Implement Based On HEVC Algorithms

With the development of the Internet, especially the rapidly growing popularity of smart phones and applications, video information is already everywhere According to the survey, people watch the demand for network video and network length short video has reached nearly 80%. While the increase in the number of videos, based on the resolution of the video users are increasingly high requirements. Despite recent network bandwidth and storage capacity is increasing rapidly, but can not meet the requirements of mass storage and transmission of video data.Currently video coding technology from the first generation video coding standard H.261 / MPEG-1, H.263, MPEG-2, MPEG-4, H.264 / AVC, until today, the third generation of highly efficient video coding HEVC( High Efficiency Video Coding) standard appears, it can guarantee video quality under the premise of the compression rate doubled. But as applied to the actual network traffic in future video encoding technology, which has brought us a high compression ratio, while the coding complexity is also increased exponentially, so HEVC paper studies how to reduce the complexity of the algorithm and bit rate.This paper uses two methods to reduce HEVC algorithm complexity and bit rate. The first is a fast HEVC intra- prediction algorithm based on motion search. Prediction according to the intra prediction unit and the current reference cell in the reference frame similarity, the best reference frame as one of the current intra prediction mode candidate intra mode prediction unit, so as to skip some of the redundant the rate of intra prediction modes- calculated distortion cost, to reduce the computational complexity of the encoding results. HEVC proposed algorithm and the reference software HM7.0 the algorithm compares the experimental results can be seen from the rapid HEVC intra prediction based on the motion search is reduced by about 25% of the intra- encoding time.The second is using a time-domain correlation judgment HEVC fast inter-mode-based approach. Statistical analyzes the coding unit(CU) between the adjacent frames corresponding block and the prediction unit(PU) in the time domain correlation and inter-frame prediction HEVC highest probability of PU mode, based on the statistical analysis of the results presents a time-domain correlation based fast inter-mode decision method HEVC skip some redundant CU and PU separation layer prediction mode, CU only take up two PU layer prediction mode, thus greatly reducing the need for rates- the number of distortion cost calculations. Like the previous method, but also to compare the time-domain correlation HEVC fast inter mode decision method and HEVC reference software 7.0 algorithm, experimental results show that the proposed method under the loss in bit rate and video quality small case, 52.3-63.5 % of coding saves timeThe proposed method will largely reduce the computational complexity of encoding, more conducive to the future application of video transmission and storage.