JPH0272779A - Image encoding system - Google Patents

Abstract

PURPOSE:To efficiently encode an image signal by dividing an input image signal into block lines consisting of plural lines, and deciding an encoding means at every unit of block line. CONSTITUTION:The image signal inputted to an input terminal 1 is divided into the block lines consisting of the plural lines at a block division part 2. When the encoding means to use the interframe correlation of a moving picture is selected, difference between the image signal and a predicted image using a preceding frame is taken. Also, when the encoding means to use in-frame correlation is selected, the image signal is sent to a DCT transformation part 3 as it is, and DCT transformation that is one of the orthogonal transformation is applied on the signal. After that, each transform coefficient is quantized by a quantizer 4, and is sent to a variable length encoder circuit 9, and is variable length-encoded, and is outputted to an output terminal 10. By employing such system, it is possible to attain high compressibility, and also, perform the processing of fast forward directional reproduction or fast backward directional reproduction.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an image encoding method for highly efficiently encoding moving images and storing the encoded images on a compact disc (CD) or the like.

[Prior art]

Conventionally, as a method for randomly accessing and decoding accumulated encoded image data, or performing high-speed forward playback or high-speed reverse playback, there are two methods: (a) using only intra-frame correlation for all frames; How to encode.

(b) Three-dimensional discrete cosine (hereinafter referred to as DCT
There are methods such as using conversion (called ``transformation'').

An example of method (b) is to estimate the movement direction between frames, move each frame in the estimated movement direction, and then configure blocks three-dimensionally in the horizontal direction, vertical direction, and time axis direction. , there is one that performs three-dimensional DCT transform and encodes each block. Regarding this, for example,
2r+d Internationals orks
hop On S, tgnal Processin
g of HDTV V, Vol, 1. rAdapt-i
ve Transform Coding of HD T
V picturesJ.

Here, intraframe encoding refers to performing encoding processing of image information only within each frame. In addition, interframe encoding refers to encoding in three dimensions (horizontal direction, vertical direction, 1. time axis direction) that also utilizes the correlation between consecutive frames.

DCT transformation is one of the orthogonal transformations and is a transformation that uses only a removal function, and mainly uses the electric cassette vector density on the coefficients after the transformation to achieve efficient encoding. , refers to using only the removal function.

[Problem to be solved by the invention]

However, the method (a) has problems such as not being able to utilize inter-frame correlation of moving images and not being able to achieve high compression.

Furthermore, the method (b) has problems such as an enormous number of encoding circuits and decoding circuits.

In addition, within the frame conventionally used for television (TV) conferences etc., the frame l? In the method of using iJ correlation according to predetermined rules, random access and high-speed forward playback are possible because the intra-frame encoding and inter-frame encoding of moving images are switched only from the viewpoint of encoding characteristics.

There were problems such as high-speed reverse playback being impossible.

The present invention has been made to solve the above problems.

An object of the present invention is to provide an image encoding method that can efficiently encode image signals by using the image encoding method used in TV conferences and the like.

Another object of the present invention is to provide an image encoding method that can randomly access and decode encoded data.

Another object of the present invention is to provide an image encoding method that allows high-speed reproduction in forward and reverse directions.

The above and other objects and novel features of the present invention will become apparent from the description of this specification and the accompanying drawings.

[Means to solve the problem]

In order to achieve the above object, the present invention provides a high-efficiency image encoding method that switches between a means for encoding using correlation within a frame of a moving image and a means for encoding using a correlation between frames of a moving image. In.

block dividing means for dividing an image signal into block lines each consisting of a plurality of lines; Flag generation means for determining whether the information is encoded and generating flag information corresponding to the determination result.

Using the flag information, it is determined that the block line is encoded at predetermined block line intervals by an encoding means that utilizes the intra-frame correlation of the moving image, and other block lines are encoded using the inter-frame correlation. encoding switching means that determines to perform encoding using an encoding means that utilizes the encoding means; and means that encodes using intra-frame correlation or inter-frame correlation of a moving image based on the output of the encoding switching means. The most important feature is that it has the following.

In addition, the code amount for each block line is calculated, and the code from the block line coded using the intra-frame correlation of the moving image to the block line coded using the next intra-frame correlation is calculated. The present invention is characterized in that it includes encoding switching means that determines the encoding means so that the sum of the amounts satisfies a predetermined value.

[Effect]

According to the above means, during encoding, the input image signal is divided into block lines consisting of a plurality of lines by the block dividing means, the encoding means is determined for each block line by the encoding switching means, and the encoding means is determined based on the determined result. Correspondingly, the flag generation means generates a flag, and the encoding is performed using the intra-frame correlation or inter-frame correlation of the moving image based on the determination result of the encoding means, so that the image signal can be efficiently encoded. can be converted into

In addition, the code amount for each block line is calculated, and the total code amount from the block line coded using intra-frame correlation to the next block line coded using intra-frame correlation is calculated. By providing a coding switching means that determines the coding method so as to satisfy a predetermined error value, coded data can be randomly accessed and decoded, and the coded data can be accessed and decoded in the forward and reverse directions. can be played at high speed.

[Embodiments of the invention]

An embodiment of the present invention will be specifically described below with reference to the drawings.

In addition, in all the figures for explaining the embodiment, parts having the same functions are given the same reference numerals, and repeated explanations thereof will be omitted.

[Example ■]

FIG. 1 is a block diagram showing a system configuration of an image encoding method according to Embodiment 2 of the present invention.

In FIG. 1, 1 is an input terminal, 2 is a block dividing section,
3 is a DCT conversion unit, 4 is a quantizer, 5 is a code switching unit, 6 is a flag generation unit, F is a frame memory, 8 is a motion compensation processing unit, 9 is a variable length encoding circuit, 10 is an output terminal, 1
1 is an output terminal.

The image encoding method of this embodiment (■) is as shown in FIG.
The image signal input to the input terminal 1 is sent to the block dividing section 2.
The line is divided into block lines each consisting of multiple lines. For the image signal of that block line, if a coding method that uses inter-frame correlation of moving images is selected, a difference is calculated from a predicted image that uses the previous frame, and a code that uses intra-frame correlation is calculated. If the converting means is selected, the data is sent as is to the DCT transform unit 3, and D
CT conversion is performed. Then, each transform coefficient is converted into a quantizer 4
The signal is quantized at , and sent to the variable length encoding circuit 9 where it is variable length encoded and output to the output terminal 10 . In addition, when the output of the quantizer 4 is encoded using interframe correlation, the sum with the predicted image signal is taken to generate the same decoded image signal as that decoded on the receiving side, and the motion It is sent to the frame memory 7 for use in compensation processing. The code switching section 5 measures the number of block lines sent out from the block dividing section 2, and switches the encoding means for the current block line between intra-frame and one using inter-frame correlation. Specifically, the predicted image signal sent from the frame memory 7 is usually subtracted from the input image signal, but
The predicted image signal sent from the frame memory 7 is cut for every fixed number of block lines, and the input image signal is directly sent to the DCT conversion section 3. Information on the encoding type of the code switching section 5 is sent to the flag generating section 6, where a uniquely detectable flag is generated and outputted to the output terminal 11. The frame memory 7 stores the encoded image of the previous frame, and the motion compensation processing unit 8 performs motion compensation processing on the current image signal and the image signal of the previous frame to form a predicted image signal. . The motion predicted image obtained by the motion compensation unit processing 8 is stored in the frame memory 7, and
The signal is sent to the code switching unit 5.

The motion compensation process, for example, divides an image signal into blocks each consisting of a predetermined plurality of pixels, and divides a certain range between some pixels included in the block and a block of the encoded image. In this method, block matching is performed within the blocks, and a motion vector indicating the direction in which the blocks have moved is determined from between the blocks where the error between the blocks is the smallest.

A predicted image signal of the current image signal can be constructed from the blocks with smaller errors.

In the above-mentioned embodiment (2), motion compensation processing was performed as a method of utilizing inter-frame correlation, but of course.

Processing other than this, for example, processing that takes inter-frame differences may also be used.

Next, the effects of the image encoding method of Example I will be explained using FIG. 2.

FIG. 2 is a diagram for explaining the relationship between frames and block lines in the image encoding method of Example 2,
21 is the first frame, 22 is the second frame, 23 is the third frame
Frame 24 is the fourth frame, and 25 is a portion encoded using intra-frame correlation.

According to the image encoding method of the first embodiment, as shown in FIG. 2, one frame of input signal is divided into N block lines, and the code is switched using intra-frame correlation every N+1 blocks. An encoding means is selected. In this case, the first block line is the first block line in the first frame 21, the second block line is the second block line in the second frame 22,
Block lines encoded using intra-frame correlation appear in sequence. During random access, consecutive N
By decoding these frames, the subsequent encoded data can be completely decoded. In addition, during high-speed playback, if a flag is detected and only the encoded part is decoded using intra-frame correlation, and one frame is constructed by connecting block lines of N frames, both forward and reverse directions are fast. Can be played.

[Example ■]

FIG. 3 is a block diagram showing the system configuration of the mouth image encoding method according to the embodiment of the present invention.

In FIG. 3, 12 is a code switching section, and 13 is a code amount calculation section.

The image encoding method of this embodiment H is as shown in FIG.
The encoded output of the variable length encoding circuit 9 of the embodiment I and the code length information of the flag information of the flag generating section 6 are stored in the code amount calculating section 1.
3, the encoded data amount is calculated and the code switching unit 1
Send to 2. The code switching unit 12 compares the encoded data amount with a predetermined value, and if the conditions are met, switches to an encoding method using intra-frame correlation. In the case of the present embodiment (2), the intra-frame correlation is used to make the amount of code between encoded block lines constant.

This can be done by calculating the number of encoded data in the code amount calculating section 13, and using an encoding means that utilizes intra-frame correlation when the number of encoded data becomes constant in the code switching section 12. The operations of the other processing units in FIG. 3 are similar to those corresponding to the same reference numerals in the embodiment I shown in FIG.

As can be seen from the above description, according to the image encoding method of the present embodiment H, data of block lines encoded using intra-frame correlation can be arranged at regular intervals during storage. For example, if the interval is set to one track on the recording medium, one track at a time will be skipped, only the encoded block line data will be decoded using intra-frame correlation, and the data will be connected. All you have to do is play it back, and the access time can be shortened.

In this case, both forward playback and high-speed reverse playback are easily possible, and if the same track is played back multiple times, slow playback is also possible.

In this embodiment (2), the code switching unit is configured so that the amount of code between block lines encoded using intra-frame correlation is constant, but if the amount of recorded data for one track is different, If the code is switched according to the amount of data, the same effect as in the first embodiment (2) can be expected.6. For block division, one frame may be divided into a plurality of blocks, or one frame may be regarded as one block line.

The present invention has been specifically explained above based on examples, but
It goes without saying that the present invention is not limited to the embodiments described above, and can be modified in various ways without departing from the spirit thereof.

〔Effect of the invention〕

As described above, according to the present invention, if block lines encoded using only intra-frame correlation are generated at constant block line intervals or constant code amount intervals,
Not only can a high compression ratio be achieved, but also random access, high-speed forward playback, and high-speed reverse playback processing necessary when video images are encoded and stored on a compact disc (CD) or the like can be performed.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the system configuration of an image encoding method according to Embodiment I of the present invention, and FIG. 2 is a block diagram showing the relationship between frames and block lines in the image encoding method according to Embodiment I of the present invention. FIG. 3 is a block diagram showing a system configuration of an image encoding method according to Example H of the present invention. In the figure, 1...input terminal, 2...block division section, 3
. . . DCT conversion section, 4 . . quantizer, 5 . . . code switching section, 6 . . . flag generation section, frame memory. 8... Motion compensation processing unit, 9... Variable length encoding circuit,
10...Output terminal, 11...Output terminal, 12...Sign switching section. 13... Code amount calculation unit.

Claims (2)

[Claims] (1) In a high-efficiency image encoding method that switches between a means of encoding using correlation within frames of a moving image and a means of encoding using a means of encoding using correlation between frames of a moving image, an image signal is made up of multiple lines. A block dividing means for dividing into block lines, determining whether the divided blocks are encoded using intra-frame correlation or inter-frame correlation, and determining the determined result. a flag generation means for generating flag information corresponding to the flag information; and a decision to use the flag information to encode the block line at a predetermined block line interval with an encoding means using intra-frame correlation. However, there is a coding switching means that determines that other block lines are to be coded by a coding means that utilizes inter-frame correlation, and a coding switching means that determines whether the intra-frame correlation or inter-frame correlation is to be coded based on the output of the coding switching means. An image encoding method characterized by comprising a means for encoding using the image. (2) Calculate the code amount for each block line, and sum the code amount from the block line coded using intra-frame correlation to the next block line coded using intra-frame correlation. 2. The image encoding method according to claim 1, further comprising encoding switching means for determining the encoding means so that the value satisfies a predetermined value.