Key Technology Research Of Colthing Image Retrieval System Based On The Content

With the rapid development of Internet, the information grows rapidly. How to find the helpful information in the vast database has been the hotspot in recent years. The information of the image has the characteristics of richness and visualization, and it is easily understood, which make the image retrieval important to the practical applications.In this paper, the image features are extracted from two ways, i.e., the color and the shape. First, the image is partitioned into different blocks according to the characteristics of the data of the image library. Then the color feature of the image is extracted according to the color histogram of the obtained blocks. To analyze the performance, we compare it with the global color histogram method. Secondly, in the part of shape feature, we extract contour by using the active contour model (Snake model).Then, describe the shape features with the Freeman chain code base on the contour. Finally, the color and the shape features are fused to accomplish the integrated retrieval of image, and the weight values for different features can be adjusted by the user.