• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2008, Vol. 30 ›› Issue (8): 44-45.

• 论文 • 上一篇    下一篇

基于SVM的图像纹理特征分类研究

汤井田[1] 胡丹[1] 龚智敏[2]   

  • 出版日期:2008-08-01 发布日期:2010-05-19

  • Online:2008-08-01 Published:2010-05-19

摘要:

支持向量机(SVM)是一种表现卓越的分类方法,而灰度共生矩阵(GLCM)则是一种很好的纹理分析方法,故而本文提出了一种使用灰度共生矩阵进行特征提取的应用支持向量机的纹理特征分类法。实验结果表明,与直接应用灰度信息进行分类的支持向量机算法相比,本文方法可以取得更为准确的分类结果。

关键词: 支持向量机 灰度共生矩阵 特征提取 纹理分类

Abstract:

Support vector machine (SVM) has excellent performance in classification. And the Gray Level Co-occurrence Matrix (GLCM) is a promising method for texture analysis. So an algorithm of texture classification by SVM is proposed, which uses GLCM to extract features. Compared with the method using images' gray information directly for SVM classification,the method proposed in this paper can classify the texture features more exactly.

Key words: support vector machine,gray level co-occurrence matrix, feature extraction, texture classification