KR100242653B1 - Intra-frame coding method and apparatus thereof - Google Patents

Abstract

The intra-frame encoding method and apparatus is to improve encoding efficiency by performing intra coding and motion prediction encoding on each macroblock of a current picture. The present invention compares the variance value of the pixel difference value between the reference macroblock and the current macroblock by the motion prediction of the current macroblock and the variance value of the pixel value in the current macroblock in the already encoded search region of the current encoding image. It is configured to selectively perform intra coding and motion prediction coding based on the comparison result. Accordingly, the present invention reduces the spatial redundancy by performing motion prediction encoding on a background region having little change in the image when performing intra-frame intra encoding, thereby improving the image quality by increasing the coding efficiency.

Description

In-frame encoding method and apparatus

The present invention relates to an intra-frame encoding method and apparatus for increasing encoding efficiency by selectively performing intra coding and motion prediction encoding on each macroblock in an intra-frame encoding.

Recently, international standardization of video compression technology, such as MPEG (IV) -IV, which encodes video at very low bit rate for video telephony or video conferencing system, has been actively conducted. have. MPEG-IV uses high-speed transmission channels of 64 kbps or less, so high compression is essential. To this end, motion compensation is performed to reduce redundancy in the time axis direction to take the difference between images, and then discrete cosine transform (DCT) and variable length coding are used to reduce redundancy in the space axis direction. That is, in general, continuous moving pictures are almost similar to the before and after picture and the current picture. Therefore, the difference between the current image to be encoded and the P image (forward prediction image) and the front image in time, or the B image (bidirectional prediction image) and the difference between the interpolation image made in front, rear or front and rear in time In this method, the amount of information to be transmitted is reduced by reducing the redundancy in the time axis direction by encoding and transmitting the difference value. However, not only take the difference of the same position but also use the motion compensation. This is a method of further reducing the amount of information to be transmitted by searching for the minimum difference between the previous image and the difference between the block positions in units of 16x16 pixels and taking the difference therebetween. In practice, in a P picture, a small amount of data is selected and encoded in units of 16 × 16 blocks (macroblocks) from the difference between the predicted picture after motion compensation and the difference.

However, in the case of an I-picture (intra-frame encoding) that intra-codes all of the pictures, intra-frame macroblocks are all intra-coded, thereby increasing the amount of bits of encoded data. As a result, not only the video encoding efficiency of the low data rate is lowered, but also the image quality is deteriorated.

Accordingly, an object of the present invention is to provide a method for improving encoding efficiency by allowing an intra mode and a motion prediction mode to be appropriately selected and used for each macroblock in intra-frame encoding in order to solve the above problems.

Another object of the present invention is to provide an apparatus implementing the above-described intra-frame encoding method.

1 is a block diagram showing an intra-frame encoding apparatus according to a preferred embodiment of the present invention.

FIG. 2 is a diagram for explaining a process of motion prediction in intra-frame encoding of the FIG. 1 apparatus. FIG.

* Explanation of symbols for main parts of the drawings

11: motion predictor 12: encoding selector

13: encoder

The intraframe encoding method of the present invention for achieving the above objects comprises the steps of: (1) predicting a motion of image data of a current macroblock in a search region formed to a maximum size in an already encoded region of the current image; (2) Among the intra mode and the motion prediction mode based on the comparison of the variance value of the difference value between the current macroblock obtained by motion prediction and the pixel value of the corresponding reference macroblock in the search region with the variance value of the pixel value of the current macroblock. Selecting one encoding mode; and (3) performing intra encoding on image data of the current macroblock when selecting an intra mode, and performing motion prediction encoding on image data of the current macroblock when selecting a motion prediction mode. Steps.

An intraframe encoding apparatus for achieving another object of the present invention performs motion prediction on image data of a current macroblock by using an encoded region of an image to be currently encoded to have the smallest pixel difference value from the current macroblock. The encoding mode is selected by comparing the motion predictor for obtaining the reference macroblock with the variance value of the pixel value of the current macroblock and the variance value of the pixel difference value between the reference macroblock and the current macroblock obtained by the motion predictor. And an encoder for encoding the encoding mode selected by the encoding selector in macroblock units for the current image to be encoded.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 shows an intra-frame encoding apparatus according to a preferred embodiment of the present invention. The apparatus of FIG. 1 is a motion predictor 11 which receives a video signal of a current picture in units of macroblocks and performs motion prediction using an already encoded region of the current picture, and a variance of pixel values of the current macroblock. And a coding selector 12 which receives a dispersion value of pixel difference values between the current macroblock and the reference macroblock obtained by the motion prediction and selects one encoding mode between the intra mode and the motion prediction mode, and the input current. An encoder 13 for encoding a macroblock in a selected encoding mode is provided. The operation of the first FIG. Device having such a configuration will be described with reference to FIG.

In FIG. 1, when a video signal of a current image to be encoded is input, the motion predictor 11 predicts the input current image in units of macroblocks having a size of 16 × 16 pixels. The motion predictor 11 finds reference macroblocks having the same image information as the current macroblocks in the current image in the maximum search areas MaxSx and MaxSy among the encoded areas in the current image. In FIG. 2, the current macroblock is indicated by a hatched rectangle, and the reference macroblock having the same image information as the current macroblock is also indicated by a hatched rectangle. The motion predictor 11 has the same number of pixels as the current macro block in the search area, and is obtained by moving the corresponding block and the corresponding block located at the upper left in the search area by one pixel to the right and / or downward. Each of the moved blocks is compared with the current macroblock. Methods used for comparison include the well-known Mean Absolute Error (MAE) method or Mean Square Error (MSE) method. When the reference macroblock having the smallest difference value is determined by comparing pixel values by MAE or MSE, the motion predictor 11 moves a motion vector indicating a spatial position difference between the reference macroblock and the current macroblock. Occurs. The motion predictor 11 also outputs the difference value and the pixel value of the current macroblock to the encoding selector 12. The encoding selector 12 compares the variance value of the pixel value of the current macroblock applied from the motion predictor 11 with the variance value of the pixel difference value between the reference macroblock and the current macroblock obtained by the motion predictor. Select the encoding mode for. If the variance value of the pixel value of the current macroblock is larger than the variance value of the difference value obtained by the motion prediction, the encoding selector 12 moves to the area where the current macroblock is almost unchanged, such as a uniform background area in the current image. Recognize and select the motion prediction mode. The encoding selector 12 is a region in which the current macroblock has image information different from the region already encoded in the current image if the dispersion value of the pixel value of the current macroblock is less than or equal to the difference value and the dispersion value obtained by the motion prediction. In other words, it selects the intra mode by recognizing it as an area out of the motion prediction range due to the large motion size. The encoder 13 encodes the current image input in macroblock units based on the encoding mode selected by the encoding selector 12. The encoder 13 performs intra coding on the current macroblock when the intra mode is selected for the current macroblock in the current image, and outputs the intra coded by the motion predictor 11 for the current macroblock when the motion prediction mode is selected. The obtained difference value is encoded and output, and the generated motion vector is output.

As described above, the intra-frame encoding method and apparatus of the present invention can adaptively perform intra coding and motion prediction encoding on each macroblock during intra encoding, so that the bits of the encoded data are higher than when performing only intra encoding. Since the amount can be reduced, the coding efficiency can be increased.

Claims (8)

An intraframe coding method comprising the steps of: (1) predicting a motion of image data of a current macroblock in a search area formed to a maximum size in an already encoded area of a current picture; (2) One of the intra mode and the motion prediction mode based on the comparison of the variance value of the difference value between the pixel value of the current macroblock and the corresponding reference macroblock in the search region obtained by motion prediction with the variance value of the retrieved value of the current macroblock. Selecting an encoding mode of the; And (3) performing intra encoding on the image data of the current macroblock when selecting the intra mode, and performing motion prediction encoding on the image data of the current macroblock when selecting the motion prediction mode. 2. The method of claim 1, wherein the step (1) estimates a reference macroblock having the smallest difference from the current macroblock by comparing pixel values between the current macroblock and the search region. . The motion prediction mode according to claim 1, wherein the step (2) is performed if the dispersion value of the pixel value of the current macroblock is greater than the difference value of the dispersion value between the pixel value of the current macroblock and the reference macroblock obtained by motion prediction as a result of the comparison. And selects an intra mode if the variance value of the pixel value of the current macroblock is less than or equal to the difference value obtained by the motion prediction and the variance value. An intra-frame encoding apparatus, comprising: a motion predictor configured to perform motion prediction on image data of a current macroblock by using an encoded region of an image to be currently encoded to obtain a reference macroblock having the smallest pixel difference value from the current macroblock; An encoding selector for selecting an encoding mode by comparing a dispersion value of pixel values of a current macroblock with a dispersion value of pixel difference values of a current macroblock and a reference macroblock obtained by the motion predictor; And an encoder for encoding the current image to be encoded in an encoding mode selected by the encoding selector in units of macroblocks. The reference macro block according to claim 4, wherein the motion predictor has a smallest difference from the current macroblock by comparing pixel values between the current macroblock and a search area formed to a maximum size in an already encoded region of the current image. And an intra frame encoding apparatus. 6. The apparatus of claim 5, wherein the motion predictor compares pixel values between a current macroblock and a search region using one of a mean absolute error (MAE) method and a mean square error (MSE) method. . The method of claim 4, wherein the encoding selector selects a motion prediction mode when the variance value of the pixel value of the current macroblock is greater than the variance value of the pixel difference value between the reference macroblock and the current macroblock obtained by the motion predictor. And an intra mode is selected if the variance value of the pixel value of the current macroblock is less than or equal to the variance value of the pixel difference value. 10. The method of claim 7, wherein the encoder performs motion prediction encoding on the image data of the current macroblock when the motion prediction mode is selected by the encoding selector. Intra encoding is performed on the image data of the current macroblock when the intra mode is selected. Intra-frame encoding apparatus characterized in that for performing.

Images