Abstract: In this paper，a new method of content-based image retrieval is presented，which integrates color and texture feature.Color feature is extracted by the improved colors histogram；texture analysis is done using Gabor wavelet.In order to improve retrieval performance，a relevance feedback mechanism based on support vector machines is introduced.Experiments show that the proposed system is not only efficient but also effective.

摘要： 由于单一特征不足以准确地描述图像特征，提出了一种结合颜色特征和纹理特征的图像检索方法。针对传统颜色直方图中图像对所有像素具有相同重要性的问题进行了改进，提出像素加权的改进颜色直方图方法；然后采用Gabor小波提取纹理特征。为提高图像检索的性能，提出了一种支持向量机的相关反馈方法。实验结果表明，该方法具有较好的检索性能。