Low-complexity in-frame estimation mode selection method

Abstract

The present invention discloses a low-complexity method of intraprediction mode selection, two groups of sample points representing texture characteristics are extracted from the current block, according to the cost values of adjacent blocks, a threshold is calculated. The method includes four steps; firstly, the method judges if the current block is located at the special position of an image frame, if yes, an optimal prediction mode at the special position is calculated and then exported, and calculation is finished; secondly, if not, the cost value of the current block under the most possible prediction mode is calculated, if the cost value is less than the threshold, then the mode is exported, and the calculation is finished; thirdly, otherwise, a mode with the least cost value is selected from all non most possible prediction modes and used as a temporary optimal prediction mode; fourthly, the cost value of the most possible prediction mode is compared with the cost value of the temporary optimal prediction mode, and the mode with the least cost value is used as an optimal prediction mode and exported. Because the present invention selects a characteristic point to replace the whole data block, on the premise that the video encoding quality is not obviously affected, the present invention notably reduces the calculation load of the intraprediction mode selection and increases the real-time performance of video encoding.

Description

technical field [0001] The invention belongs to the field of video information compression, and in particular relates to a low-complexity intra-frame prediction mode selection method. Background technique [0002] The current advanced video coding standards include the intra-frame prediction function, which uses the sample points of adjacent blocks to extrapolate to realize the prediction of the current block, so as to better eliminate the spatial redundancy in the single-frame image, so that only the The residual of the predicted block and the current block is encoded. Especially in the region where the change is flat, the bit rate can be greatly reduced by using intra prediction. [0003] When a macroblock is coded in intra-frame mode, a prediction block P is constructed using previously coded and reconstructed blocks. For the luma component, a prediction block P can be created for each block or macroblock. For example, in the H.264 encoding standard, there are 9 option...

AI Technical Summary

Problems solved by technology

This method needs to determine the magnitude and angle of the edge vectors of all pixels in the block, and calculate the edge direction histogram for each pixel, which requires a lot of calculation and is not practical.

Images