenient for optical head access, such as the head of the GOP.
It operates so that P is fixed and arranged, and for example, it is arranged like the sequence a in FIG. Rearrange like this,
The low resolution component can be reproduced by reading the first half, which is the region occupied by the L component. The resolution residual component has a smaller amount of data than the non-resolution residual component, and can be efficiently layered.
That is, here, the first encoding means is used as an example of the first encoding means for encoding according to a predetermined condition and the image information other than the image information encoded by the first encoding means. A second coding means for coding the residual component by coding is provided to perform efficient hierarchization.

FIG. 3 is a diagram showing an example of the data structure result. In FIG. 3, the sequence a is the sequence generated by the encoding process of the present embodiment, and the sequence b is the second.
A sequence generated by the encoding process of the embodiment of
The sequence c is a sequence generated by the encoding process of the third embodiment. In the sequence b, L represents a low frequency component, H represents a high frequency component, C in the sequence c represents a component quantized and encoded by coarse quantization, and A represents a coarse quantization residual component. Further, as shown in the sequence a of FIG. 8, the above operation may be performed only on the I picture and the P picture, and only the component may be fixedly arranged at the head of the GOP.

As described above, if only the low-resolution component is fixedly arranged at the head of the GOP, the ratio of the L component to the whole is greatly reduced, and a margin is given to the reading speed from the medium, and the skip search is performed. Etc. can be easily realized. Further, if only the R components of the I picture and P picture are arranged at the beginning of the GOP such as the sequence a in FIG. 8, only the low resolution data of the I picture and P picture are decoded. Works like. In the embodiments described above, the thinning ratio is horizontal 1/2 times,
Although the case of vertical 1/2 times has been described, it goes without saying that this ratio may be other than this and may be applied to any ratio.

Regarding the encoding method, MPEG1,
Although there are MPEG2, JPEG, etc., it is not always necessary to use a common encoding method in the hierarchy of resolutions. Because
When encoding with reduced resolution, the MPEG1 system that does not mention interlacing is sufficient for encoding. In addition, even in the JPEG method for still images, since the stack for each frame is a moving image, it is possible to occupy a certain specific position of the GOP and still perform normal decoding. Further, although the description has been made with the hierarchy of two resolutions, it goes without saying that more layers may be used. It should be noted that the data of the low resolution component is encoded by the first encoding means in FIG. 1, the output from this first encoding means is interpolated, and the difference component from the image before pixel thinning is subtracted. It is also possible to provide a difference component encoding means for performing the encoding for the difference component obtained in step 6 to encode the difference component.

The frame to be read from the image memory is usually brought from the prediction reference frame as shown in the figure, but since there is a low resolution frame, the time axis is adjusted properly (the memory address is determined). It needs to be stored in memory and read. It goes without saying that additional information such as an audio signal and a header and an error correction code may be added to the residual component data by providing information adding means.

Example 2. Another embodiment of the present invention will be described with reference to FIG. In this embodiment, the DCT
The low-frequency region and the high-frequency region of the block are divided into layers, and only the low-frequency region is arranged at the head of the GOP. FIG. 4 is a block diagram of a digital video signal coding processing unit. In FIG. 4, reference numerals 26 and 27 respectively denote a first variable length coder (VLC1) and a second variable length coder (VLC1).
LC2). It should be noted that the same or corresponding parts in the drawings are designated by the same reference numerals and the description thereof will be omitted.

Next, the operation will be described. The video data of this interlace has an effective screen size of horizontal 7 for example.
This is data of 04 pixels and 480 vertical pixels. Since the I picture is decoded without performing the interframe calculation using the output of the frame memory that is intraframe coded, the video data is output through. This video data is orthogonally transformed into frequency components by the DCT calculator 8, converted into a block scan from the low frequency region, and then quantized by the quantizer 10.
The appropriate quantization is done by.

The data arrangement of DCT coefficients inside the DCT block is shown in FIG. The lower frequency component in the upper left corner of Fig. 5,
The DCT coefficient data of the high frequency component is arranged at the lower right. Among the DCT coefficient data arranged in this DCT block, the DCT coefficient data in the low frequency region up to the DCT coefficient data at a specific position (for example, the shaded area in FIG. 5) serves as the low frequency region extracting means. The data is entropy-encoded through the first variable length encoder 26 and output to the data reconstructor 25. In addition, the second variable length encoder 27
Performs variable length coding on the DCT coefficient data after the DCT coefficient data at the specific position. That is, the data is partitioned and coded in this manner in the frequency domain.

Regarding the coding of the motion vector or the DC component, only the first variable length encoder 26 is required, and the second variable length encoder 27 is not necessary. This is because during normal reproduction, the output data of the first variable length encoder 26 and the output data of the second variable length encoder 27 may be combined and decoded.

Although the coding area is determined at the fixed position of the DCT coefficient, other methods may be used, for example, a fixed number of events may be used. That is, the unit that gives the Huffman code that is a variable length code is an event,
The coding area may be set with a preset number of events, such as three units. If an example of the output bit stream of the data reconstructor 25 is the sequence b of FIG. 3, an image in the low frequency region can be reproduced by reading only the low frequency region in the first half. Although FIG. 8 is a diagram showing an example of the processing sequence, the arrangement may be changed to the arrangement like the sequence b in FIG.

On the other hand, the data quantized by the quantizer 10 is inversely quantized by the inverse quantizer 14 and inversely transformed into the data in the spatial domain by the inverse DCT unit 16. The I picture is decoded without performing the interframe calculation using the output of the frame memory that is intraframe coded. Therefore,
In the case of an I picture, there is no input from the image memory 20 of the adder 19, so the data is passed through. The output of the adder 19 is used as data stored in the image memory 20.

At least I pictures and P are stored in the image memory.
Pictures need to be stored. Because usually MPE
This is because the I picture and the P picture are necessary as reference data for decoding the B picture in both G1 and G2.

According to this structure, the ratio of the L component to the whole is greatly reduced, and the read speed from the medium of the head has a margin so that the skip search can be realized. Further, as will be described later, if only I pictures and P pictures are fixedly arranged, it becomes possible to operate so as to easily decode only the low frequency component data of the I pictures and P pictures. Since the data amount in the high frequency region is smaller than the data amount in all the regions, the data can be constructed more efficiently than the data in the low frequency region is extracted and stored before the data in the entire region.

When the I picture finishes coding by passing through the subtracter 6 and coding, the B picture is coded by bidirectional prediction with the last P picture of the GOP which is immediately preceding in time. . Motion vector detection and prediction mode by comparing the output of the preprocessor 1 and the data (the arrow is omitted in the drawing) from the memory of the reference frame,
Judging the frame structure etc., based on the judgment result,
The data of the reference frame memory in which the output of the preprocessor 1 and the data of the reference frame memory are most suitable are read from the image memory 20 as the data for the forward direction and the data for the backward direction, and the read data And the output result of the B picture preprocessor 1 are subtracted by the subtracter 6 (the result is referred to as a temporal residual component for both P picture and B picture). DCT calculation is performed on the time residual component, the result is quantized, and further variable length coding is performed.

Embodiment 3 FIG. Another embodiment of the present invention will be described with reference to FIG. In the present embodiment, the coarse quantized component of the DCT coefficient and the coarse quantized residual component are divided into hierarchical layers as an example of data other than the coarse quantized component, and the coarse quantized component is GOP.
I'm trying to place it at the beginning of. FIG. 6 is a block diagram showing a digital video signal encoding processing unit. In FIG. 6, 28 is a subtractor and 29 is an adder. It should be noted that the same or corresponding parts in the drawings are designated by the same reference numerals and the description thereof will be omitted.

Next, the operation will be described. The input image data of this interlace is, for example, data having an effective screen size of horizontal 704 pixels and vertical 480 pixels. The I picture is decoded without performing the interframe calculation using the output of the frame memory that is intraframe coded. Therefore, in the case of an I picture, since the image memory 20 which is the input of the subtractor 6 does not output anything, the video signal is the subtracter 6
Through. This data is orthogonally transformed into frequency components by the DCT calculator 8, converted into a block scan from the low frequency region, and the quantizer 10 reduces the amount of encoded data to half or less by an appropriate rough quantum. Be converted. The quantized data is entropy-coded through the variable length encoder 22 and output to the data reconstructor 25.

On the other hand, the data quantized by the quantizer 10 is inversely quantized by the inverse quantizer 14 (the result is referred to as a rough quantization result), and this inversely quantized data has another code. Inverse D is sent to the processing unit (dotted line frame part in FIG. 6).
The CT calculator 16 inversely transforms the data into spatial domain data. In another encoding processing unit (here, since I picture encoding is described, there is no output from the image memory 21, but normally the decoding result of this encoding processing unit is stored in the image memory 21. Therefore, the motion vector detector 2 determines the motion vector detection or prediction mode based on the data, and the DCT block mode is determined. The data at the position corresponding to the determined mode is referred to and the subtraction of the subtracter 7 is performed. You will be on the input side.)

The output of the subtractor 7 is subjected to DCT, and the result of coarse inverse quantization and the residual are obtained by the subtractor 28 (this is called a coarse quantized residual). The coarse quantization residual is finely quantized by the quantizer 11 (fine quantization which is almost the same as ordinary encoding in consideration of code amount control) and variable length encoding, while it is inversely quantized and inverse DCT is performed. Image memory 2 after being decoded
Stored in 1. As for the coding result and the coding result of the rough quantization, the data reconfigurator 25 determines the allocation of necessary data and adds a header or the like.

If the sequence c in FIG. 3 is used as an example of this output data, the decoding result of the image quantized roughly can be obtained only by reading the first half of the GOP. Further, since the data of the coarse quantization residual is smaller than all the finely quantized data, it is possible to configure the data more efficiently than extracting the coarsely quantized data and storing it before the finely quantized data.

As another example, the sequence c in FIG.
It is also possible to perform a shift-type process such as the arrangement of. With this configuration, the ratio of the C component (the component encoded by performing the coarse quantization) to the whole is greatly reduced, and the read speed from the medium has a margin, so that the skip search can be realized. Also, as will be described later, if only I pictures and P pictures are arranged in a solid manner, only I-pictures and P-pictures that have undergone coarse quantization operate to be decoded.

When the I picture finishes coding by passing through the subtractor 6 and coding, the B picture is coded by bidirectional prediction with the last P picture of the GOP immediately preceding in time. . Motion vector detection and prediction mode by comparing the output of the preprocessor 1 and the data (the arrow is omitted in the drawing) from the memory of the reference frame,
Judging the frame structure etc., based on the judgment result,
The data of the reference frame memory in which the output of the preprocessor 1 and the data of the reference frame memory are most suitable are read from the image memory 20 as the data for the forward direction and the data for the backward direction, and the read data And the output result of the B picture preprocessor 1 are subtracted by the subtracter 6 (the result is referred to as a temporal residual component for both P picture and B picture). The data for the forward direction and the data for the backward direction are read from the image memory 21, and the data and the output of the preprocessor 1 are subtracted by the subtractor 7 to perform orthogonal transform entropy coding. Note that P pictures are also encoded using the same process.

FIG. 7 is a diagram showing an example of the statistic amount of coded data. The variation of the code amount when the number of frames in the GOP: N = 15, the period of I pictures and P pictures: M = 3. Is shown. It can be seen from this figure that I pictures and P pictures account for about 50% of the whole, so if only this part or only I pictures are divided into layers by resolution, frequency, and quantization as in the above embodiment. Further, since the code amount to be reproduced is further reduced, the moving time of the optical head can be shortened, and the functions such as skip search can be easily realized.

The processing sequence at that time is shown in FIG.
FIG. 8 encodes the sequence of the I picture, P picture, and B picture of the original image, and performs the processing of the above-described first, second, and third on only the I picture and P picture of the original B picture. Is encoded without dividing the hierarchy.
A sequence obtained by processing I pictures and P pictures according to the processing of the first embodiment is referred to as a sequence b, and a sequence obtained by processing I pictures and P pictures according to the processing of the third embodiment is referred to as a sequence c.

In each sequence, the low-resolution component (R), low-frequency component (L), coarse-quantized component (C) I-picture, and P-picture component are respectively reconstructed by the data reconstructor 2.
5 configures the data so that it is fixedly placed at the head of the GOP. For sequence a, I picture,
Only the low resolution component of the P picture (in the sequence a in FIG.
Since low-resolution images of I pictures and P pictures can be decoded in the core area portion after data reconstruction), skip search can be easily supported. Of course, it is not necessary that the data other than the core area are arranged as shown in the figure, and it goes without saying that the data may be arranged in the order of the frame numbers at the time of encoding in FIG.

As for the sequence b, only the low frequency components of the I picture and P picture (see sequence b in FIG. 8).
Middle, core picture after data reconstruction), I picture,
Since a low-frequency image of P picture can be obtained, it is possible to easily cope with skip search. Further, for the sequence c, only the coarse quantization components of the I picture and P picture (see FIG.
Since coarsely quantized images of I pictures and P pictures can be decoded in the core area portion after data reconstruction in sequence c), it is possible to easily cope with skip search.

For example, in the configuration shown in FIG. 1, for the B picture, if the coding loop including the preprocessor 4 is not used and only the coding loop including the preprocessor 1 is used for coding, Good, in the configuration shown in FIG.
All frequency components may be encoded in the first variable length encoder (VLC1) 26. Further, in the configuration shown in FIG. 6, the quantizer 10 may perform fine quantization for encoding.

Ideally, basic data such as low-frequency data is collected at the beginning of the GOP, but it may be slightly shifted so that it overlaps the beginning of the unit forming the error correction code. Needless to say. In addition,
Thus, the basic data arrangement according to the error correction unit can be similarly performed in other embodiments.

Example 4. Another embodiment according to the present invention will be described with reference to FIGS. FIG. 9 is a diagram showing an example of an arrangement of DCT blocks and an arrangement of frequency components in a bit stream of one block. In FIG. 9, a shows a macro block header and a DCT of a luminance signal in the arrangement of all DCT blocks. Block Y1-Y
4. It is expressed that one macroblock is formed by the DCT block U1 of the color difference signal (BY) and the DCT block V1 of the color difference signal (RY). Also, regarding the arrangement of DCT blocks of low-frequency components, a is a macro block header and a DCT block Y1L of the luminance signal.
It is expressed that Y4L, the DCT block U1L of the color difference signal (BY), and the DCT block V1L of the color difference signal (RY) form one low-frequency component macroblock. It should be noted that the same or corresponding parts in the drawings are designated by the same reference numerals and the description thereof will be omitted.

Regarding the arrangement of the high frequency component DCT blocks, C is a luminance signal DCT block Y1H to Y4H,
It is expressed that the DCT block U1H of the color difference signal (BY) and the DCT block V1H of the color difference signal (RY) form one macroblock of the high frequency component. D represents the concept of arrangement of frequency component data of one block bit stream. FIG. 10 is a block diagram of a digital video signal decoding processing unit and an explanatory view showing the operation concept thereof. In FIG. 10, 30 is a mode switching unit as a mode switching unit, 31 is a data rearranging unit as a data rearranging unit. , 32 are decodable discriminators, 33 is a variable length decoder (VLD), and 34 is a switch. The decodable discriminator 32 and the switch 34 constitute a data operating means. 35 is an inverse quantizer,
36 is an inverse DCT calculator, 37 is an image memory, 38 is an adder, and 39 is an inverse scan converter.

Next, the operation will be described. The data shown in FIG. 9 is, for example, a code string assembled into 8 bits (1 byte) in the vertical direction, and each macroblock has information about that macroblock, which is called a macroblock header. This information includes, for example, an increment address, a quantization scale code, a motion vector, a marker bit, a macroblock pattern (coded block).
k Pattern) and the like.

After this macroblock header, each DC
The T blocks of encoded data follow. The method of filling this data is such that the bytes are first arranged from the bit stream and each byte is arranged in order.
Since the code length of the T block is variable, the block boundary or the boundary between the header and the data is not completed in byte units, and a boundary often exists in the middle of byte units. By the way, although the data of each block has a variable length, the closer to the macro block header, the lower the frequency of the data.

This data is a unit for giving an event (1 variable length code, and normally, in the case of a DC component, 1 is a DC component.
Since an AC component constitutes an event, and the AC component is run-length encoded, a combination of a non-zero DCT coefficient and a run length constitutes one event. At the end of the block, a code called EOB (end of block, end of block) completes one event. ) Is used as a maximum value to divide the low-frequency component (L) and the high-frequency component (H) into the encoded data as shown in (a) and (c) of FIG.

Next, the operation of FIG. 10 will be described. First, the mode switcher 3 sends a mode signal, such as during a skip search from a microcomputer or during normal continuous playback.
Input to 0. On the other hand, a reproduction signal from a disk is amplified by an amplifier and output from a PLL or the like after being reproduced by a phase-locked clock. Then, the output data is discriminated, digitally demodulated, and error-corrected. After that, the audio signal is separated from the system layer including the video signal data and the audio signal data, and the video bit stream is extracted and input to the data rearranger 31.

The output (control signal) of the mode switch 30 is supplied to the data rearranger 31 and the decodable discriminator 32. The data rearranger 31 obtains a control signal and reconnects the data before division from the L component and the H component of FIG. 9, or outputs the data to the variable length decoder (VLD) 33 only with the L component. The variable length decoder (VLD) 33, together with the decodable discriminator 32, extracts the break of the event in the L component region, decodes up to the break and outputs the decoded break to the switch 34. This switch 34 is connected so that 0 is not inserted during normal reproduction. The switch 34 controlled by the output of the decodable decision unit 32 inputs the decoded low frequency component to the DCT block, while DC
The entire DCT block is configured such that 0 is inserted on the high frequency side of the T block.

At the time of decoding, the DC configured as described above
The data of the T block is inversely DCT'ed so that the output of the adder 38 is passed through in the case of an I picture, and is corrected by the motion vector of the reference I picture or P picture in the case of a P picture. In the case of the B picture, the reading of the image memory 37 is performed according to the case of each picture so that both the I picture and the P picture which became the reference are corrected by the motion vector and added. Controls the dashi and adds 38
Is added by

The DCT mode and the prediction mode motion vector at this time are controlled based on the information obtained by decoding the code of the header. According to the process as described above, the motion-compensated predicted data is decoded and stored in the image memory 37, and the image is restored in the original configuration order of the GOP. The inverse scan converter 39 performs buffering and conversion from block scan to raster scan in the order of image output.

In this embodiment, the macro block is skipped or the fixed length data is shortened to a fixed length.
It is needless to say that there is no problem even if the L component data is connected to the next block, because the L component can be reliably extracted by sequentially detecting the L component. Further, if the data length exceeds a predetermined value, it goes without saying that EOB may be added as L component data even up to the event delimiter. Although not particularly shown in the above description of the embodiment, information adding means for adding additional information such as an audio signal and a header and an error correction code may be further provided to add the data to the high frequency region. Not to mention good things.

Example 5. Next, another embodiment of the present invention will be described with reference to FIGS. FIG. 11 is a block diagram of the digital video signal encoding processing unit and an explanatory view showing the operation concept thereof. In FIG. 11, 40 is a rate controller and 41 is an EOB searcher. Here, a first variable length encoder 26 and a second variable length encoder 27 are provided as encoding means. It should be noted that the same or corresponding parts in the drawings are designated by the same reference numerals and the description thereof will be omitted.

Next, the operation will be described. The interlaced input image data is buffered by the preprocessor 1 to convert a raster scan into a block scan.
The I picture is decoded without performing the interframe calculation using the output of the frame memory that is intraframe coded. Therefore, in the case of an I picture, since the image memory 20 which is the input of the subtractor 6 does not output anything, the video signal passes through the subtractor 6.

This data is orthogonally transformed into frequency components by the DCT calculator 8, converted into a block scan from the low frequency region, and appropriately quantized by the quantizer 10. D of a specific position in this quantized data
The data in the low frequency region up to the CT coefficient data is entropy coded through the first variable length encoder (VLC1) 26 and output to the data reconstructor 25.

The second variable length encoder (VLC2)
27 variable-length-encodes the DCT coefficient after the data at the specific position. Regarding the coding of the motion vector and the DC component, at least, the variable length encoder 26 may be used.
At this time, in order for the break of the L component to change at the rate without restriction on the code, it is necessary to add EOB to both the L component and the H component even though it is one block. In this way, by disposing the EOB code at the delimiter between the L component and the H component, the break of the L component can be changed according to the rate.

On the other hand, the data quantized by the quantizer 10 is inversely quantized by the inverse quantizer 14 and inversely transformed into spatial component data by the inverse DCT unit 16.

The I picture is decoded without performing the interframe calculation using the output of the frame memory which is intraframe coded. Therefore, in the case of I picture, adder 1
Since there is no input from the image memory 20 of No. 8, data is passed through. The output of the adder 18 is used as data stored in the image memory 20. It is necessary to store at least I picture, or I picture and P picture data in the image memory. Because usually MPEG
1, in MPEG2 I for decoding B picture
This is because the picture and P picture are necessary as reference data.

When the coding of the I picture is completed,
Next, a B picture is coded by bidirectional prediction with the last P picture of the previous GOP. The output of the preprocessor 1 is compared with the data (the arrow is omitted in the drawing) from the memory of the reference frame to determine the motion vector detection and the prediction mode, the frame structure, etc. The data of the memory of the reference frame in which the output of the device 1 and the data of the memory of the reference frame are most suitable are read out from the image memory 20 together with the data for the forward direction and the data for the backward direction, and the preprocessing of the data and B picture The output result of the device 1 is subtracted by the subtracter 6 (the result is a P picture,
B pictures are also called temporal residual components). This time residual component is DCT, quantized and variable length coded.

When divided into the low frequency region and the high frequency region, the rate is not constant due to the frequency components included in the pattern. Therefore, since the data rate in the low frequency region is not constant, the head actuator controllable range cannot be completely compensated. Here, the rate controller 40
Controls the area of the low frequency component to be variable according to the rate, and controls the size of the low frequency area to be variable with respect to the target rate as shown in FIG.

That is, the first variable length encoder (VLC1)
While monitoring the output of 26, if the amount is larger than the target rate set by the application, the area occupied by the data in the low frequency region is reduced, and the first variable length encoder (V
If the code amount of LC1) 26 is small, the area of the low frequency region is increased. While monitoring the code amount in this way, the setting of the occupied area in the low frequency region is appropriately changed for the first variable length encoder (VLC1) 26 and the second variable length encoder (VLC2) 27.

In addition to the above, for example, by performing preliminary encoding in advance, a criterion for setting the occupied area of the low frequency region is determined and the target rate is set based on the result of which region has more or less codes. Is also good.

FIG. 12 is a block diagram of the digital video signal decoding processing section, and shows a block diagram of the decoding processing for decoding the coded data coded as described above. Next, description will be made with reference to FIG. A mode signal such as a skip search or normal continuous reproduction is input to the mode switch 30 from a microcomputer or the like. On the other hand, the reproduced signal from the disk is amplified by the amplifier and discriminates the reproduced signal by the clock applied with the PLL to perform digital demodulation, error correction processing is performed to separate the audio signal from the system layer, and the video bit stream is extracted. And inputs it to the data rearranger 31.
The output of the mode switching unit 30 as the mode switching unit is supplied to the data rearrangement unit 31 and the decodable determination unit 32. The data rearranger 31 operates to reconnect the data before division from the L component and the H component of FIG. 9 by obtaining this control signal, or only the L component without connecting to the H component. A variable length decoder (VL) as a decoding means.
D) Output to 33.

In the present invention, the L component is not theoretically cut off in the middle of an event, but variable length decoding is performed just in case, considering decoding of a signal with poor signal quality such as skip search. The rectifier (VLD) 33 and the decodable discriminator 32 confirm the break of the event in the L component region, decode up to that break, and output it to the switch 34. The switch 34 operates so as to be always turned on in the case of reproduction data having good signal quality as in normal reproduction. Here, the decodable discriminator 32 and the switch 3
4 constitutes a data operating means.

The switch 34 is controlled by the output of the decodable discriminator 32 so that 0 is inserted on the high frequency side of the block from the low frequency component for which decoding is successful, and the DC is controlled.
In the case of an I picture, the output of the adder 38 is passed through in the case of forming a T block and inverse DCT (IQ) the data, and in the case of a P picture, the motion vector in the reference I picture is corrected. The reading of the image memory 37 is controlled and added by the adder 38 so that the addition is performed, and in the case of the B picture, the I picture and the P picture are corrected by the motion vector and added. Further, the DCT mode or the prediction mode motion vector at this time is controlled by decoding the code of the header. The data thus motion-compensated and predicted is decoded and stored in the image memory 37, and the images are arranged in the original configuration order of the GOP. The reverse scan converter 39 performs buffering and conversion in the image output order and from block scan to raster scan.

In the above description, an example of controlling the size of the DCT coefficient region has been described.
The number of events may be controlled. In this case, the EOB is reached without reaching the predetermined number of events for the L component.
However, since the EOB searcher 41 monitors the appearance of EOB, the L component can be reliably captured. Here, in particular, the data rearranger 31 and the EOB searcher 41 reconstruct the data based on each of the low frequency region data, the high frequency region data, and the EOB. That is, the data rearranger 31 and the EOB searcher 41 constitute data reconstructing means.

Of course, since the energy after DCT is small, it is needless to say that it is desirable to treat the non-coded non-coded block in the same manner for the L component and the H component. For the H component, it is ideal that the data excluding the L component is run-length encoded, but the H component may be encoded by setting the L component to 0. Normal MPE
Since this can be handled with the same configuration as the G variable length encoder (VLC), this can be simplified in terms of circuit.

Example 6. Another embodiment of the present invention will be described with reference to FIG. FIG. 13 is a block diagram showing a digital video signal decoding processing unit, and in FIG.
2 is a multiplexer, 43 is a switch, 44 is a first variable length decoder (VLD1), 45 is a second variable length decoder (VLD2), 46 is a first inverse quantizer (IQ1), 4
Reference numeral 7 is a second inverse quantizer (IQ2), 48 and 49 are adders, 50 and 51 are image memories, and 52 is a resolution inverse converter. Note that FIG. 13 also shows a low-resolution decoding unit as a decoding unit. It should be noted that the same or corresponding parts in the drawings are designated by the same reference numerals and the description thereof will be omitted.

Next, the operation will be described. It can be considered that FIG. 13 corresponds to a processing block of video data of a reproduction signal from the disc when the encoded data described with reference to FIG. 6 is recorded on the optical disc or the like. A mode signal such as a skip search or normal continuous reproduction is input from a microcomputer or the like to a mode switching unit 30 as a mode switching unit. On the other hand, the reproduced signal from the disk is amplified by the amplifier and discriminates the reproduced signal by the clock applied with the PLL to perform digital demodulation, and error correction processing is performed to separate the audio signal from the system layer and extract the video bit stream. To do.

The extracted video bit stream is input to the multiplexer 42. Multiplexer 4
2 sends low resolution component data to the second variable length decoder 45, and sends other data to the first variable length decoder 44 via the switch 43.

The switch 43 is controlled by the mode switch 30, and the resolution residual component is reproduced halfway though the mode requires only the reproduction image output of the low resolution component by skip search or the like. When it comes in, it operates so as to stop the transmission of extra data. Also, during normal reproduction, the switch 33 operates so as to remain connected.

The second variable length decoder 45 decodes the Huffman code and the run length code, and the second dequantizer 47.
Is inversely quantized, and the inverse DCT unit 36 converts the frequency component into a spatial component.

From the converted data, in the case of an I picture, the adder 49 is passed through and stored in the image memory. In the case of a P picture, for the first P picture, the I picture stored in the image memory is referred to, and after the second P picture, the (previous) P picture stored in the image memory is referred to, and the motion vector The position is corrected and read, and the adder 49 performs motion compensation prediction. In the case of B picture, the same operation is performed based on I picture and P picture.

The motion vector in the figure, the quantization parameter for the inverse quantization, the prediction mode, etc. are output from the variable length decoder and are the same as those shown in FIG. The loop shown by the dotted line block in FIG. 13 is a component block for decoding the low resolution component, and the decoding result is interpolated by the resolution inverse converter 52 as the interpolation image generation means to be decoded as the resolution residual component. It is input to the image memory 50 to supplement the result.

During normal reproduction, the resolution residual component is decoded
The image is output by the inverse scan converter 39 in combination with the decoding result of the low resolution component (in accordance with the division processing when this is performed in time division). Through the switch 43, the first variable length decoder 44 can decode the frequency component, and the inverse quantizer 46 dequantizes the inverse D
It is decoded by the CT unit 36 into resolution residual component data in the spatial domain.

The image memory 50 refers to the pixel interpolation data of the low resolution component, and further the P picture is the I picture and the B picture.
The picture is position-corrected by the motion vector by referring to the I picture and the P picture, is read from the image memory 50, and the motion compensation prediction is decoded by the adder 48.

In the case of skip search or the like, in order to prevent the resolution residual component from being reproduced halfway by the switch 43, the output of the resolution residual component is stopped so that the output of the inverse DCT unit 36 is output. Prevents extra data from being output. Therefore, only the data obtained by pixel interpolation of the low resolution component passes through the image memory 50 (image memory 5
The same operation is performed even if a switch is provided in the input unit of 0), and the image is output through the reverse scanning conversion 39.

Example 7. Next, another embodiment of the present invention will be described with reference to FIGS. Figure 14 is a GOP
FIG. 4 is a block diagram showing an address generation and disk control unit, and particularly shows a processing block when the rate information is recorded in a sequence header or the like. In FIG. 14, 53 is a register, 54 is a GOP address calculator,
An optical head / disk rotation control converter 55 serves as a head position conversion means and a disk rotation control conversion means. FIG. 15 is a block diagram showing a GOP address generation and disc control unit including a reproduction process. Particularly, the GOP is reproduced from a disc in which the above rate information is hardened and recorded at several positions on the disc. The configuration for is shown. In FIG. 15, reference numeral 56 is a reproduction amplifier, 57 is a digital demodulator, and 58 is an error corrector (EC
C) and 59 are system layer processors, and 60 is a rate information memory, and the system layer processor 59 and the rate information memory 60 constitute data rate information extraction means. Reference numeral 61 is a GOP number counter, and the GOP address calculator 54 and the GOP number counter 61 constitute position information calculating means. It should be noted that the same or corresponding parts in the drawings are designated by the same reference numerals and the description thereof will be omitted.

Next, the operation will be described. As described in the conventional example, by making the rate per GOP variable, the overall rate of one program can be optimized, so that the image quality is considerably improved. However, for this reason 1 GOP
If you do not know whether the beginning of the GOP is the beginning of the GOP unless you see the contents of the data, and you want to appreciate it from the place where you have seen the software you saw halfway, while searching the data on the disk in detail, find the position I have no choice but to look for it.

Therefore, in such a case, in the present invention, first, rate control of a variable rate is performed by, for example, 1 Mbit, 1.5 Mbit, 2 Mbit, 2.5 Mbit per GOP.
Discrete rate targets such as 3M bits are set, and rate information of all GOPs is recorded on the disc. In particular, as a predetermined area on a recording medium such as a disc, a TOC (Table Of Conte
Abbreviation of nts: A recording area is assigned to the first part of the disc, and information such as title and recording time is recorded. ), Quasi-TOC, etc., is the most efficient if the rate information for each GOP is recorded.

Also, rate information about GOP may be incorporated in the sequence header of the video bit stream. For example, 2 hours of software is 14.4k GO
In this case, the rate information needs to be able to be distinguished for the five types of rates, so 3
It can be expressed in bits. Therefore, 5.4k bytes (14.4
Information of (k pieces × 3 bits / 8 bits / byte) enables recording at all GOP rates on the disc.

The rate information of each GOP is shown in FIG.
It is possible to access the desired GOP at high speed by storing it in the rate information memory 60 shown in FIG. 5 and integrating the information length corresponding to the value.

A description will be given below with reference to FIG. The variable length decoder 33 decodes the Huffman code and the run length code and decodes the header to determine the motion vector, the type of the picture, and the like.

On the other hand, the sequence header is also decoded,
The rate information is input to the GOP head address calculator 54. In addition, the address information of the currently accessed GOP is stored in the register 53, the GOP start address of the next access destination is calculated, and stored in the register 53.
The position of the optical head is obtained from the address using the optical head and disk rotation control converter 55 up to the head of the next GOP to be accessed. Next, the control signal for the next access is calculated from the difference between the currently accessed GOP and the start address of the access destination GOP. The position control of the optical head actuator and the disk rotation control are performed based on this control signal.

The reproduction process will be described with reference to FIG. Directly or indirectly from the TOC area or its equivalent (after specifying the rate information address,
The address part is accessed and the rotation of the optical head and the optical disk is controlled so as to read the rate information), the reproduction signal from the optical head is amplified by the reproduction amplifier 56, and this signal is detected by the digital demodulator 57. Then, it discriminates into a digital signal and performs digital demodulation.

The reproduced signal, which has been digitally demodulated and becomes digital data, is input to the error corrector 58, and the error contained in the reproduced signal is corrected. The data after this error correction is separated and processed by the system layer processor 59 into an audio bit stream, a video bit stream and other data.

For example, this signal indicates which data type (AV
The stream path is divided according to whether it applies to (video and audio) data, text data, binary data such as programs, etc.). Among them, the rate information as described above is stored in the rate information memory 60.

On the other hand, the GOP number counter 61 is used to generate information as to which GOP is to be processed, and the optical head actuator and the disk are rotated based on the address calculated by the GOP head address calculator 54. Control the speed.

In the above description, the GOP counter receives the signal from the system layer processor 59. However, a part such as a microcomputer that takes the user I / F substitutes the process, or skips during reproduction. It may be more efficient to input address data from a variable length decoder or the like that processes a video bit stream when moving to a search or the like.

Example 8. Next, the operation of another embodiment of the present invention will be described below with reference to FIGS. 16, 17, and 18. FIG. 16 is a diagram showing a signal processing block in the case where the frequency division and the quantization division of the digital signal reproduction unit are performed, and particularly the reproduction processing in the case where the rate information is hardened and recorded in several places on the disc. It is a block diagram about the structure used for. It should be noted that the same or corresponding parts in the drawings are designated by the same reference numerals and the description thereof will be omitted.

At the beginning of disk reproduction, the rotation of the optical head and the optical disk is controlled so that the TOC area or an area equivalent thereto is directly or indirectly read (the address for describing the rate information is specified), and the reproduction signal from the optical head is reproduced. The signal is amplified by 56, this signal is detected by the digital demodulator 57, the digital signal is discriminated, and the digital demodulation is performed. The reproduced signal converted into digital data by this is input to the error corrector 58, and the error contained in the reproduced signal is corrected. The error-free data is separated into an audio bit stream and a video bit stream by the system layer processor 59, and other data is also processed.

For example, the stream path is divided by determining whether this signal is AV data, text data, or binary data such as a program. Among them, the rate information as described above is stored in the rate information memory 60. On the other hand, the GOP counter generates information about how many GOPs to process,
It is calculated by the OP head address calculator 54 and controls the rotation speed of the optical head actuator and the disk.

In this way, the address of the GOP to be accessed is determined even during skip reproduction, and the disk is skipped. When the desired GOP can be accessed, the next address is calculated in the same manner and, for example, the configuration described in the fifth embodiment is used. The data in the low frequency region obtained by using the data is reproduced and displayed on the screen.

A mode signal such as a skip search or normal continuous reproduction is input to the mode switch 30 from a microcomputer or the like. As described above, the video bitstream is extracted and input to the data rearranger 31. The output of the mode switching unit 30 is supplied to the data rearrangement unit 31 and the decodable determination unit 32. The data rearranger 31 operates to reconnect the data before division from the L component and the H component of FIG. 9 by obtaining this control signal, or only the L component without connecting to the H component. Output to the variable length decoder (VLD) 33.

In this embodiment, the L component is not theoretically cut off in the middle of the event, but VLD is used as a precaution in consideration of decoding a signal with poor signal quality such as skip search. With the decodable decision unit 32, the break of the event in the L component area is confirmed, and even the break is decoded and output to the switch 34.

The switch 34 is controlled by the output of the decodable discriminator 32 so that 0 is inserted on the high frequency side of the low frequency component of the successfully decoded low frequency component, and the DC is controlled.
In the case of an I picture, the output of the adder 38 is passed through in the case of forming a T block by inverse DCT, and in the case of a P picture, the motion vector of the reference I picture is corrected and added. In the case of the B picture, the reading of the image memory 37 is controlled so that the I picture and the P picture are corrected by the motion vector and added, and the addition is performed by the adder 38.

The DCT mode or prediction mode motion vector at this time is controlled by decoding the code of the header. The data thus motion-compensated and predicted is decoded and stored in the image memory 37, and the images are arranged in the original configuration order of the GOP. The reverse scan converter 39 performs buffering and conversion in the image output order and from block scan to raster scan. Also, the switch 34 is
It is not connected so as to insert 0, and control is performed so that only the reproduction data is reproduced.

Further, in the case of dividing and encoding in the low frequency region and the high frequency region in this way, the quantization table that emphasizes the low frequency side and the quantization table that emphasizes the high frequency side, or one quantum evenly. It is easily conceivable that a fine quantization table is prepared for the quantization table regardless of the frequency domain. In such a case, a variable length decoder and an inverse quantizer like the local decoder in FIG. 6 are used. It can be realized by having 2 each. The data rearranger 31 must then be a multiplexer.

Next, the operation will be described with reference to FIG. FIG. 17 is a diagram showing a signal processing block in the case where division is performed by the bit length of the digital signal reproducing section, and particularly the reproduction processing in the case where the above rate information is recorded at several places on the disc. It is a block diagram for explaining an example. The rotation of the optical head and the optical disk is controlled so as to read directly or indirectly (specify the address describing the rate information) from a predetermined area on the recording medium, for example, the TOC area or an area equivalent thereto at the beginning of disk reproduction, and the optical head Playback signal from the playback amplifier 5
The signal is amplified by 6, and this signal is detected by the digital demodulator 57, discriminated into a digital signal, and digital demodulated.

The reproduced signal converted into digital data by this is input to the error corrector 58, and the error contained in the reproduced signal is corrected. The error-free data is separated into an audio bit stream and a video bit stream by the system layer processor 59, and other data is also processed.

For example, the stream path is divided by judging whether this signal is video / audio data, text data, or binary data such as a program. Among them, the rate information as described above is stored in the rate information memory 60. On the other hand, what G
Information about whether or not to process the OP is generated by the GOP number counter and calculated by the GOP head address calculator 54 to control the rotation speed of the optical head actuator and the disk.

In this way, the address of the GOP to be accessed is obtained even during skip reproduction, and the GOP is skipped on the disk. When the desired GOP can be accessed, the next address is calculated in the same manner. The data in the low frequency region obtained by using the data is reproduced and displayed on the screen.

A mode signal, such as during a skip search or during normal continuous reproduction, is input to the mode switch 30 from a microcomputer or the like. Also, the video bit stream is extracted and input to the data rearranger 31. The output of the mode switching unit 30 is supplied to the data rearrangement unit 31 and the decodable determination unit 32. Data sorter 31
Operates to obtain the control signal and reconnect the undivided data from the L component and the H component of FIG.
The L component alone is output to the variable length decoder (VLD) 33 without being connected to the component.

The VLD 33, together with the decodable discriminator 32, extracts the break of the event in the L component area, decodes up to that break, and outputs it to the switch 34. Switch 3
4 is a decodable discriminator 3 so that 0 is inserted on the high frequency side of the low frequency component that has been successfully decoded.
2 is controlled by the output of 2 to form a DCT block, and its data is subjected to inverse DCT to adder 3 in the case of an I picture.
The output of 8 is passed through, and in the case of the P picture, the motion vector in the reference I picture is corrected and added, and in the case of the B picture, the motion vector is corrected and added from the I picture and P picture. Further, the reading of the image memory 37 is controlled and added by the adder 38.

The DCT mode or prediction mode motion vector at this time is controlled by decoding the code of the header. The data thus motion-compensated and predicted is decoded and stored in the image memory 37, and the images are arranged in the original configuration order of the GOP. The reverse scan converter 39 performs buffering and conversion in the image output order and from block scan to raster scan. Further, the switch 34 is not connected so as to insert 0 at the time of normal reproduction, but operates so as to reproduce only reproduction data.

Next, the operation of FIG. 18 will be described. FIG.
FIG. 8 is a diagram showing a signal processing block in the case where division is performed according to the resolution of the digital signal reproducing unit, and an embodiment of a reproducing process particularly when the above rate information is recorded at several places on the disc is recorded. It is a block diagram for explaining. The rotation of the optical head and the optical disk is controlled so that the TOC area or an area equivalent thereto is read directly or indirectly (the address for describing the rate information is specified) at the beginning of the disk reproduction, and the reproduction signal from the optical head is amplified by the reproduction amplifier 56. Then, the digital demodulator 57 detects this signal, discriminates it into a digital signal, and performs digital demodulation.

The reproduced signal converted into digital data by this is input to the error corrector 58, and the error contained in the reproduced signal is corrected. The error-free data is separated into an audio bit stream and a video bit stream by the system layer processor 59, and other data is also processed. For example, the stream path is divided by judging whether this signal is AV data, text data, or binary data such as a program. Among them, the rate information as described above is stored in the rate information memory 60.

On the other hand, information about how many GOPs to process is generated by the GOP number counter and calculated by the GOP head address calculator 54 to control the rotational speed of the optical head actuator and the disk. In this way, the address of the GOP to be accessed is obtained even during skip reproduction, and the disc is skipped. When the desired GOP can be accessed, the next address is calculated in the same manner and obtained using, for example, the configuration described in the fourth embodiment. The data in the low frequency region is reproduced and displayed on the screen.

A mode signal such as a skip search or a normal continuous reproduction is input to the mode switch 30 from a microcomputer or the like. The video bit stream is extracted and input to the multiplexer 42. The multiplexer 42 outputs the data of the low resolution component to the second variable length decoder 4
5 to the first variable length decoder 44 via the switch 43. The switch 43 is controlled by the mode switch 30, and although the mode requires only the reproduction image output of the low resolution component by skip search or the like,
When the resolution residual component is reproduced halfway, it is turned off. In addition, the switch 43 is turned on when performing reproduction with good signal transmission quality such as normal reproduction.
To be done.

The second variable length decoder 45 decodes the Huffman code and the run length code, and the second dequantizer 47
Is inversely quantized, and the inverse DCT unit 36 transforms the frequency domain into the spatial domain. If the data is an I picture, the adder 49 is passed through and stored in the image memory. If the data is a P picture, the P picture is referenced from the image memory, the position is corrected by the motion vector, and the data is read out. Decode motion compensated prediction. If it is a B picture, the same operation is performed for I picture and P picture.

The motion vector in the figure, the quantization parameter for the inverse quantization, the prediction mode and the like are output from the variable length decoder, but the flow of information is the same as in FIG. 12 and is omitted. The loop on the lower side of the drawing is decoding of the low resolution component, and the decoding result is input to the image memory 50 in order to interpolate the pixel by the resolution inverse converter 52 and complement the decoding result as the resolution residual component. It

Example 9. Next, another embodiment of the present invention will be described with reference to FIGS. 19, 20 and 21. 19 is a block diagram for the encoding process, and FIG. 20 is a block diagram for the decoding process. 19 and 20, reference numeral 62 is a video signal encoder as an encoding means, 63 is an audio signal encoder, 64 and 67.
Is a memory, and 65 and 68 are memory controllers. The memory 64 and the memory controller 65 constitute a data supplement means. Further, the video signal encoder 62 and the memory controller 65 constitute a code amount comparing means. 66 is a system layer bitstream generator, 6
Reference numeral 9 denotes a variable length decoder, 70 denotes a decoded signal processing unit after the variable length decoder, and the variable length decoder 69 and the decoded signal processing unit 70.
Functions as data decoding means, and the data rearranger 31 in FIG. 20 functions as data reconstruction means. It should be noted that the same or corresponding parts in the drawings are designated by the same reference numerals and the description thereof will be omitted.

First, the operation of the configuration shown in FIG. 19 will be described. A memory 64 is arranged between the video signal encoder 62 and the system layer bitstream generator 66 to embed data between GOPs of the encoded video signal, and then generate a system layer bitstream. Each GOP is input to the device 66, while the audio signal is encoded by the audio signal encoder 63 and then input to the system layer bit stream generator 66 together with the video signal, and an operation such as adding a header is performed. It

The data embedding operation in the memory 64 will be described below. The memory controller 65 functions as a control circuit of the memory 64, and performs control so that the coded video signal data is refilled between the GOPs without blanks. The signal processing will be described below with reference to FIG. FIG. 21 is an explanatory view showing the concept of processing of the digital signal recording / reproducing apparatus, for example, nG.
When the OP generates excess data for the access position of the optical head or the control unit of the error control, the (n + 1) GOP ends halfway with the access position of the optical head or the control unit of the error control. When a blank is generated in the data area, as shown in FIG. 21A, the blank nGOP data after the (n + 1) GOP is filled with the overfilled nGOP data, and similarly, the nGOP is filled, so that the data cannot be inserted a little. Remaining data of (n + 1) GOP and (n +
2) The data of GOP is embedded in the blank part of (n + 3) GOP (the embedding direction is from left to right on the paper surface).

As another control method, instead of sending the surplus data backward as described above, the (n + 2) GOP controls the access position of the optical head and error control, as shown in (a) of FIG. A little over the unit (n +
3) If the GOP ends halfway with respect to the access position of the optical head or the control unit of error control, (n +
2) The blank area after the GOP was exceeded (n + 3)
The data of GOP was filled in, and similarly (n + 3) GOP was filled in, so it was a little too small to enter (n + 2) G.
The remaining data of the OP is controlled such that the data of which the data of the (n + 1) GOP cannot be filled is embedded in the blank portion of the nGOP (the embedding direction is from right to left on the paper).

Next, the operation of the configuration shown in FIG. 20 will be described. The memory 67 is controlled by the memory controller 68 so as to restore the data rearranged according to the rules as described above with reference to FIGS. For example, in the case of restoring the data as shown in FIG. 21A, the nGOP portion following the (n + 1) GOP is connected after the nGOP data on the left side of the paper, and then the (n + 1) GOP data is connected, ( n + 2) GOP
The data of (n + 1) GOP that follows the data of (1) is connected to the data of (n + 1) GOP to restore the state of the original GOP data.

The rearrangement rule is, for example, TO
It is necessary to be defined in advance as a formatting rule of the medium, such as recording as flag information in a grouped area following the C area, and if it is not defined, it must be specified somewhere on the medium. I have to.

Example 10. Next, another embodiment of the present invention will be described with reference to FIGS. 22, 23 and 24.
22 is a diagram showing a signal processing block when frequency division or quantization division of the digital signal reproducing unit is performed, and FIG. 23 is a bit length division or quantization dividing of the digital signal reproducing unit. It is a figure showing the signal processing block in the case. FIG. 24 is a diagram showing a signal processing block when division by resolution or division by quantization of the digital signal reproducing unit is performed.
In FIG. 24, 71 is an IP selection indicator, 72 is a decodable judging device, and 73 is a switch. Note that, in FIG. 24, corresponding portions are expressed as an example of the first decoding means, the second decoding means, and the third decoding means. It should be noted that the same or corresponding parts in the drawings are designated by the same reference numerals and the description thereof will be omitted.

Next, the operation will be described. In FIG. 22 and FIG. 23, first, the rotation of the optical head and the optical disk is controlled so that the TOC area or an area equivalent thereto is directly or indirectly read (the address in which the rate information is described) is read at the beginning of the disk reproduction. The reproduction signal of is amplified by the reproduction amplifier 56, and the digital demodulator 57
This signal is detected by and discriminated into a digital signal,
Perform digital demodulation. The reproduced signal converted into digital data by this is input to the error corrector 58, and the error contained in the reproduced signal is corrected. The error-free data is separated into an audio bit stream and a video bit stream by the system layer processor 59, and other data is also processed. 22, FIG. 23 and FIG. 24 output a control signal to the IP selection indicator 71 from the mode switching unit 30 as the mode switching unit for switching the decoding unit based on the special reproduction speed. This control signal and the skip search speed are controlled so as to switch between displaying only the I picture and displaying the I picture and the P picture.

When the skip search speed is 100 times or the like, if both the I picture and the P picture are output on the screen, GOPs must be thinned out and output, which is unnatural to the motion of the reproduced screen. Come out. In this case, in order to eliminate the unnaturalness, it is necessary to switch to the reproduction mode for only the I picture, and the decodable discriminator 32 (72 in FIG. 24) is instructed to stop decoding the P picture as well as the B picture ( In FIG. 24, the switch 73 plays this role, and controls the image memory 37 (50 and 51 in FIG. 24) so that only the I picture is displayed.

Normally, screen display of I-pictures and P-pictures is good up to about 5 times speed, but screen display of only I-pictures is better at about 15 times or more. This is because if all I-pictures and P-pictures are displayed at 15 times speed, the next GOP that can be played is 5 GO even if the screen is updated every frame.
The continuity of movement is extremely deteriorated because of the P destination. Also, G
When the number of frames in OP: N = 15, I picture, P picture cycle: M = 3, for example, in the case of 7 × speed reproduction, all P pictures are decoded, but the screen output is I.
By outputting a picture and a second P picture (third and ninth pictures in GOP), finer skip search can be performed.

As described above, when recording is performed by dividing the frequency domain at a predetermined position of each GOP data read from the recording medium, when recording is performed by dividing by the resolution, quantization is performed. The data state is divided and recorded based on a predetermined condition such as when recording is performed by dividing by the level. When the data is reproduced as the first data and the data other than the basic first data is reproduced as the second data, if the decoding is performed for all the first and second data, the first data is decoded. And (at least) obtaining a reproduced image corresponding to the low frequency region of the I picture or the number of thinned pixels, the first data is decoded to at least the low frequency region of the I picture or P picture or the thinned image. Of the case of obtaining a reproduced image corresponding to the prime number, has decoding means for obtaining a reproduced image either, whether used which decoding means may be switched on the basis of the special playback speed during special playback.

Needless to say, the setting of the display method of the I picture and P picture may be changed between the reproduction in the forward direction and the reproduction in the reverse direction. Since P pictures can be decoded only in the positive direction of time, in the reverse reproduction, it is necessary to store the screen existing before the P picture to be decoded, and the memory for that is added. Need to use. Therefore, in order to facilitate this without using extra memory, for example, I pictures and P pictures are reproduced in the forward skip search, and only I pictures are reproduced in the backward skip search. And play it back.

Example 11. Another embodiment of the present invention will be described with reference to FIGS. 25, 26, 27 and 28. FIG. 25 is a diagram showing a signal processing block when the digital signal reproducing unit performs division by frequency or quantization, and FIG. 26 shows a signal processing block when dividing by the bit length of the digital signal reproducing unit. FIG. Further, FIG. 27 is an explanatory diagram showing the concept of the processing at the time of skip search. In FIGS. 25, 26, and 27, 74 is a field display controller. The system layer processor 59 functions as video data extracting means. Further, in FIG. 25 and FIG. 26, corresponding portions are displayed as an example of the video data decoding / reproducing means. Reference numeral 30 is a mode switch as a mode switching means. It should be noted that the same or corresponding parts in the drawings are designated by the same reference numerals and the description thereof will be omitted.

Next, the operation will be described. In FIGS. 25 and 26, first, the rotation of the optical head and the optical disk is controlled so that the TOC area or an area corresponding thereto is directly or indirectly read (the address in which the rate information is described) is read at the beginning of the disk reproduction. The reproduction signal of is amplified by the reproduction amplifier 56, and the digital demodulator 57
This signal is detected by and discriminated into a digital signal,
Perform digital demodulation. The reproduced signal converted into digital data by this is input to the error corrector 58, and the error contained in the reproduced signal is corrected. The error-free data is separated into an audio bit stream and a video bit stream by the system layer processor 59, and other data is also processed.

At the time of skip search, for example, I picture and P
When the pictures are continuously output on the screen, the screen output order is as shown by the arrow in FIG. At this time, the even field of the I picture and the odd field of the P picture are continuous during the skip search, but there is an interval of 4 fields in the encoded data. That is, during this period, the reproduction speed is 5 times higher than that between the odd field and the even field of the I picture. Therefore, the playback speed is 1 for each field.
It is an image of an unnatural movement that moves back and forth between double and quintuple.

The even field of the I picture is replaced with the same screen as the odd field, or the average of the upper and lower scanning lines is filled in considering interlacing to create a screen and output the same. The image memory 37 shown in FIGS. 25 and 26 is controlled by the field display controller 74 so as to have the structure.
By doing so, the skip amount between fields, which is the interval between the fields encoded when each field of the image to be reproduced is recorded, can be obtained almost uniformly, so that jerkiness (unnatural movement) becomes inconspicuous.

FIG. 28 is an explanatory view showing the concept of the processing at the time of reverse reproduction, and particularly showing the field order of reverse reproduction. The reverse playback field order will be described below with reference to FIG. When the reverse reproduction is performed, the reverse reproduction is usually performed for each frame in which an odd field and an even field are paired. Specifically, when moving from the odd field to the even field, reproduction is performed in the same direction as the time on the image (reproduction operation following the passage of C in FIG. 28A), and when moving from the even field to the odd field, the image is reproduced. The reverse feed of two fields is performed in the opposite direction to the above time (a skip operation that follows the progress of d in FIG. 28A).

By the way, if the reproduction path as described above is taken, the reverse feed is performed at a triple speed, resulting in a reproduced image having a visually inferior motion in which the sequence of motion is not directly felt. When the image display memory 37 is controlled by the field display controller 74 in FIGS. 25 and 26 so as to display it in the opposite direction in the order of odd-even, odd-even ... by one screen for each field as shown in FIG.
Since the skip amount between fields can be obtained almost evenly,
Jerkiness becomes less noticeable. However, it is necessary to maintain the normal odd-even field relationship (rather than reversing the field sync signal) for the sync signal at this time.

Since the display method is independent for each field in this way, the image memory does not take the output of the adder 38 as it is, but takes it out independently from the image memory. In order to perform such an operation, a buffer may be separately provided and the order may be changed, or a three-port memory in which address control can be independently set may be used. Furthermore, the above operation can be realized even if the data is read out after being multiplexed with a memory having a very high operation speed.
In addition, the inverse scan converter 39 has at most one field plus one.
Needless to say, the inverse scan converter 39 may be provided with such a buffering function because it has a slice memory.

Further, in slow reproduction of special reproduction, jerkiness becomes conspicuous because the same frame is output several times in succession. Therefore, the frames are reconfigured and output so that the reproduction times are equidistant. For example, in the case of slow playback at 1/3 times speed, for example, after the decoded I frame is output three times, the decoded B frame is not output three times, but the first one frame is the I frame. One frame is constructed from the odd field of (the even field side may be obtained by averaging the upper and lower lines).

By doing so, the interlaced image does not shift in the vertical direction of the screen due to the line shift, and the image becomes stable. ), And the next one frame outputs the original I frame, and the next one frame constitutes one frame from the even field of the I frame (the odd field side may take the average of the upper and lower lines).
Then, the next one frame constitutes one frame from the odd field of the B frame (the even field side may take the average of the upper and lower lines, etc.), and the next one frame outputs the original B frame. If one frame is composed of an even field of the B frame (the odd field side may be obtained by averaging the upper and lower lines, etc.), it is possible to throw at equal intervals in time.

[0197]

According to the digital video signal recording apparatus of the first aspect of the present invention, it is possible to perform coding so that the minimum code amount to be accessed during special reproduction is reduced, and the minimum data amount to be accessed during special reproduction can be obtained. This has the effect of enabling efficient placement on the disc.

According to the digital video signal recording apparatus of the second aspect, the spatially thinned video data is coded by the first coding means and only the data of the first coding means is accessed. For example, even if it is decoded, there is an effect that it can be encoded so that the scene can be fully understood.

According to the digital video signal recording apparatus of the third aspect, the video data in the partial frequency domain is coded for each block by the first coding means, and only the data of the first coding means is coded. If accessed, there is an effect that encoding can be performed so that a scene can be fully understood even if it is decoded.

According to the digital video signal recording apparatus of the fourth aspect of the present invention, the first encoding means coarsely quantize the data so that the encoding can be performed without lowering the resolution. Accessing only this has the effect of enabling encoding so that the scene can be fully understood even if it is decoded.

According to the digital video signal recording apparatus of the fifth aspect, the low frequency region of the video signal can be divided by the data extracting means, and the code amount can be easily limited to a fixed length or less. In addition, there is an effect that if the data in the low frequency region is decoded, the rough content of the image can be encoded so that it can be understood.

According to the digital video signal reproducing apparatus of the sixth aspect, the normal reproducing data rearranging means joins the coded data divided into two to obtain a complete decoded image, and the special reproducing image is reproduced. Has the effect of obtaining an image to the extent that the rough content of the image can be grasped by decoding only the data in the low frequency region of the coded data divided into two, and depending on the operating status of the device. The effect is that it can be decrypted.

According to the digital video signal reproducing apparatus of the seventh aspect, during the special reproduction, the data operating means allows
When decoding the low frequency region of the coded data divided into two, only the decodable data is decoded, and the bits that could not be decoded are discarded to avoid decoding the abnormal data. Further, there is an effect that a decoded image can be obtained without data conversion by replacing the high frequency region of the block with a fixed value and decoding.

According to the digital video signal recording apparatus of the eighth aspect, both the low frequency region and the high frequency region of the block are coded by EOB as if the block is apparently ended, so that only the data in the low frequency region is decoded. In this case, it is possible to obtain coded data that can be decoded without requiring an extra circuit such as discarding the data.

According to the digital video signal reproducing apparatus of the ninth aspect, the data reproducing means at the time of normal reproduction allows
A complete decoded image can be obtained from the coded data separated by EOB, respectively.
By decoding only the data in the low frequency region, it is possible to obtain a rough image enough to understand a scene. Further, the above two modes have an effect that they can be operated according to the operating conditions of the device. Further, when the data manipulating means decodes the low frequency region of the encoded data, the remaining high frequency region of the block is replaced with a fixed value for decoding, so that decoding can be performed without data conversion.

According to the digital video signal recording apparatus of the tenth aspect of the present invention, the low resolution encoding means encodes the spatially thinned video data, and only the data of the encoding means is accessed for decoding. Even so, there is an effect that it can be encoded so that the scene can be fully understood. The differential component coding means interpolates the decoded data in the low resolution coding means and compares it with the image before the low resolution conversion to obtain the differential component, which is not obtained by the low resolution coding means. By encoding high resolution video data, video information other than low resolution can be encoded.

According to the digital video signal reproducing apparatus of the eleventh aspect, during normal reproduction, the video signal decoding means causes the low resolution encoded data, the data before thinning to low resolution, and the low resolution data. By synthesizing the coded data of the high resolution component of the difference component of, there is an effect that an image of perfect resolution can be decoded.

According to the digital video signal reproducing apparatus of the twelfth aspect, the low resolution encoded data divided into two, the data before thinning to the low resolution and the low resolution portion are divided by the mode switching means during the normal reproduction. The image with full resolution can be decoded by synthesizing the coded data of the high resolution component of the difference component with the data, and the rough image is decoded by decoding only the low resolution coded data during special playback. The effect is that it can be converted. Further, there is an effect that the decoding means can be switched as described above depending on the operation state of the device.

According to the digital video signal reproducing apparatus of the thirteenth aspect, when the video generating means decodes only the low resolution coded data during the special reproduction, the video data of the low resolution component is interpolated to the original. The effect is that it can be restored to the image size of.

According to the digital video signal recording apparatus of the fourteenth aspect, since the data rate setting means sets only a very limited value in the variable rate video signal encoding means, the GOP data rate information ( G
There is an effect that it becomes possible to express (corresponding to the code amount of OP) with a small number of bits.

According to the digital video signal recording device of the fifteenth aspect, since the data rate setting information when encoded at a variable rate is recorded on the recording medium separately from the video data, the data rate information is collected together. Since it becomes possible to read the information, it is possible to record information such that the position of a predetermined GOP on the disk can be immediately calculated.

According to the digital video signal recording apparatus of the sixteenth aspect, encoding of high image quality can be realized by the variable rate, and the GOP whose rate is greatly increased by the variable rate is spatially thinned. By the encoding means for encoding the video data, it is possible to perform the encoding so as to reduce the minimum access area at the time of reproduction.

According to the digital video signal recording apparatus of the seventeenth aspect, it is possible to realize high-quality coding by the variable rate, and the coding means can block the block even in the GOP in which the rate is greatly increased by the variable rate. By the encoding means for encoding the video data of the partial frequency domain every time, there is an effect that the encoding can be performed so as to reduce the minimum access area at the time of reproduction.

According to the digital video signal recording apparatus of the eighteenth aspect, it is possible to realize high-quality coding by the variable rate, and the coding means is coarse even in the GOP in which the rate is greatly increased by the variable rate. Quantization has the effect of encoding complete resolution data and encoding so as to reduce the minimum access area during reproduction.

According to the digital video signal reproducing device of the nineteenth aspect, the data rate information extraction means
It becomes possible to extract the data rate information of each GOP, and the disc position information calculating means calculates the position information on the disc having the GOP to be accessed from the obtained data rate information, so that the image quality can be changed. This has the effect of facilitating special playback and search of rates.

According to the digital video signal reproducing apparatus of the twentieth aspect, the position information on the disk having the GOP to be the access target is calculated according to the special reproduction speed, and the optical head position converting means determines the special reproduction speed. It is possible to control the position of the optical head to the position of the target GOP according to the effect, and it is possible to perform special reproduction at a variety of speeds with high image quality and variable rate.

According to the digital video signal recording apparatus of the twenty-first aspect, even when the data is coded and recorded at a variable rate, the access time is shortened by GOP at a position where the head can easily access. It becomes possible to record by encoding so as to increase the amount of data. Further, at that time, it is possible to fill a waste part such as a blank part on the disc as much as possible, and it is possible to use it for improving the image quality or contribute to prolonging the recording time.

According to the digital video signal reproducing apparatus of the twenty-second aspect, the coded data in which the data of another GOP is embedded in the blank area can be reconstructed from the coded data of the original GOP to be decoded without being converted into data. In addition, the read data amount can be increased in the special reproduction, and a high-quality reproduced image can be obtained.

According to the digital video signal reproducing apparatus of the twenty-third aspect, the mode for decoding and displaying only the I picture and the mode for displaying the I and P pictures are switched according to the speed of special reproduction. In slow special reproduction (eg, double speed), special reproduction of I and P pictures can be realized, fine special reproduction with less frame skipping can be realized as compared with special reproduction of only I pictures, and I can be used in fast special reproduction. It is possible to obtain a digital video signal reproducing device having an excellent man-machine interface that responds to a reproduction speed such as special reproduction of a picture.

According to the digital signal reproducing apparatus of the twenty-fourth aspect, during reverse reproduction, the field display is displayed so as to move in reverse, and at the time of frame skip reproduction such as fast forward, the even field is the same as the odd field. Is output, the number of fields to be skipped is made constant, and a special reproduction image that is easy to see can be obtained.

According to the digital video signal reproducing device of the twenty-fifth aspect, by the data rate information extracting means,
It becomes possible to extract the data rate information of each GOP, and the disc position information calculating means calculates the position information on the disc having the GOP to be accessed from the obtained data rate information, so that the image quality can be changed. This makes it possible to easily perform rate trick play and search, and at the time of trick play, according to the trick play speed, a mode for decoding and displaying only I pictures and I,
Since the mode for displaying the P picture is switched, the special reproduction of the I and P pictures can be realized in the comparatively slow special reproduction (for example, double speed), and the special reproduction with less frame skipping than the special reproduction of only the I picture can be realized. It is possible to obtain a digital video signal reproducing device having an excellent man-machine interface that can realize reproduction and that corresponds to the reproduction speed such as performing special reproduction of an I picture during fast special reproduction.

[Brief description of drawings]

FIG. 1 is a block diagram of a digital video signal encoding processing unit according to a first embodiment.

FIG. 2 is a diagram showing the concept of resolution conversion according to the first embodiment on an image.

FIG. 3 is an explanatory diagram showing an example of a data configuration result of Examples 1, 2, and 3.

FIG. 4 is a block diagram of a digital video signal encoding processing unit according to a second embodiment.

FIG. 5 is an explanatory diagram showing an example of a data arrangement of DCT coefficients inside a DCT block.

FIG. 6 is a block diagram of a digital video signal encoding processing unit according to a third embodiment.

FIG. 7 is an explanatory diagram showing an example of a statistic amount of encoded data according to the first, second, and third embodiments.

FIG. 8 is an explanatory diagram showing an example of a processing sequence of Examples 1, 2, and 3.

FIG. 9 is an explanatory diagram showing an example of the arrangement of DCT blocks and a schematic arrangement of frequency components in a bitstream of one block according to the fourth embodiment.

FIG. 10 is a block diagram of a digital video signal decoding processing unit according to the fourth embodiment and an explanatory diagram showing the operation concept thereof.

FIG. 11 is a block diagram of a digital video signal encoding processing unit according to a fifth embodiment and an explanatory diagram showing the operation concept thereof.

FIG. 12 is a block diagram of a digital video signal decoding processing unit according to a fifth embodiment.

FIG. 13 is a block diagram of a digital video signal decoding processing unit according to the sixth embodiment.

FIG. 14 is a block diagram of a GOP address generation and disk control unit according to the seventh embodiment.

FIG. 15 is a block diagram of a GOP address generation and disk control unit including the reproduction processing according to the seventh embodiment.

FIG. 16 is a signal processing block diagram in the case where frequency division and quantization division are performed in the digital signal reproducing unit according to the eighth embodiment.

FIG. 17 is a signal processing block diagram in the case where division is performed by the bit length of the digital signal reproducing unit according to the eighth embodiment.

FIG. 18 is a signal processing block diagram when division is performed according to the resolution of the digital signal reproducing unit according to the eighth embodiment.

FIG. 19 is a block diagram of encoding processing according to the ninth embodiment.

FIG. 20 is a block diagram of decoding processing according to the ninth embodiment.

FIG. 21 is an explanatory diagram showing the concept of processing of the digital signal recording / reproducing apparatus of the ninth embodiment.

FIG. 22 is a signal processing block diagram in the case where frequency division or quantization division is performed in the digital signal reproducing unit according to the tenth embodiment.

FIG. 23 is a signal processing block diagram in the case where division is performed by a bit length or division by quantization of the digital signal reproducing unit according to the tenth embodiment.

FIG. 24 is a signal processing block diagram in the case of performing division by resolution or division by quantization in the digital signal reproducing unit of the tenth embodiment.

FIG. 25 is a signal processing block diagram in the case where frequency division or quantization division is performed in the digital signal reproducing unit according to the eleventh embodiment.

FIG. 26 is a signal processing block diagram in the case where division is performed by the bit length or division by quantization of the digital signal reproducing unit of the eleventh embodiment.

FIG. 27 is an explanatory diagram showing the concept of processing during skip search according to the eleventh embodiment.

FIG. 28 is an explanatory diagram showing the concept of processing during reverse reproduction according to the eleventh embodiment.

FIG. 29 is a block diagram showing an example of a conventional digital signal recording / reproducing apparatus.

FIG. 30 is a block diagram of a video signal encoding unit in a conventional digital signal recording / reproducing apparatus.

FIG. 31 is an explanatory diagram showing the concept of moving image processing of video signal encoding in a conventional digital signal recording / reproducing apparatus.

FIG. 32 is a block diagram of a video signal decoding unit in a conventional digital signal recording / reproducing apparatus.

FIG. 33 is a block diagram of a video signal encoding unit in a conventional digital signal recording / reproducing apparatus.

[Explanation of symbols]

3 resolution converter, 4 preprocessor, 5 motion vector detector, 7 subtractor, 9 DCT calculator, 11 quantizer,
13 variable-length encoder, 15 inverse quantizer, 17 inverse D
CT calculator, 18 adder, 19 adder, 20 image memory, 21 image memory, 23 rate controller, 24
Resolution inverse converter, 25 data reconstructor, 26 first variable length encoder, 27 second variable length encoder, 28
Subtractor, 29 adder, 30 mode switcher, 31 data rearranger, 32 decodable decider, 33 variable length decoder, 34 switch, 35 inverse quantizer, 36
Inverse DCT calculator, 37 image memory, 38 adder,
39 inverse scan converter, 40 rate controller, 41 EO
B search device, 42 multiplexer, 43 switch, 4
4 first variable length decoder, 45 second variable length decoder,
46 first dequantizer, 47 second dequantizer, 4
8 adder, 49 adder, 50 image memory, 51
Image memory, 52 resolution inverse converter, 53 register,
54 GOP address calculation, 55 optical head / disk rotation control converter, 56 reproduction amplifier, 57 digital demodulator, 58 error corrector, 59 system layer processor, 60 rate information memory, 61 GOP number counter, 62 video signal encoding Device, 63 audio signal encoder, 64 memory, 65 memory controller, 6
6 system layer bit stream generator, 67 memory, 68 memory controller, 69 variable length decoder, 70 decoded signal processing unit, 71 IP selection indicator,
72 Decodable discriminator, 73 switch, 74 field display controller.

Continuation of front page (51) Int.Cl. ⁶ Identification number Reference number within the agency FI Technical display location H04N 7/32 H04N 7/137 Z (72) Inventor Tadashi Kasezawa 1st place, Babajo, Nagaokakyo, Kyoto Mitsubishi Electric Video System Development Laboratory, Inc. (72) Inventor Masato Nagasawa, No. 1 Baba Institute, Nagaokakyo-shi, Kyoto Mitsubishi Electric Corporation Video System Development Laboratory

Claims (25)

[Claims] 1. A digital video signal recording apparatus for recording a digital video signal encoded by using motion compensation prediction and orthogonal transformation on a recording medium, wherein a digital video encoded by using the motion compensation prediction and orthogonal transformation. First encoding means for encoding a video signal of an encoded image including at least an intra-frame encoded image of the signal according to a predetermined condition, and the first encoding means of the video signal are used. Second encoding means for performing encoding on the residual component by encoding, and the respective output data output from the first and second encoding means, for each image group data, each image A digital video signal recording device, comprising: a data arranging means for arranging at a predetermined position in group data. 2. A configuration in which the first encoding means encodes video information thinned out at a predetermined interval with respect to a video signal composed of an encoded image including at least an intra-frame encoded image. The digital video signal recording device according to claim 1. 3. The digital video signal recording apparatus according to claim 1, wherein the first coding means is configured to code only the low frequency region subjected to the orthogonal transformation. 4. The digital video signal recording apparatus according to claim 1, wherein the first encoding means is configured to roughly quantize and encode at a quantization level. 5. A digital video signal recording apparatus for recording a digital video signal encoded by using motion compensation prediction and orthogonal transformation on a recording medium, wherein the digital video encoded by using the motion compensation prediction and orthogonal transformation. Low-frequency region extracting means for extracting data for a low-frequency region from a data string obtained by dividing a video signal, which is a coded image including at least an intra-frame coded image, of the signal into predetermined bits. Digital video signal recorder. 6. A digital video signal reproduced from a recording medium, the digital video signal being coded using motion compensation prediction and orthogonal transformation, and composed of data in which low-frequency region data and high-frequency region data are divided into regions. In a video signal reproducing device, data rearranging means for rearranging the data in the low frequency region and the data in the high frequency region in a predetermined order, and a mode for decoding the rearranged data or selectively the low frequency region And a mode switching means for selecting one of the modes for decoding the data. 7. When decoding in a mode of decoding only data in a low frequency region, only decodable data is decoded, and undecodable data near a predetermined bit number boundary is discarded and obtained. 7. The digital video signal reproducing apparatus according to claim 6, further comprising data operating means for substituting fixed values for the data in the high frequency region and performing inverse orthogonal transformation. 8. A digital video signal recording apparatus for recording a digital video signal encoded using motion compensation prediction and orthogonal transform on a recording medium, wherein the digital image encoded using the motion compensation prediction and orthogonal transform. When the predetermined number of bits as the data in the low frequency region is reached for the coded data of each block of the video signal formed of the coded image including at least the intra-frame coded image of the signal, the block end code is added, and A digital video signal recording apparatus comprising: a coding means for coding the above-mentioned coded data exceeding a predetermined number of bits as high frequency region data. 9. A digital image in which low-frequency region data encoded on the basis of motion-compensated prediction and orthogonal transform and high-frequency region data encoded by block end codes are read from a recording medium. In a signal reproducing device, data reconstructing means for reconstructing data based on low-frequency region data, high-frequency region data, and block termination code, and a mode for selectively reconstructing the reconstructed data or a selective low-frequency region A mode switching means for selecting one of the modes for decoding only the data of the area, a decoding means for decoding the coded data reconstructed based on the output of the mode switching means, and a high frequency area With a data manipulating means for substituting fixed values for inverse orthogonal transformation. Place. 10. A digital video signal recording apparatus for recording a digital video signal encoded by using motion compensation prediction and orthogonal transformation on a recording medium, wherein a digital video encoded by using the motion compensation prediction and orthogonal transformation. Coding means for coding low-resolution component data obtained by performing pixel thinning on a video signal composed of a coded image including at least an intra-frame coded image, and interpolation of output data of the coding means Then, it has a differential component coding means for coding a differential component with respect to an image before pixel thinning, and divides the output of the coding means and the output of the differential component coding means into a predetermined area to store data. A digital video signal recording apparatus, comprising: an information adding unit configured to add an audio signal, a header, and the like and add an error correction code. 11. A difference component between low-resolution data encoded based on motion-compensated prediction and orthogonal transformation and thinned out from a pixel, and an image before pixel thinning and an image interpolated after pixel thinning. In a digital video signal reproducing device for reading a digital video signal having data corresponding to the above from a recording medium, there is provided a decoding means for synthesizing and decoding the low resolution component data and the difference component data. Digital video signal playback device 12. A mode switching means for switching between a mode for combining low-resolution component data and differential-component data for decoding and a mode for decoding only low-resolution components. Digital video signal playback device. 13. The digital video signal reproducing apparatus according to claim 11, further comprising an interpolated video generation means for generating a video interpolated after decoding when the low resolution image is decoded. 14. When recording a digital video signal encoded by using motion compensation prediction and orthogonal transformation on a recording medium, image deterioration at the time of encoding and decoding based on the motion compensation prediction and orthogonal transformation. A pattern determining means for determining the degree of the above, an adaptive encoder for adaptively changing and encoding the data rate based on the determination output from the pattern determining means, an audio signal, additional information such as a header, and an error correction code. A digital video signal recording apparatus having information adding means for adding, wherein the digital video signal recording apparatus has a data rate setting means for setting a discrete value to a data rate to be adaptively changed. 15. When recording a digital video signal encoded by using motion compensation prediction and orthogonal transformation on a recording medium, image deterioration when encoding and decoding based on the motion compensation prediction and orthogonal transformation. A pattern determining means for determining the degree of the above, an adaptive encoder for adaptively changing and encoding the data rate based on the determination output from the pattern determining means, an audio signal, additional information such as a header, and an error correction code. A digital video signal recording apparatus having information adding means for adding, wherein the data rate information is multiplexed on a header or the like or is written in a predetermined area on a recording medium. 16. When recording a digital video signal encoded by using motion compensation prediction and orthogonal transformation on a recording medium, image deterioration at the time of encoding and decoding based on the motion compensation prediction and orthogonal transformation. In a digital video signal recording device having a pattern determining means for determining the degree of the above and an information adding means for adding additional information such as an audio signal and a header and an error correction code, To the video signal, the first coding means for coding the video signal thinned out at a predetermined interval, and the residual component by the coding using the first coding means in the video signal. A second encoding means for performing encoding with respect to at least one of the first and second encoding means based on the determination output from the pattern determining means. A digital video signal recording device, characterized in that the encoding is performed by adaptively changing the data rate in one of the encoding means. 17. When recording a digital video signal coded by using motion compensation prediction and orthogonal transformation on a recording medium, image deterioration caused by coding and decoding based on the motion compensation prediction and orthogonal transformation. In a digital video signal recording device having a pattern determining means for determining the degree of the above and an information adding means for adding additional information such as an audio signal and a header and an error correction code, A first encoding means for encoding only the low frequency region subjected to the orthogonal transformation, and a residual component of the image signal encoded by the first encoding means. And a second encoding means for performing encoding of at least one of the first and second encoding means based on the determination output from the pattern determining means. A digital video signal recording device, characterized in that the encoding is performed by adaptively changing the data rate in one of the encoding means. 18. When recording a digital video signal encoded by using motion compensation prediction and orthogonal transformation on a recording medium, image deterioration when encoding and decoding based on the motion compensation prediction and orthogonal transformation. In a digital video signal recording device having a pattern determining means for determining the degree of the above and an information adding means for adding additional information such as an audio signal and a header and an error correction code, at least a coded image including an intra-frame coded image A video signal that is roughly quantized at a quantization level to code, and a residual component of the video signal that is coded using the first coding means. A second encoding means for performing encoding, and based on the determination output from the pattern determining means, at least one of the first and second encoding means A digital video signal recording apparatus, characterized in that the encoding is performed by adaptively changing the data rate in the other encoding means. 19. A digital video signal reproducing apparatus for reading data in which a data rate of coded data encoded based on motion compensation prediction and orthogonal transformation is adaptively variable according to a picture, in a normal reproduction mode. Based on the mode switching means for switching between the special reproduction mode and the reproduction mode, the data rate information extracting means for extracting the data rate information, and the data rate information output from the data rate information extracting means in the special reproduction mode. A digital video signal reproducing apparatus comprising: position information calculating means for calculating the position on the recording medium where the special reproduction data exists. 20. A digital video signal reproducing apparatus according to claim 19, further comprising head position converting means for controlling the head position at a position on the recording medium according to the output from the position information calculating means and the special reproduction speed. apparatus. 21. A digital video signal recording apparatus for recording a digital video signal encoded using motion compensation prediction and orthogonal transformation on a recording medium, wherein a digital coded based on the motion compensation prediction and orthogonal transformation is used. Encoding means for controlling and encoding the code amount corresponding to the area allocated to one image group formed by the video signal, and code amount comparing means for comparing the output from the encoding means with a predetermined code amount. A digital video signal recording device having data supplementing means for embedding extra data in a blank area of an image group having a blank area based on an output from the code amount comparing means. 22. Digital video signal data encoded based on motion compensation prediction and orthogonal transformation, and data of another image group is embedded in a blank area of an image group formed based on this encoded data. In a digital video signal reproducing apparatus for reading the data from a recording medium, a data reconstructing means for reconstructing the embedded video signal encoded data into an original image group, and decoding the reconstructed data from the data reconstructing means And a data decoding means for converting the digital video signal reproducing apparatus. 23. The data is encoded based on motion-compensated prediction and orthogonal transformation, and a predetermined array of first and second encoded data is arranged at a predetermined position of each image group data. A digital video signal reproducing apparatus for generating first and second decoded data corresponding to the first and second encoded data according to a condition, wherein a first and second data are decoded to obtain a reproduced image. A second decoding means for decoding the first data to obtain a reproduced image corresponding to the low frequency region of the intra-coded image or the number of thinned pixels or coarse quantization; Third decoding means for decoding the first data to obtain a reproduced image corresponding to at least the intra-frame coded image, the low frequency region of the inter-frame predictive coded image, or the number of thinned pixels or coarse quantization A digital video signal reproduction having at least one of the decoding means and a mode switching means for switching which of the decoding means to be used during special reproduction based on the special reproduction speed. apparatus. 24. Video data for extracting data corresponding to a video signal from a reproduction code in a digital video signal reproducing apparatus for reproducing video information coded based on motion compensation prediction and orthogonal transformation from a recording medium. Extraction means,
Video data decoding / reproducing means for decoding and reproducing the video data output from the video data extracting means, and a normal reproduction mode when reproducing the video data, or reproducing either one of an odd field and an even field A digital video signal reproducing apparatus having a mode switching means for switching between a display mode and a mode in which a field structure of an odd field or an even field is reversed to perform reproduction display. 25. Encoding is performed based on motion compensation prediction and orthogonal transformation, and the first and second encoded data are arranged at a predetermined position of each image group data,
A digital video for generating first and second decoded data corresponding to the first and second encoded data from data whose encoded data rate is adaptively variable according to a picture according to a predetermined condition. In the signal reproducing device, a first decoding means for decoding the first and second data to obtain a reproduced image, and a low frequency region or thinning of the intra-frame encoded image by decoding the first data. Second decoding means for obtaining a reproduced image corresponding to the number of pixels or coarse quantization, and at least the intra-frame coded image or the inter-frame predictive coded image in the low frequency region or thinning out by decoding the first data. At least one of the third decoding means for obtaining a reproduced image corresponding to the number of pixels or rough quantization is used, and any of the above decoding means is used for special reproduction. Digital video signal reproducing apparatus characterized by having a mode switching means for switching based on the trick play speed or.