基于系数关系共生矩阵和SVM的纹理分析  

Texture Analysis Based on Coefficient Relationship Co-occurrence Matrix and SVM

作　　者：刘清[1,2] 林土胜[1] 王晓军[2] 

机构地区：[1]华南理工大学电子与信息工程学院,广州510640 [2]井冈山大学工学院,江西吉安343009

出　　处：《光电工程》2009年第4期128-132,共5页Opto-Electronic Engineering

基　　金：国家自然科学基金资助项目(60472006);广东省自然科学基金团队研究项目(04205783);江西省教育厅2008年度科技计划资助项目(GJJ08414)

摘　　要：纹理是图像中非常重要的特征。提出了一种新的纹理特征提取算法,即对纹理图像进行离散小波框架变换后,利用同一变换尺度下的小波高频系数与低频系数之间的依存关系信息,构造系数共生矩阵,在此基础上进行纹理特征提取,而不是独立地提取各子带系数特征。考虑支撑向量机(SVM)在小样本数据库和泛化能力方面的优势,在分类实验中采用支撑向量机分类器,实验结果表明,基于这种共生矩阵特征提取分类算法能得到很好的分类结果。Texture is an important image feature. A novel texture feature extraction technique is proposed based on coefficient co-occurrence matrix of discrete wavelet frame transformed image, which captures the information about relationship between each high frequency subband and low frequency subband of the decomposed image at the corresponding level. It is not independent to extract the information of each subband coefficient. Considering that the Support Vector Machine （SVM） has advantages of resolving the small-sample statistics and generalizing ability, the classification performance is analyzed by using the SVM classifier. The experimental results demonstrate the effectiveness of our proposed texture feature in achieving the improved classification performance.

关 键 词：离散小波框架变换 系数共生矩阵 纹理特征 支撑向量机 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]