Research On Low Complexity Intra And Inter Compression Algorithm Based On HEVC

Today,the information era,with the development of computer and Internet technology,video applications are becoming more and more extensive and frequent.Wireless video transmission is a combination of radio transceiver and video information processing technology,suitable for many occasions that are difficult to rely on cable transmission.Today,various application areas and occasions put higher demand on wireless video transmission,and the increase in video resolution and frame rate directly leads to flare up in bit rate of compressed video.In order to achieve the timely transmission of video,while enhancing the ability of wireless transceiver at the same time,using a new generation of video compression standard,the HEVC,can save half of the bit rate.In order to apply HEVC in wireless video transmission system,it is necessary to solve the problem of its high computational complexity.The current low complexity intra and inter prediction algorithms based on HEVC generally aim at limited step,although the rate-distortion performance almost unchanged,the coding time saving is also limited.Taking into account the practical application of wireless video transmission,in the case of a slight sacrifice of rate-distortion performance,this paper achieves higher coding time saving rate.Based on the understanding of the intra and inter video compression principle as well as the study of the optimal rate-distortion algorithms,this paper makes low-complexity improvements on CU splitting procedure and mode selection procedure in the intra and inter prediction.In this paper,we first propose a CU fast partitioning algorithm based on image spatial correlation,which use the depth information of adjacent CUs to set the depth range for the current CU.This algorithm can be applied to both intra prediction and inter prediction.For the intra prediction,this paper firstly proposes a fast algorithm which can skip the RMD process and eliminate the angle mode according to the statistical characteristic between the reference pixel variance and the intra mode.In the RMD and RDO mode selection procedure,this paper realizes the quick optimi zation of intra mode using image texture feature and MPM statistical feature.Then,the early termination of CU partitioning is realized by using the RD cost of the final intra prediction mode.Finally,the low complexity intra prediction integrated algorithm is designed by combining the improvement points above.For the inter prediction,based on the understanding between rate-distortion cost and prediction residual as well as quantization parameter,it is considered in this paper that when the skip or the cbf condition is satisfied,any other following mode can be skipped without any constraints for higer coding time saving.In addition,considering that the asymmetric partition mode or AMP plays a minor role,this paper disables the AMP mode.Finally,combined with the inter prediction strategy above and the CU fast partition algorithm,the low complexity inter prediction integrated algorithm is designed.Experiments shows that,compared with the existing low complexity algorithm,the intra and inter integrated algorithm proposed in this paper can reduce more coding time with slight increase in the loss of rate-distortion performance.