novel method is presented for image feature repres entation. The underlying color information of images is extracted to obtain colo r feature values based on the presented method. With the edge detection of objects in images, e dge orientation values of pixels in images are calculated. Then color feature values and edge orientation values are quantized. Based on quantizatio n result analysis of neighboring pixels, an eight dimensional feature vector is constructed for each pixel. Different weights are given to different positions based on the relationship between a central pixel and neighboring pixels. According to feature vectors of pixels, characteristic values are calculated. Fi nally, the pixels number with the same characteristic value is counted to f orm a histogram, which is the basis for image retrieval. Experimental result s demonstrate that the proposed method can effectively describe color distributi on and spatial structure of objects in images. Moreover, more image details can be saved t o enhance the discrimination power. The method is proved to be much more effective than other methods.