KR0181067B1 - Moving picture encoder of having compatibility - Google Patents

Abstract

The present invention relates to an encoder for encoding image data by accommodating a compressor method that is not identical to the transmitting encoder, even if the receiving decoder is employed. To solve this problem, the present frame n and the previous frame n Implement the predicted frame (a) based on -1), implement the predicted frame (b) based on the current frame (n) and the preceding frame (n + 1), from which each frame (a) a motion predictor for selectively providing (b), a first frame memory for storing and reconstructing the predicted frame (a), a predicted frame (a) and a predicted frame (between the current and previous frames reconstructed therefrom) b) a first motion predictor for outputting the reduced frame according to b) and a first VLC for variable length coding the motion displacement information predicted by the first motion predictor to determine the difference between the receiver decoder and the transmitter code period. It has the effect that efficient coding possible, even if the techniques used apcheuk.

Description

Compatible video encoder

1 is a conventional image processing apparatus.

2 is a video encoding apparatus having compatibility according to the present invention.

* Explanation of symbols for main parts of the drawings

207: motion prediction unit 208: dance compensation unit

301: First frame memory 302: First motion predictor

303: First VLC 304: First buffer

The present invention relates to a video encoder having compatibility that accepts and encodes image data even when a compressor method that is not identical to the transmitting encoder is employed in the receiving decoder.

When a video signal is represented in a digital form, a large amount of information data must be transmitted in the case of a high quality video device. However, since the available frequency band of the transmission channel is limited, it is impossible to transmit a large amount of digital data. Therefore, compression of video data is essential in the case of transmitting a large amount of video data. Therefore, one of the methods for compressing a large amount of data is a probabilistic coding method, and the other is known as the most efficient hybrid coding method combining a temporal and spatial compression method.

The hybrid coding method uses motion compensated DPCM (differential pulse code modulation), two-dimensional DCT (discrete cosine transform), quantization of DCT coefficients, variable length coding (VLC), and the like.

The motion compensation DPCM predicts the motion of the object between the current frame and the previous frame of the input video signal, and predicts the current frame from the previous frame based on the motion information predicted therefrom. How to pay

The two-dimensional DCT removes spatial redundancy between object image data of an image signal. The two-dimensional DCT can effectively reduce the amount of data to be transmitted using quantization, VLC (variable length coder), and the like. That is, the motion predicted by the motion compensation DPCM can be represented as a two-dimensional motion displacement, that is, a vector. Here, one of various approaches for estimating the displacement of an object is a block unit method, and another is a pixel unit motion estimation.

In block-based motion estimation, a block matching block is determined by comparing a block of a current frame with blocks of a previous frame. From this, the interframe displacement for the entire block is estimated for the current frame being transmitted. On the other hand, in the pixel unit method, the motion displacement is calculated using all of the pixels for implementing the image, which causes a disadvantage in that a lot of data is generated.

FIG. 1 is a block diagram showing an apparatus for compressing and encoding a general video signal.

Referring to FIG. 1, a digital image signal input in the form of a bitstream is implemented in units of frames, and the frame forming unit 100 sequentially outputs the frames, and is implemented by the frame forming unit 100. The subtractor 101 for obtaining the difference signal between the current frame and the predicted frame to be described later, the DCT 102 for obtaining the DCT conversion coefficient using the cosine function, and the difference signal between the frames by the listener 101 from the DCT 102. The step size of the AC coefficients of the high frequency component is determined among the DCT transform coefficients, and the quantizer 103 and the transform coefficients quantized by the quantizer 103 are quantized according to the step size. An inverse quantizer 104 for restoring the original transform coefficients, an inverse DCT 105 for inverse discrete cosine transforming the transform coefficients inversely generated by the inverse quantizer 104, and a frame for storing the previous frame The frame is based on the motion displacement predicted by the memory 106, the motion predictor 107, and the motion predictor 107 that predict the motion between the current frame to be input and the previous frame stored in the frame memory 106. The motion compensator 108 compensates from the previous frame stored in the memory 106, the adder 109 for adding the compensated motion displacement and the inverse DCT data, and the quantized transform coefficients in the quantizer 103 are varied. It consists of a VLC (110) for variable-length encoding the motion displacement predicted by the length prediction or the motion predictor (107) and a buffer 111 for transmitting the variable-length coded transmission data by the VLC (110) to the transmission channel. And the intra / inter operation mode is set based on the on / off operation of the switches 112-113. In the conventional video encoding apparatus configured as described above, the video data transmitted through the limited transport channel is encoded by applying the compressor method already set and transmitted to the receiving decoder. At this time, the encoded decoder can restore the encoded video data only when the receiving decoder receives the same compression scheme as the transmitting encoder. Accordingly, the present invention has been made to solve the above disadvantages, and an object of the present invention is to enable the receiving decoder to perform this even when a compressor method that is not identical to the transmitting encoder is employed in the receiving decoder. It is to provide a video encoder with compatibility.

According to the present invention for achieving the above object, detecting a predicted frame (movement displacement information) based on the current frame and the previous frame, and encoding on the basis of the difference signal between the predicted frame and the input current frame In the video encoding device,

A frame (a) predicted based on the current frame (n) and the previous frame (n-1) is implemented, and a frame predicted based on the current frame (n) and a frame (n + 1) preceding this frame. a motion predictor 207 for implementing (b) to selectively provide each implemented frame (a) (b); A first frame memory (301) for storing, restoring and outputting the predicted frame (a) provided from the motion predictor (207); A first frame outputting the reduced frame based on the predicted frame a between the current and previous frames reconstructed from the first frame memory 301 and the predicted frame b provided from the motion predictor 207. A motion predictor 302; And a first VLC 303 for variable length encoding the motion displacement information predicted by the first motion predictor 302.

Hereinafter, the present invention will be described in detail with reference to the illustrated drawings. 2 is a block diagram showing a video encoding apparatus having compatibility according to the present invention.

Referring to FIG. 2, the video encoding apparatus compatible with the present invention includes a frame forming unit 200, a motion compensating unit, an encoding unit, a decoding unit, a VLC 210, a buffer 211, a switch 212, 213, 1 frame memory 301, a first motion predictor 302, a first VLC 303, and a first buffer 304.

The frame forming unit 200 is configured to implement the input image signal in a frame unit and provide the same to the subtractor 201 and the motion predicting unit 207 which will be described later.

The motion compensating means includes a subtractor 201, an adder 209, a frame memory 206, a motion predictor 207, and a motion compensator 208, and inputs a previous frame and an input stored in the frame memory 206. The motion estimation unit 207 obtains the motion of the current frame feeling, and provides the motion compensation unit 208 with this movement displacement, and adds a prediction frame (motion displacement information) that is compensated for from the previous frame. 209 and subtractor 201.

In addition, the motion predictor 207 predicts a motion displacement between the previous frame and the input current frame, and provides the predicted frame to the first frame memory 301 which will be described later.

A motion displacement is predicted between the current frame and a frame one frame ahead of the current frame, and is provided to the first motion predictor 302 which describes the predicted frame.

The encoding means includes a DCT 202 quantization unit 203, and encodes a differential signal between an input current frame and a predicted frame by DCT and quantization to dequantize the quantizer 204, the VLC 210, and the first VLC. 303 is configured to be provided.

The decoding means is composed of an inverse quantization unit 204 and an inverse DCT 205, and decodes the image data quantized by the quantization unit 203 by inverse quantization and inverse DCT. The compensated displacements are summed and stored in the frame memory 209 as the previous frame.

The first frame memory 301 temporarily stores the predicted frame a between the current frame n and the previous frame n-1 by the motion predictor 207, and restores the stored predicted frame by reconstructing the VLC ( 210 and the first motion predictor 302.

The first motion predictor 302 includes a predicted frame a reconstructed from the first frame memory 301, a current frame n, and a frame n + 1 preceding the frame by the motion predictor 302. The motion displacement information is detected and provided to the VLC 210 and the first VLC 303 based on the predicted frame b of the liver.

The VLC 210 is configured to provide variable length encoding of the video data encoded by the encoding means to the buffer 224 described later.

The first VLC 303 is configured to vary length-encode the motion displacement information detected by the first motion predictor 302 to provide the first buffer 304 to be described later.

The buffer 211 provides a control signal QP according to the storage capacity of the variable length coded image data provided from the VLC 210 to the quantizer 208 and outputs the variable length coded image data to a decoder. It is composed.

The first buffer 304 provides a control signal according to the storage capacity of the variable length coded data provided from the first VLC 303 to the quantizer 208, and outputs the variable length coded image data to a decoder. do.

Referring to the present invention configured as described in detail as follows.

First, an image signal input in the form of a bit stream is input to the frame forming unit 200 in a digital form and then provided to the subtractor 201 and the motion predictor 207 in units of frames.

Then, the motion predictor 207 predicts the motion between the previous frame and the currently input frame stored in the frame memory 206 and then provides the image data to the motion compensator 208.

The motion compensator 208 obtains a predicted frame based on a previous frame stored in the frame memory 206, and from there, the difference signal between the predicted frame and the current frame provided from the frame forming unit 200. Encoding is then performed to the VLC 210, the first VLC 303, and the inverse quantization unit 204 of the decoding means.

The decoding means decodes the encoded image data by inverse quantization and inverse DCT, and combines the decoded image data with the motion-compensated displacement and stores the previous frame in the frame memory 206 for use in the next encoding. .

The motion predictor 207 provides the motion displacement between the current frame and the previous frame to the first frame memory 301.

In the first frame memory 301, the motion displacement predicted by the motion predictor 207 is provided to the VLC 210 and the first motion predictor 302, and the motion displacement information is varied by the VLC 210. The length is encoded and then output to the decoder through the buffer 211.

Meanwhile, the first motion predictor 302 receives the frame a predicted between the current frame and the previous frame from the first frame memory 301, and also predicts the frame a and the current frame n. And the predicted frame b between the frame n + 1 preceding this frame is provided from the motion predictor 207.

Here, the first motion predictor 302 predicts the motion between the predicted frame a and the predicted frame b between the current frame n and the frame n + 1 preceding the frame, and predicts from this. The provided motion displacement information is provided to the VLC 210 and the first VLC 303. That is, the motion displacement information by the first motion predictor 302 may be used as additional information to correct an error that may occur in the variable length encoding process.

Accordingly, in the VLC 210, variable length coding is performed on the encoded image data output from the quantization unit 203, the first frame memory 301, and the first VLC 303 of the encoding unit, and is output to the decoder through the buffer 224. Done.

In addition, the first VLC 303 also variable-length encodes the encoded image data output from the quantization unit 203 and the first motion predictor 207 of the encoding unit, and outputs the variable length encoding to the decoder through the first buffer 304. .

As described above, the present invention has the effect that efficient encoding is possible even when other compression techniques of the receiving decoder and the transmitting code period are used.

Claims (1)

A video encoding apparatus for detecting a predicted frame based on a current frame and a previous frame, and encoding the encoded frame based on a difference signal between the predicted frame and an input current frame; A frame (a) predicted based on the current frame (n) and the previous frame (n-1) is implemented, and a frame predicted based on the current frame (n) and a frame (n + 1) preceding this frame. a motion predictor 207 for implementing (b) to selectively provide each implemented frame (a) (b); A first frame memory (301) for storing, restoring and outputting the predicted frame (a) provided from the motion predictor (207); A first frame outputting the reduced frame based on the predicted frame a between the current and previous frames reconstructed from the first frame memory 301 and the predicted frame b provided from the motion predictor 207. A motion predictor 302; And a first VLC (303) for variable length encoding the motion displacement information predicted by the first motion predictor (302).