基于SVM综合利用颜色和纹理特征的图像分类和检索

基于内容的图像检索和分类在多媒体数据库管理中得到了越来越多的重视。在体统的基于内容的图像检索方法中，语义间隔（semantic gap）常常会导致检索的效果不佳，利用支持向量机（SVM）可以很好的解决图像中的语义间隔。本文介绍了我们设计的基于SVM分别利用颜色特征和纹理特征的两种分类方法，在此基础上，我们提出了一种综合利用上述两个特征共同进行分类的方法。实验结果表明。综合特征要比单一特征分类效果更好。

SVM-based Image Classification and Retrieval Using Both Color and Texture Features

Content-based image retrieval and classification have gotten more and more attention in multimedia database management. In the traditional approach of content-based image retrieval, the results of retrieving are not satisfying because of the wide semantic gap. We propose that the support vector machine can resolve the semantic gap well. In this paper, we first discuss the method for SVM-based image classification using color texture features separately, then, we propose a new method for SVM-based image classification using combined color and texture feature. The experimental results show that the new method is better.