Image Retrieval Based on the Local Multi-Features Statistical Method

摘要

The spatial geometrical features of the objects are widely used in the image matching and retrieval. The popular algorithms of the spatial feature detection are designed for gray images, which failed to make use of the color information in color images. However, the color information is absolutely nontrivial in discrimination of different objects, vacancy of which would lead to miss judgment. Based on the illumination model, a matching algorithm using the combined color and spatial features is proposed to extract the local invariant features of the objects in this paper. The robust of our algorithm is strengthened and the time-consuming in matching is reduced compared to SIFT algorithm.