{"title":"一种改进的上下文自适应硬决策量化算法","authors":"Xi Wei, Zhelei Xia","doi":"10.1109/FSKD.2017.8393245","DOIUrl":null,"url":null,"abstract":"The traditional Hard-Decision quantization is adopted in fixed offset, without considering the correlation between quantization coefficients, so that the quantitative performance is poor. To solve this problem, an improved Context Adaptive Hard-Decision quantization was proposed which introduces the coefficient correlation. Statistics out the quantitative offsets that corresponding non-zero coefficient segment when each coefficient is quantified and actual bit rate of each nonzero coefficient in quantization. Using Bayesian two value discrimination method calculates the best threshold value which can distinguish quantitative results and build up new threshold model, then use the new threshold model to adjust quantitative offsets dynamically. The experimental results show that the improved Context Adaptive Hard-Decision quantization model which takes the rate of nonzero coefficient segment into account is more efficient comparing with traditional hard-decision quantization.","PeriodicalId":236093,"journal":{"name":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"An improved context adaptive hard-decision quantization algorithm\",\"authors\":\"Xi Wei, Zhelei Xia\",\"doi\":\"10.1109/FSKD.2017.8393245\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The traditional Hard-Decision quantization is adopted in fixed offset, without considering the correlation between quantization coefficients, so that the quantitative performance is poor. To solve this problem, an improved Context Adaptive Hard-Decision quantization was proposed which introduces the coefficient correlation. Statistics out the quantitative offsets that corresponding non-zero coefficient segment when each coefficient is quantified and actual bit rate of each nonzero coefficient in quantization. Using Bayesian two value discrimination method calculates the best threshold value which can distinguish quantitative results and build up new threshold model, then use the new threshold model to adjust quantitative offsets dynamically. The experimental results show that the improved Context Adaptive Hard-Decision quantization model which takes the rate of nonzero coefficient segment into account is more efficient comparing with traditional hard-decision quantization.\",\"PeriodicalId\":236093,\"journal\":{\"name\":\"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)\",\"volume\":\"44 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2017-07-29\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/FSKD.2017.8393245\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/FSKD.2017.8393245","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
An improved context adaptive hard-decision quantization algorithm
The traditional Hard-Decision quantization is adopted in fixed offset, without considering the correlation between quantization coefficients, so that the quantitative performance is poor. To solve this problem, an improved Context Adaptive Hard-Decision quantization was proposed which introduces the coefficient correlation. Statistics out the quantitative offsets that corresponding non-zero coefficient segment when each coefficient is quantified and actual bit rate of each nonzero coefficient in quantization. Using Bayesian two value discrimination method calculates the best threshold value which can distinguish quantitative results and build up new threshold model, then use the new threshold model to adjust quantitative offsets dynamically. The experimental results show that the improved Context Adaptive Hard-Decision quantization model which takes the rate of nonzero coefficient segment into account is more efficient comparing with traditional hard-decision quantization.