Research on Image Retrieval Based on Color and Shape Features

摘要

According to the robustness of color and shape feature extraction, a multi-feature matching algorithm which is combined of color features and shape features is proposed. In the respect of color feature, a new color histogram method based on main colors is proposed. By combining the major color retrieving method and the color histogram computing, two rapid elective filter are carried out, scope of the search is reduced and the retrieval efficiency is improved. In the respect of shape feature, the use of Fourier shape descriptor, an improved contour-based description method is proposed. According to the tangential angle of contours(curvature) is highlighted and factors such as complex coordinates and center distance are ignored, within a reasonable range, the accuracy is lowered appropriately and the query speed is improved significantly. Experiments in traffic signs image library, show that the proposed method of recognition accuracy is better than traditional methods, and efficiency has improved.