JPH07222100A - Dynamic image reproduction device - Google Patents

Abstract

PURPOSE:To decode an image continuously even when input of data to a buffer 3 is interrupted in the device reproducing compressed dynamic image information. CONSTITUTION:The dynamic image reproduction device having a read device 2 reproducing dynamic image information of a CD-ROM 1 and outputting a bit stream of coded data, a buffer 3 storing the bit stream, a buffer level read device monitoring the data quantity of the buffer 3, a comparator 7 discriminating whether or not the data quantity reaches a threshold quantity and sending coded data to a decoding means (after VLD8) to start decoding is provided with a header analyzer 5 analyzing the coded data to calculate the data providing a threshold level and an adder 13 adding a prescribed value Ba to the data calculated by the header analyzer 5 to provide the sum to the comparator 7 as a threshold level.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for reproducing compressed moving image information recorded on a recording medium such as a CD-ROM or a digital video tape.

[0002]

2. Description of the Related Art When moving image information is recorded as digital data on a recording medium, the amount of information is very large, and as it is, only very short time data can be recorded. Therefore, a method has been proposed in which moving image information is compressed by high-efficiency encoding, recorded as encoded data, and then decoded at the time of reproduction.

As a high efficiency coding system, there is a compression system called MPEG (ISO, 11172) standardized by ISO. In MPEG, data for a plurality of frames is grouped into one group (GOP), and at least one frame of data is coded by intraframe compression coding processing with a relatively low compression rate (I picture), and the remaining frames. Are encoded by interframe compression encoding processing having a relatively high compression rate (P picture, B picture). In addition, as a high-efficiency encoding method, other than H.264. 261 and H.264. 21
There are standards for videophone such as 1st grade. According to such a method, the data amount is compressed with a relatively high image quality from a fraction of the original data amount to a few tenths, and in some cases, a few hundredths.

[0004]

The moving picture information recorded on the CD-ROM according to the MPEG standard is first encoded by the disc reading device 2 as shown in FIG.
And is temporarily stored in the buffer 3. This storage is performed at a fixed rate R [bit / sec].
Next, when the amount of data in the buffer 3 reaches a certain threshold value B0 (see FIG. 3), the reading by the buffer reader 4 is started, and thereafter, the encoding data is set at a time interval determined by the image rate. Data is read from the buffer 3 in frame units.
The encoded data read from the buffer 3 is V
After the variable length code is decoded by the LD8, the inverse quantizer 9
Is inversely quantized and is inversely DCT-transformed by the inverse DCT unit 10,
Furthermore, the output of the predictor 12 is added by the adder 11 as needed. The moving image data thus decoded is output to an external device.

The threshold value B0 is the sequence of the video sequence layer of the encoded data read from the CD-ROM 1.
It is calculated by the header analyzer 5 based on the parameter given as the buffer size BS in the can header. The calculated threshold value B0 is sent to the comparator 7. The amount of data in the buffer 3 is read by the buffer level reader 6 and sent to the comparator 6. The comparator 6 compares the two, and when the amount of data in the buffer 3 exceeds the threshold value B0 as a result, information to that effect is sent to the buffer reader 4. As a result, the buffer reader 4 starts reading data for decoding. After that, the coded data is read out in frame units at time intervals determined by the image rate. The read coded data is sent to the VLD 8 side and decoded.

As a result of inputting / outputting data to / from the buffer 3 as described above, the amount of data in the buffer 3 changes as shown in FIG. For this reason, the disk reader 2 is
When a data read error occurs and the input of data to the buffer 3 is interrupted, the data in the buffer 3 is lost,
In some cases, continuous playback of images may not be possible. It is an object of the present invention to enable continuous decoding of images even when the input of data to the buffer 3 is interrupted due to the occurrence of a data read error in the disk reading device 2. And

[0007]

SUMMARY OF THE INVENTION The present invention is a reproducing means for reproducing moving image information recorded on a recording medium and outputting a bit stream of encoded data, and the bit stream.
The buffer for storing the buffer and the data amount of the buffer are monitored, and when the data amount reaches a certain threshold value, the encoding data is sent to the decoding means to start the decoding. In a moving picture reproducing apparatus provided with decoding start control means; data analyzing means for analyzing the encoded data to calculate data for giving the threshold value; and data analyzing means. And a threshold control means for giving a value after the addition to the decoding start control means to the decoding start control means as a threshold value.

[0008]

According to the encoded data reproduced from the recording medium, the data analyzing means calculates the data which gives the threshold value for starting the decoding. A certain predetermined value is added to this calculated data by the threshold value control means, and the value after addition is given to the decoding start control means as the above threshold value. The decoding start control means compares the given threshold value with the data amount in the buffer. As a result, when the data amount in the buffer exceeds the threshold value, the decoding means controls the encoding in the buffer. Send data and start decoding.

[0009]

EXAMPLES Examples of the present invention will be described below. First, the data structure of the MPEG encoding standard for storage will be described with reference to FIG. 2, and subsequently, the data provided in the header of each layer will be described.
Will be described. The moving picture data of MPEG is, in order from the lower layer, a block layer, a macroblock layer, a slice layer,
It is configured as a picture layer, a GOP layer, and a video sequence layer, and the coding data is also configured in the same layer structure.

Each block is composed of 8 lines × 8 pixels. DCT (Discrete Cosine Transform) is executed in this block unit.

A macroblock is a C block which is located at the same position on the image as four luminance blocks adjacent to each other in the left, right, top and bottom directions.
It is composed of a total of 6 blocks including two color difference blocks b and Cr. The transmission order is Y0, Y1, Y2, Y
3, Cb, Cr. What is used for the prediction data (reference image data that takes the difference: made by forward prediction, backward prediction, bidirectional prediction, etc.), and whether it is not necessary to send the difference, etc. It is judged by.

A slice is composed of one or a plurality of macroblocks which are continuous in the scanning order of an image. The slice length and starting position are arbitrary and can be changed according to the error state of the transmission path.

The picture means each picture one by one, and is usually composed of three kinds of picture, I picture, P picture and B picture. The I picture is a picture in which all blocks in the image are composed of blocks encoded by DCT without taking a difference from the prediction data. The P picture is a DCT image without taking a difference from the prediction data in the image.
This is a picture in which a block coded by (1) and a block coded by DCT of the difference between the image data located ahead in time with the input and already decoded are mixed. The B picture is a difference between a block coded by DCT without taking a difference from the prediction data in the picture and a picture data which is located ahead in time at the input and already decoded. Is a DCT-coded block, and the difference between the image data that has been temporally located at the input and has already been decoded is the DCT.
D is the difference between the block encoded in step S1, the image data that is temporally forward and already decoded, and the two interpolated images of image data that is temporally posterior and already decoded.
The block coded by CT is a mixed picture.

A GOP is one or more I pictures and 0.
A group of pictures consisting of one or more non-I pictures
It is The video sequence is composed of one or a plurality of GOPs having the same image size and image rate.

The video sequence layer is a sequence star.
It begins with a to-code and ends with a sequence-end code.
One or more sequence header sections are included between them, and each sequence header section is followed by one or more GOP layer data. The sequence header section starts with a sequence header code, and has an image rate, a bit rate of encoded data, and a buffer size necessary for decoding. Also, the number of horizontal and vertical pixels in the display area of the screen, the aspect ratio of the pixel interval,
Each parameter has a flag indicating whether or not it is below a certain constraint, and further has information such as a quantization matrix as an option.

Each GOP layer begins with a group start code. Next, the time code of the first picture of the GOP (code indicating the time from the beginning of the sequence), a flag indicating whether or not data of another GOP is necessary for reconstructing the image of the GOP, A flag indicating whether or not the data of the preceding GOP can be referred to follows, and then the picture layer data of the picture included in the GOP follows.

The picture layer begins with a picture start code. Next, a number indicating the display order of each picture, the coding type (I, P, B) of the picture, and the time required to store the coding data required at the time of displaying the picture in the buffer. , Motion vector accuracy, motion vector range, followed by one or more slice layer data.

The slice layer begins with a slice start code. Next follows the data that gives the quantisation width, followed by one or more macroblock layer data.

In the macroblock layer, dummy data for bit rate adjustment, data for controlling macroblock addressing indicating an absolute position of a macroblock in a picture, macroblock type, quantization width, and, if necessary, a front-back direction A prediction motion vector and data indicating which block of data is used as encoded data are arranged, followed by block layer data.

The block layer contains DC components and A after DCT.
The data obtained by entropy coding the C component is Y0, Y.
1, Y2, Y3, Cb, Cr are arranged in this order.

Next, the structure of the embodiment apparatus will be described with reference to FIG. This apparatus reads moving image information recorded on a CD-ROM 1 by a disk reader 2 and generates a bit stream of encoded data at a fixed rate R [bit / sec], which is then generated. , Once stored in the buffer 3, then read in accordance with the signal from the comparator 7 and read out by the decoding unit (VLD8,
Inverse quantizer 9, inverse DCT 10, adder 11, predictor 1
It is a device for sending to (2) and decoding and outputting to moving image data.

First, the data stored in the buffer 3 at the fixed rate R is read by the buffer reader 4 and sent to the header analyzer 5. In the header analyzer 5, the data of the picture layer (the time required to store the coded data required at the time of displaying the picture in the buffer 3):
Buffer 3 at the start of decoding based on the parameter BF).
An amount of data (threshold value for starting decoding) B0 that should be stored in B0 is calculated. That is, since the data is stored in the buffer 3 at the fixed rate R, the parameter BF can be converted into the data amount B0.

The above data amount B0 calculated by the header analyzer 5 is then sent to the adder 13 to have a predetermined value Ba.
Is added, and the value “B0 + Ba” after this addition is sent to the comparator 7 as a threshold value for starting decoding. Comparator 7
The data from the buffer level reader 6 for monitoring the amount of data in the buffer 3 is input to.

The comparator 7 compares the threshold value B0 + Ba with the amount of data in the buffer 3, and when the amount of data in the buffer 3 exceeds the threshold value B0 + Ba, the buffer reader 4 "decodes". Signal to start conversion ”. In response to this, the buffer reader 4 reads out the data of the buffer 3.
It is read in units of frames and sent to VLD8. This allows VL
A known decoding process is performed in the decoding unit after D8, and the decoded data is output to the outside. After the decoding is started, the reading of the data from the buffer 3 is executed in frame units at every time determined by the image rate.

As a result of reading the data from the buffer 3 in this way, the amount of data in the buffer 3 of this apparatus changes as shown in FIG. That is, the data in the buffer 3 is
When the data amount exceeds "B0 + Ba", decoding is started and the data of the first picture is read out, and thereafter, the data is read out in frame units at each time determined by the image rate. Be decrypted. In addition, as shown in FIG.
As the capacity of the buffer 3, the added value B is larger than the original capacity B.
The capacity "B + Ba" which is large by the amount of "a" is secured.
The original capacity B is the capacity of a conventional device which is defined based on the data (buffer size required for decoding: parameter BS) in the sequence header. A value obtained by adding the addition value Ba to B, “B +
“Ba” is secured as the capacity of the buffer 3.

Next, a read error in the disk reading device 2
The case where the input of the bit stream to the buffer 3 is interrupted by the above will be described with reference to FIGS.

As shown in FIG. 5, when the input of the bit stream to the buffer 3 is interrupted at time t1, the amount of data in the buffer 3 does not increase thereafter. In this state the next frame
When the memory data is read, the amount of data in the buffer 3 decreases by the amount of the amount of data, but at this time, the amount of data in the buffer 3 is larger than Ba. However, when the frame data is read out one after another at time t2, the amount of data in the buffer 3 becomes smaller than Ba. However, the amount of data in the buffer 3 is not "0". As described above, in the conventional device, the amount of data in the buffer 3 becomes "0" at time t2, and it is impossible to continuously decode the image. However, in the case of the present device, Since there is a margin for the addition value Ba, it is possible to continuously decode.

FIG. 6 shows that the amount of data in the buffer 3 monitored by the buffer level reader 6 is the time t2.
As described above, when the addition value Ba falls below the above-mentioned addition value Ba, it is judged that "error has occurred", and a warning to that effect is sent to the buffer reader 4 to interrupt the reading of data from the buffer 3. Here, for display, it is assumed that the immediately previous image data is repeatedly used. By processing in this way, the amount of data in the buffer 3 can be recovered. In FIG. 6, waiting (data to the buffer 3 is
The accumulation of data is for two frames, but it may be waited until the amount of data in the buffer 3 exceeds "B0 + Ba". It should be noted that the "error occurrence" can be detected when the amount of data in the buffer 3 becomes "0" in the past, too.
In that case, there is a problem that the decoding is interrupted at that point.

Therefore, the "error occurrence" is detected before the amount of data in the buffer 3 becomes "0", and the above-described action is taken (the reading of the data from the buffer 3 is interrupted, If it is desired to repeatedly use the immediately preceding image data for display), the device may be configured as follows.

That is, reproducing means for reproducing the moving picture information recorded on the recording medium and outputting the bit stream of the encoded data (the disc reading device 2. or other means such as a digital video tape). (Recording medium reading device), the buffer 3 for storing the bit stream, the amount of data in the buffer 3, and when the amount of data reaches a certain threshold value B0, the decoding means (VLD8) , Inverse quantizer 9, inverse DCT device 1
0, adder 11, predictor 12) and decoding start control means (buffer level reader 6, comparator 7, buffer reader 4) for sending the above encoded data to start decoding. In a moving image reproducing apparatus; data analyzing means (header analyzer 5) for analyzing the encoded data to calculate data giving the threshold value B0, and calculating by the data analyzing means Threshold control means (adder 8, generating or inputting a predetermined value Ba) which adds a predetermined value Ba to the obtained data and gives the value after the addition as the threshold to the decoding start control means. (Means for giving a predetermined value Ba), and further, when the data amount of the buffer 3 monitored by the buffer level reader 6 falls below the above value Ba in the decoding start control means, the buffer reader 4 From buffer 3 by
The reading of the data is interrupted, and a function of sending a signal to the external device to repeatedly use the immediately preceding image data as the image data for display is provided.

[0031]

As described above, according to the present invention, a certain predetermined value is added to the threshold value for starting decoding, which is calculated based on the encoded data read from the recording medium, and the value after the addition is the threshold value. Used as. Therefore, the buffer data at the start of reading (at the start of reading for decoding) is added by the additional value.
The amount of data becomes larger than before. Therefore, for example, the data in the buffer is read by the read error in the disk reading device.
Even if the input of data is interrupted, the margin until the data in the buffer is lost becomes large, and the possibility that the image can be continuously decoded increases.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an apparatus according to an embodiment.

FIG. 2 is a data configuration diagram showing an MPEG data structure layer by layer.

FIG. 3 is an explanatory diagram showing changes in the amount of data in the buffer for a conventional device.

FIG. 4 is an explanatory diagram showing changes in the amount of data in the buffer for the embodiment apparatus.

FIG. 5 is an explanatory diagram showing changes in the amount of data in the buffer for the apparatus of the embodiment, and shows a case where the input of data to the buffer is interrupted.

FIG. 6 is an explanatory diagram showing changes in the amount of data in the buffer for the apparatus of the embodiment, and shows a case where the same image is read when the input of data to the buffer is interrupted.

FIG. 7 is a block diagram showing a configuration of a conventional device.

[Explanation of symbols]

 1 CD-ROM

Claims (1)

[Claims] 1. A reproducing means for reproducing moving picture information recorded on a recording medium to output a bit stream of encoded data, and a buffer for storing the bit stream.
Decoding start control means for monitoring the amount of data in the buffer, and sending the encoded data to the decoding means to start decoding when the amount of data reaches a certain threshold value. In a moving image reproducing apparatus provided with the data analysis means, which analyzes the encoded data to calculate data for giving the threshold value, and data calculated by the data analysis means. To the decoding start control means, and a predetermined value is added as a threshold value to the decoding start control means.