An Approach for Image Retrieval Based on Wavelet Coefficient Co-occurrence Matrix

摘要：

This paper integrated the advantages of co.occurrence matrix and wavelet transform and proposed a new method based on wavelet coefficient co-occurrence matrix.The new approach constructed four co-occurrence matrices based on wavelet coefficient matrix,and then it picked up the texture parameters of the four co-occurrence matrices for image retrieval,it not only considered the spatial correlation info,but also took full advantage of excellent character of wavelet transforill.The experimental results show that the new method has an advanced general performance and its computational complexity is lower than wavelet histogram technology.

关键词： co-occurrence matrix image retrieval texture character wavelet coefficient wavelet transform

作者: Beizhan Wang Qingqi Hong Cuihua Li Qingshan Jiang LiangSi

会议类型: 国际会议

会议名称: The International Conference Information Computing and Automation(2007国际信息计算与自动化会议)

会议地点: 成都

会议语种:英文

页码: 320-323

在线出版日期: 2007-12-19（万方平台首次上网日期，不代表论文的发表时间）