Wavelet-Contourlet Texture Image Retrieval System

摘要：

To improve the retrieval rate of contourlet transform texture image retrieval systems and reduce the redundancy of contourlet which cost two much time in building feature vector database, a waveletcontourlet transform retrieval system was proposed. In the system, wavelet-contourlet was constructed by wavelet transform at coarser scales and contourlet transform at finer scales, the mixed transform was used to transform the space domain texture image into multiscale domain, the features which characterize the content of each image used were absolute mean energy, standard deviation and kurtosis. The feature vectors were constructed by cascading the absolute mean energy, standard deviation and kurtosis of each sub-band wavelet-contourlet coefficients and the similarity measure used was Canberra distance. Experimental results on 109 brodatz texture images show that using the features can lead to a higher retrieval rate than several contourlet transform retrieval systems which utilize the combination feature of standard deviation and absolute mean which is most commonly used today.

关键词： content based image retrieval wavelet-contourlet transform absolute mean energy kurtosis standard deviation

作者: Xiang-Ying Li Yan-Li Li Xin-Wu Chen Wei Luo

作者单位: School of Computer and Information Technology Xinyang Normal University Xinyang, China

会议类型: 国际会议

会议名称: 2010 International Conference on Future Information Technology(2010年未来信息技术国际会议 ICFIT 2010)

会议地点: 长沙

会议语种:英文

页码: 889-893

在线出版日期: 2010-12-14（万方平台首次上网日期，不代表论文的发表时间）