熵编码的区域上下文分析和二进制化改进-region context analysis and binarization improvement based on entropy coding.docxVIP

下载本文档

6
0
约4.27万字
约 52页
2018-07-31 发布于上海
举报
版权申诉

熵编码的区域上下文分析和二进制化改进-region context analysis and binarization improvement based on entropy coding.docx

1、原创力文档（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。。
2、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
3、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。
4、该文档为VIP文档，如果想要下载，成为VIP会员后，下载免费。
5、成为VIP后，下载本文档将扣除1次下载权益。下载后，不支持退款、换文档。如有疑问请联系我们。
6、成为VIP后，您将拥有八大权益，权益包括：VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
7、VIP文档为合作方或网友上传，每下载1次，网站将根据用户上传文档的质量评分、类型等，对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档

熵编码的区域上下文分析和二进制化改进-region context analysis and binarization improvement based on entropy coding

上海交通大学硕士学位论文上海交通大学硕士学位论文万方数据万方数据 RESEARCH ON REGIONAL CONTEXT MODEL AND DYNAMIC BINARIZATION IN ENTROPY CODING ABSTRACT The application of super high-density video has been widely spread in the field of mobile communication and multi-media on network. To meet the demand, HEVC made the official public release which imported more direction contexts, truncated rice in the binarization procedure of CABAC and other algorithms based on H.264/AVC. To further improve the video compression efficiency by appropriate context model and dynamic adaptivity, both the regional context model and dynamic binarization in entropy coding, which are in the procedure of context modeling and encoding, are researched by this paper. There are a lot of context model algorithms which make the probability statistics by the context model, such as context tree weighting and prediction with partial matching. When applied to multi-direction context model, these context model algorithms make the model redundancy double-exponential increase. To solve the problem in multi-direction context model, this paper adopts the algorithm which centers the current pixel to be predicted, sorts the pixels in the region according to the line distance from the center pixel and calculates the normal coefficient by the multiple of the amount that is max line distance minus current line distance, to the power of the count of the pixels which have the same line distance. This algorithm takes account of the impact of prediction accuracy from both the line distance and the count of pixels with the same line distance. In the experiments of the compression for the test sequences of H.264/AVC, it turned out that the region context model improved the compression efficiency. III To reduce the count of multiple operation in arithmetic coding, CABAC made the state machine which stood for some probability. In the procedure of encoding, it will change the state from 0 to 63. The probability of next state is the multiple of the