组合颜色、空间和纹理特征的图像检索

Combining Color,Spatial and Texture Feature for Image Retrieval

提出了一种组合颜色、空间信息和纹理特征的图像检索方法 .用颜色块直方图来表达图像的颜色特征 ;通过提取主颜色块的面积比、质心、分布方差等特征得到图像的空间特征 ;利用 Gabor滤波器组提取图像的纹理特征 .实验结果表明 ,基于组合特征的图像检索方法要优于单纯的基于颜色。

In this paper,a new method combining color,spatial and texture feature for efficient image retrieval is presented.Color feature is extracted by the histogram of dominant colors.Spatial analysis is performed by computing the ratio of pixels,the barycenter and the standard deviation of the distribution of dominant colors.Texture analysis is done using a bank of Gabor filters.It is demonstrated that the synergy resulting from the combination of color,spatial and texture is superior than using just color and texture.