Intra-frame prediction fast mode selecting method based on autoregressive model in video standard

Abstract

The invention discloses an intra-frame prediction fast mode selecting method based on an autoregressive model in the video standard to mainly solve the problem of high selection complexity in an intra-frame prediction mode in the H.265 / HEVC standard. The method comprises the step of carrying out the rough code selection on a current prediction unit to obtain m candidate modes, the step of adding a most possible mode to obtain m1 candidate modes, the step of carrying out ascending sort on cost function values of the m1 candidate modes, and the step of comparing the difference value of two adjacent cost values, the specific value of the average value of the two adjacent cost values and a threshold value obtained based on the autoregressive model to select n ultimate candidate modes in a self-adaptation mode to carry out rate-distortion optimization. The method is easy to operate, the compression performance of an image is kept, running time is shortened, a technical basis is provided for achievement of the H.265 / HEVC standard, and the method can be used in all intra-frame prediction mode selections in a video compression encoding terminal based on the H.265 / HEVC standard.

Description

technical field [0001] The invention belongs to the technical field of digital signal processing, relates to a prediction mode selection realization method in image video compression coding, and can be used in the intra-frame prediction coding process in the H.265 / HEVC video standard. Background technique [0002] In 2004, one year after the establishment of the H.264 / MPEG-4AVC standard, VCEG, a video expert group of ITU-T, the International Telecommunication Union's Telecommunication Standardization Organization, began to study technologies that could become the next-generation video compression standard. For this purpose, the Joint Collaborative Team on Video Coding (JCT-VC) jointly established by VCEG and MPEG, the motion expert group of the International Organization for Standardization ISO / IEC, jointly developed the HEVC standard. In recent years, JCT-VC has been working on the formulation of H.265 / HEVC, and released the first draft of the public version in July 2012. ...

AI Technical Summary

Problems solved by technology

This method realizes the fast selection of the intra prediction mode, and has a certain compression of the running time, but its threshold value is determined by the size of the CU, and cannot be adaptively adjusted according to the context. For images with rich texture content, the encoding quality is reduced.

[0012] To sum up, the first two technologies of the above three technologies use the texture direction characteristics of the image block to judge the best prediction mode, so when the texture direction characteristics of the image block are not obvious, the direction judgment error is caused and the image compression performance is reduced; The texture direction calculation process is more complicated, which increases the amount of additional calculations; while the threshold value in the third technique is determined by the size of the CU, and cannot be adaptively adjusted according to the content of the context. When the texture content is rich, the coding performance is affected.

Images