首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于邻近值的HEVC帧内预测优化
引用本文:李 曾,何小海,孙伟恒,熊淑华,Karn Pradeep.基于邻近值的HEVC帧内预测优化[J].国际商务研究,2019(10).
作者姓名:李 曾  何小海  孙伟恒  熊淑华  Karn Pradeep
作者单位:四川大学 电子信息学院,成都 610065,四川大学 电子信息学院,成都 610065,四川大学 电子信息学院,成都 610065,四川大学 电子信息学院,成都 610065,四川大学 电子信息学院,成都 610065
基金项目:国家自然科学基金资助项目(61871279);成都市产业集群协同创新项目(2016-XT00-00015-GX)
摘    要:高性能视频编码(High Efficiency Video Coding,HEVC)标准中的帧内编码模式利用当前帧中像素点之间的空间相关性作出有效预测。为了解决待编码像素远离参考像素时预测不准确的问题,提出了一种基于邻近值的HEVC帧内预测优化算法。该算法的主要思想是,对于当前像素,先根据传统HEVC帧内编码方法得到其预测值,再使用该像素点左边、左上、上边位置的修正值以及该像素本身的预测值对该预测值进行修正。因为将当前像素与周围像素的相关性进行了有效的数学建模,所以HEVC帧内编码性能得到了提升。实验结果显示,所提算法与HEVC标准相比,最高节省了2.7%的码率,平均节省的码率为1.3%。

关 键 词:高性能视频编码(HEVC)  帧内编码  邻近值  马尔科夫模型

Optimization for HEVC intra prediction based on neighboring pixels
LI Zeng,HE Xiaohai,SUN Weiheng,XIONG Shuhua and Karn Pradeep.Optimization for HEVC intra prediction based on neighboring pixels[J].International Business Research,2019(10).
Authors:LI Zeng  HE Xiaohai  SUN Weiheng  XIONG Shuhua and Karn Pradeep
Institution:College of Electronics and Information Engineering,Sichuan University,Chengdu 610065,China,College of Electronics and Information Engineering,Sichuan University,Chengdu 610065,China,College of Electronics and Information Engineering,Sichuan University,Chengdu 610065,China,College of Electronics and Information Engineering,Sichuan University,Chengdu 610065,China and College of Electronics and Information Engineering,Sichuan University,Chengdu 610065,China
Abstract:The high efficiency video coding(HEVC) makes effective prediction by using the spatial correlation between pixels in the current frame.An optimization algorithm for HEVC intra prediction based on neighboring pixels is proposed to solve the problem of inaccurate prediction when encoding pixels are far away from the reference pixels.The main idea about this algorithm is that the predicted value of current pixel is firstly predicted according to the traditional HEVC intra prediction.Then the predicted value is modified by using the modified values of its left,upper left and upper positions as well as the predicted value of current position.Since the correlation between current pixel and its surrounding pixels is effectively modeled,the algorithm achieves efficient performance for HEVC.Experimental results show that,compared with the standard HEVC,the proposed algorithm achieves 1.3% bit saving on average and the highest up to 2.7%.
Keywords:high efficiency video coding(HEVC)  intra coding  neighboring pixel  Markov model
点击此处可从《国际商务研究》浏览原始摘要信息
点击此处可从《国际商务研究》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号