Wavelet-based Coefficients Relationship Co-occurrence Histogram Algorithm of Texture Teature Extraction

摘要：

We propose a novel texture feature extractiontechnique based on coefficientsco-occurrencehistogram of discrete wavelet frame transformed image,which capture the information about relationshipbetween each high frequency subband and the lowfrequency subband of the decomposed image at thecorresponding level It is not independently utilizingthe information of each subband coefficient.Theclassification performance is analyzed using the k-NNclassifier.And the experimental results demonstrate theeffectiveness of our proposed texture feature inachieving the improved classification performance.Comparisons with the Gabor filter and a recentlyproposed approach are also provided.

作者: Liu Qing Lin Tu-sheng

作者单位: School of Electronic and lnformation Engineering,South China Univ.of Tech.,Guangzhou 510640,Guangdon School of Electronic and lnformation Engineering,South China Univ.of Tech.,Guangzhou 510640,Guangdon

会议类型: 国际会议

会议名称: 2008 3rd International Conference on Intelligent System and Knowledge Engineering(第三届智能系统与知识工程国际会议)(ISKE 2008)

会议地点: 厦门

会议语种:英文

页码: 1050-1053

在线出版日期: 2008-11-17（万方平台首次上网日期，不代表论文的发表时间）