JP2000224545A - Moving picture recorder and moving picture reproducing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce a data size as by leaving digital data as it is without re- encoding MPEG(moving picture expert group) encoding data. SOLUTION: A moving picture processor is provided with an information reducing means 102 reducing prescribed information from inputted moving picture data and an information reducing quantity managing means 103 managing reduced information quantity. When a residual record able capacity onto an information recording medium is smaller than the size of MPEG coding data to receive, prescribed information, which is included in an inter-frame predictive coding block within MPEG coding data and deterioration in the image quality of which is comparatively decreased is reduced to reduce the data size of MPEG encoding data.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an MPEG (Moving P), an international standard for moving picture coding for storage media.
The present invention relates to a moving image recording apparatus for recording an encoded image on a storage medium and a moving image reproducing apparatus for reproducing the encoded image from the storage medium.

[0002]

2. Description of the Related Art There are several MPEG standards, such as the MPEG1 standard, which is a moving image coding standard for storage media, and the MPEG2 standard, which is a general-purpose coding standard applied to various applications such as broadcasting, communication, and storage media. Such standards are included.

In general, the MPEG coding technique uses predictive coding in which a signal value of a pixel constituting a certain frame is represented as a motion vector using a difference between a signal value of a pixel in a preceding frame and a signal value of a pixel in a preceding frame. In predictive coding, there are forward prediction using only past reproduced images, and bidirectional prediction using forward prediction from past reproduced images and backward prediction from future reproduced images. In order to realize this bidirectional prediction, three types of picture types, i.e., I picture, P picture, and B picture, are defined in the MPEG coding as frame units.

FIG. 2 shows the structure of MPEG-encoded image data. That is, one image data is used as a sequence layer, and a sequence header and a GOP (Group of Pic
tures). MPE
In G encoding, since encoding is performed based on data of preceding and succeeding frames, a unit called a GOP, which is a group of a plurality of frames, is determined to enable random access.

That is, the GOP layer includes the above-mentioned I picture, P
Each picture is divided into units called slices. A slice is composed of one or more macroblocks, and each macroblock is composed of four luminance blocks and two chrominance blocks. The luminance block is composed of 16 × 16 pixels, and the chrominance block is composed of 8 × 8 pixels.

[0006] In an I-picture, all the macroblocks constituting one frame are constituted only by intra-coded blocks which do not use the temporal correlation with the preceding and succeeding frames. The P picture is composed of an intra-frame coded block and an inter-frame forward prediction coded block. A B picture is composed of an intra-frame coded block, an inter-frame forward prediction coded block, an inter-frame backward prediction coded block, and an inter-frame bi-directional predicted coded block.

FIG. 3 shows an outline of a format structure of a macroblock layer and a block layer in MPEG1 encoding, and the main parts will be described below. In the macroblock layer of FIG. 3, MBTYPE (Macroblock Ty
pe) indicates the type of the above-described block. For example, in the case of a B picture, two or three types of prediction directions (forward, reverse, Direction, bidirectional), there are three types of motion vector only, motion vector + transformation coefficient, motion vector + transformation coefficient + quantization step update, so there are a total of 11 types. Note that the transform coefficient is a value obtained by performing a DCT (Discrete Cosine Transform) on a pixel value included in an intra-frame coded block, and for other blocks, an image predicted by a motion vector and an actual image. This is for interpolating the difference, and is obtained by subjecting the difference to DCT. MHF (Mot
ion Horizontal Forward Code), MVF (Motion Vertic
al Forward Code) is used when MBTYPE is interframe forward and bidirectional encoding, and the horizontal component of the forward motion vector of the macroblock, respectively.
Represents the vertical component. MHB (Motion Horizontal Backward
Code), MVB (Motion Vertical Backward Code) is MB
TYPE is used in the case of inter-frame reverse direction and bidirectional coding, and represents the horizontal component and the vertical component of the reverse motion vector of the macroblock, respectively. CBP
(Coded Block Pattern) indicates whether six blocks in the macro block have transform coefficients. If MTYPE is an intra-coded macroblock, it is obvious that transform coefficients exist in all six blocks in the macroblock, so CBP is unnecessary. In other MTYPE, CBP is It indicates whether a transform coefficient exists in the block.

Next, the structure of the transform coefficient in the block layer of FIG. 3 is shown. DDSL (DCT DCSize Luminance), D
DSC (DCT DC Size Chrominance), DDCD (DCT DC D
(ifferential) is used in the case of an intra-coded macroblock, and indicates the difference between the DC components of the transform coefficients obtained by the DCT transform. DDSL indicates the number of bits of the difference of the luminance block, and DDSC indicates the number of bits of the difference of the chrominance block. The DDCD represents the actual difference value according to the number of bits. DCF (DCT CoefficientFirst) indicates a DC component itself other than the intra-coded macroblock. DCN (DCT Coefficient Nex
In (t), DCT coefficients of 8 × 8 pixels constituting one block are encoded in a predetermined order by a combination of a run (consecutive zeros) and a level (a non-zero value thereafter).

FIG. 4 shows the configuration of a video encoder and video decoder based on MPEG. In the video encoder of FIG. 4, the source encoder performs compression of the video signal,
CT (Discrete Cosine Transform) conversion, quantization, motion compensation, bidirectional prediction, and the like are performed. The video signal multiplexing encoder generates the data format shown in FIG. 2 from the compressed image data, and the transmission buffer is a memory for controlling the transmission data amount. The system multiplexing encoder synchronizes video, audio, and text data.

In a video decoder, a system multiplex decoder separates video, audio, and text data, and a reception buffer is a buffer for guaranteeing decoding processing time for received data. The video signal multiplexing decoder performs extraction of compressed image data, and the information source decoder performs inverse DCT transform, inverse quantization,
The original video signal is restored by motion compensation, bidirectional prediction, or the like.

[0011]

However, when moving image data encoded in this way is recorded in a moving image recording apparatus, there are the following problems.

That is, if the remaining recordable capacity of the information recording medium mounted on the moving image recording apparatus is smaller than the data amount of the moving image to be recorded from now on, it is difficult to record all of the data. Or the received MPE
After once decoding the G-coded data into the original video signal, it was necessary to re-encode using a video encoder so as to have a recordable capacity. Usually, a moving picture recording device is equipped with only a video decoder, so mounting a video encoder leads to an increase in cost, and the amount of calculation for encoding by the video encoder is enormous. It takes time.

An object of the present invention is to solve the above-mentioned problems, and to reduce the recording capacity of MPEG encoded data without decoding and re-encoding received MPEG encoded data.

[0014]

According to a first aspect of the present invention, there is provided a moving image recording apparatus, comprising: an information reducing unit configured to reduce predetermined information in the moving image input to the moving image recording apparatus; And information reduction amount management means for managing the reduced information amount, based on the recordable capacity on the information recording medium, the data amount of the received moving image, and the information reduction amount from the received moving image. Then, it is determined whether or not the amount of information is to be reduced by the information reducing means. If the amount of information is to be reduced, information of a predetermined area is reduced from the received moving image.

According to a sixth aspect of the present invention, there is provided a moving picture reproducing apparatus which is used when a moving picture recording apparatus according to the first aspect records macroblocks in a B picture by reducing the number of macroblocks. The information interpolating means in the moving image reproducing device,
The information specifying the macroblock reduced in the B picture is referred to from a predetermined area in the moving image, and the information of the motion vector in the reduced macroblock is obtained by using the I picture or P picture before and after the B picture.
Interpolate based on information included in the picture.

[0016]

Embodiments of the present invention will be described below with reference to the drawings.

(Embodiment 1) Embodiment 1 of the present invention will be described using a moving image recording apparatus shown in FIG.

In FIG. 1, reference numeral 101 denotes data receiving means for receiving MPEG encoded data from the outside, and 102 denotes information reducing means for reducing the amount of information in the received MPEG encoded data. Reference numeral 103 denotes an information reduction amount management unit that manages the amount of information reduced by the information reduction unit 102. The information reduction amount management unit 103 also determines whether the information reduction unit 102 should continue reducing information. Is shown. An information recording unit 104 records MPEG encoded data other than the information reduced by the information reducing unit 102 on the information recording medium 105. An overall control unit 106 controls the entire moving image processing apparatus.

Next, the operation of the moving picture processing apparatus thus configured will be described. Before receiving the MPEG encoded data, the overall control unit 106 on the moving image processing apparatus acquires the entire data amount of the MPEG encoded data to be received through the data receiving unit 101. Further, information about the remaining capacity that can be recorded on the information recording medium 105 is obtained through the information recording unit 104. At this time, if the remaining recordable capacity on the information recording medium 105 is larger than the amount of MPEG encoded data to be received, the MPE by the information reducing means 102
The received data is recorded on the information recording medium 105 without reducing the predetermined information of the G encoded data.

Next, if the remaining recordable capacity on the information recording medium 105 is smaller than the amount of MPEG encoded data to be received, the following processing is performed.

That is, first, a value obtained by subtracting the recordable capacity on the information recording medium 105 from the data amount of the MPEG encoded data to be received is set in the information reduction amount management means 103.
Next, in the information reduction means 102, the DCF of the block layer shown in FIG. 3 in the inter-frame prediction coding macroblock included in the B picture for the MPEG coded data having the configuration shown in FIG. The information about the transform coefficient of the DCN part is deleted, and the information is passed to the information recording unit 104. In FIG. 3, as described above, CBP is a predetermined M
The information reduction means 102 is effective in the case of TYPE, and specifies a block including a transform coefficient in a macroblock. CBP is invalidated by rewriting the CBP.

The information value reduced by the information reduction means 102 is sequentially subtracted from the value set in the information reduction amount management means 103, thereby updating the set value of the information reduction amount management means 103. Then, the updated information reduction amount management means 103
Is less than or equal to 0, the remaining MPEG encoded data can be recorded on the information recording medium 105 as it is, so that the information reducing means 102 does not reduce the predetermined information.

Here, I picture included in one GOP,
The average of the ratio of the P picture and the B picture is a: b: c, and the average of the ratio of the sizes of the I picture, P picture, and B picture per image is d: e: f. Further, it is assumed that all the macroblocks included in the B picture are inter-frame prediction coded macroblocks, and the average of the ratio of the transform coefficients in one macroblock is m (0 <m <1). in this case,
The amount of reduction in recording size due to recording with all transform coefficients in each macroblock in a B picture reduced is
cfm / (ad + be + cf). For example, a:
b: c = 4: 2: 1, d: e: f = 1: 4: 10, m = 2/3
In this case, the reduction amount of the recording size is 10/33, which is about a 30% reduction.

In the first embodiment, the information to be reduced by the information reducing means 102 is the transform coefficient of the block included in the inter-frame predictive coding macroblock in the B picture. It is also possible to reduce the number of blocks including the transform coefficients of the blocks included in the predictive coding macroblock.

Further, in the first embodiment, when the value set in the information reduction amount management means 103 reaches a predetermined threshold value, a change is made so that normal recording is performed. It is also possible to continue to reduce the amount of information.

Further, in the first embodiment, the information to be reduced by the information reducing means 102 is added to the conversion coefficients of the blocks included in the inter-frame predictive coding macroblock in the B picture, and the MHF, MVF, M
It is also possible to reduce the information amount together with the motion vector information indicated by HB and MVB.

Here, assuming that the deletion of the motion vector information in the macroblock is performed on a picture basis, a picture is constituted as shown in FIG. That is, a picture is originally composed of a picture header and picture data, and the picture header includes a picture start code as a start code of a picture layer, a temporal reference indicating the order of pictures in a GOP, and a picture for distinguishing I, P, and B pictures. Includes coding type, etc. In this picture header, a motion vector deletion flag is provided as information indicating that the motion vector information included in each macroblock in the picture has been deleted, as shown in FIG.

When this method is adopted, if the average m of the ratio of the sum of the motion vector information and the transform coefficient in one macroblock is 11/12, the recording size when the above-mentioned numerical value is applied is assumed. Is reduced to 5/12, about 4
It is a decrease in percentage.

In the first embodiment, the information reduced by the information reducing means 102 can be the entire B picture. When playing back MPEG encoded data in which the entire B picture has been deleted and recorded, in the picture header of each picture, there are missing temporal reference values which usually increase by one in one GOP. Therefore, when the video decoder detects that the temporal reference value is missing, the picture of the I picture or P picture reproduced immediately before in the time axis method from the picture that caused the missing is displayed as it is. To play.

(Embodiment 2) Next, Embodiment 2 of the present invention will be described using a moving picture reproducing apparatus shown in FIG.

In FIG. 5, reference numeral 501 denotes an information recording medium.
Reference numeral 02 denotes a system multiplex decoder that separates video data, audio data, and text data from MPEG encoded data recorded on the information recording medium 501. 50
Reference numeral 3 denotes a reception buffer that stores the video data separated by the system multiplex decoder 502. 504 is receive buffer 5
This is a video signal multiplexing decoder that cuts out data after MPEG compression from the video data in 03. Reference numeral 505 denotes information interpolating means for interpolating the MPEG-compressed data extracted by the video signal multiplexing decoder 504. Reference numeral 506 denotes an information source decoder that decodes the MPEG-compressed data interpolated by the information interpolation unit 505 into an original video signal.

Next, the operation of the moving picture reproducing apparatus thus configured will be described.

The MP recorded on the information recording medium 501
It is assumed that the EG coded image may have the information of the transform coefficient and the motion vector of the block included in the inter-frame predictive coding macroblock in the B picture deleted. Here, assuming that the deletion of the motion vector information in the macroblock is performed on a picture basis, a picture is configured as shown in FIG.

After the moving image data thus constructed is read from the information recording medium 501, the system multiplex decoder 502, the reception buffer 503, and the video signal multiplex decoder 504
The information interpolating means 505 interpolates the moving image data transmitted through the above as follows.

FIG. 7 shows an example in which a B picture in which a motion vector deletion flag is set in a picture header, and an I picture and a P picture immediately before and after the B picture.

Here, the horizontal component of a motion vector of a macroblock having a P picture is m, the vertical component is n, and the horizontal component of a motion vector to be interpolated in a B picture is m.
1, and the vertical component is n1. In addition, temporary references of I picture, B picture, and P picture
1, X2 and X3. In this case, by distributing the motion vector value of the P picture proportionally, the horizontal component m1 of the macroblock located at the same position as the P picture in the B picture is (X2-X1) m / (X3-X1), and the vertical component n1 is (X2-X1) n / (X3-X1). In the same manner, motion vector interpolation is performed for all the inter-frame predictive coding macroblocks in the B picture.

In the second embodiment, the area in which the motion vector deletion flag is set is a picture header. However, the same effect can be obtained by setting it in another layer constituting an MPEG encoded image.

[0038]

According to the first embodiment of the present invention, the received MPEG-encoded data is recorded by excluding data having relatively little effect on the image quality.
It is possible to reduce the recording size of the MPEG encoded data. As a result, according to the conventional method, there is an effect that recording can be performed even on an information recording medium on which recording was not possible because the remaining recording time was insufficient by a predetermined amount.

Also, according to the second embodiment of the present invention, the transform coefficients included in the inter-frame predictive coding macroblock in the B picture from the moving picture data compressed by the MPEG according to the first embodiment of the present invention. And when reproducing data recorded with a motion vector deleted, by interpolating the information of the motion vector in the B picture based on the information of the I picture or P picture before and after the B picture, Has the effect of reducing the deterioration of

[Brief description of the drawings]

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

FIG. 2 is a diagram showing a format configuration of MPEG encoded data.

FIG. 3 is a diagram showing a format configuration of a macroblock layer and a block layer.

FIG. 4 is a diagram showing a block configuration of MPEG encoding and decoding.

FIG. 5 is a diagram showing a block configuration of a moving image reproducing apparatus according to Embodiment 2 of the present invention.

FIG. 6 is a diagram showing a configuration of a picture header.

FIG. 7 is a diagram showing a motion vector interpolation method according to the second embodiment of the present invention;

[Explanation of symbols]

 101 data receiving means 102 information reduction means 103 information reduction amount management means 104 information recording means 105 information recording medium 106 overall control means 501 information recording medium 502 system multiplex decoder 503 reception buffer 504 video signal multiplex decoder 505 information interpolation means 506 Source decoder

 ────────────────────────────────────────────────── ─── Continuing on the front page (72) Inventor Hideshi Ishihara 1006 Kazuma Kadoma, Osaka Prefecture Matsushita Electric Industrial Co., Ltd. F term (reference) 5C053 GB06 GB07 GB08 GB11 GB19 GB21 GB23 GB30 GB37 GB38 HA40 JA30 KA22 KA24 5C059 KK01 KK08 LB13 MA00 MA04 MA05 MA23 MC32 NN11 NN28 PP05 PP06 PP07 PP16 RB01 RC16 RC32 SS11 SS30 TA00 TA49 TB04 TB07 TC39 TD11 UA05

Claims (6)

[Claims] 1. A moving image recording apparatus for recording a compressed moving image on an information recording medium, comprising: information reducing means for reducing predetermined information in the moving image input to the moving image recording apparatus; A moving image recording apparatus comprising: an information reduction amount management unit that manages an information amount reduced by the information reduction unit. 2. The moving picture according to claim 1, wherein the information to be reduced by the information reducing means is a transform coefficient of each block in an inter-frame predictive coding macroblock constituting the moving picture. Recording device. 3. The information reduced by the information reduction means is a transform coefficient of each block in an inter-frame predictive coding macroblock included in a B picture constituting the moving image. Item 3. The moving image recording device according to Item 1. 4. The moving picture recording apparatus according to claim 1, wherein the information to be reduced by said information reducing means is a B picture constituting said moving picture. 5. The information to be reduced by the information reducing means is a transform coefficient and motion vector information of each block constituting an inter-frame predictive coding macroblock, which is included in a B picture constituting the moving image, The moving picture recording apparatus according to claim 1, wherein information for specifying the macro block including the reduced motion vector information is added to a predetermined area in the moving picture. 6. A moving picture reproducing apparatus for reproducing a compressed moving picture from an information recording medium, comprising: data which is multiplexed and decoded by a video signal; and data which is obtained by interpolating motion vector information. A moving image reproducing apparatus including an interpolation unit, wherein the information interpolation unit refers to information for specifying the macroblock including the reduced motion vector in the B picture from a predetermined area in the moving image. A moving image reproducing apparatus for interpolating the information of the reduced motion vector in the macroblock based on information included in I and P pictures before and after the B picture.