Research On Video Coding And Decoding Algorithm Based On Video Content Complexity

New generation of video coding standard HE VC(High efficiency video coding)'s proposal significantly improved the coding efficiency and draw the researchers attention.In the HEVC coding standard,the introduction of a new quadtree coding structure,in improving the coding efficiency at the same time to a certain extent,increased the complexity of coding,affecting the real-time encoder.In addition,the introduction of intra-block copy algorithm enhances the efficiency of HEVC screen content coding,but the tool uses a fixed-size matching block search method,which reduces the search ability of matching blocks.In order to solve the above problems,on the basis of analyzing the complexity of video content this paper studies achieve reducing the complexity of quadtree coding and improving the matching efficiency of intra-block block algorithm,improve the coding performance of HEVC encoders..The main innovations of the paper include the following two aspects.First,for the problem of the quadruple tree structure of HEVC framework coding complexity.In-depth study of the coding unit quadtree recursive partition process.The correlation between the coding unit size and the unit image complexity is found.A low complexity algorithm is proposed to use the image complexity to skip the rate-distortion optimization process or to determine the coding unit in advance.The algorithm significantly reduces the complexity of coding without substantially affecting the quality of video compression.Second,the study of the screen content of the intra block copy technology show the tool is based on the use of fixed-size matching block search method,for some complex continuous text area matching block search ability is weak,reduced coding efficiency.This paper designs a matching block search method based on adaptive size.The adaptive block division of the coding unit is carried out with the change of variance as an index.So that the size of the block is inversely proportional to the complexity of the block contents,which is conducive to improving the matching degree and coding efficiencyIn view of the above two algorithms,a lot of experiments were carried out to test,the first algorithm reduces the coding time by 35%on All Intra(AI)coding conditions.At the same time,the second algorithm increase 0.96%BD-rate,while increasing the encoding time only by an average of 3%on All intra(AI)encoding conditions.