Content-Based Image Retrieval Using Salient Points and  Spatial Distribution

In content-based image retrieval, the method based on salient points detection is one of the most active research areas for it can represent the local properties of the image. This paper proposes an improved salient points detector based on wavelet transform which can extract salient points more exactly. Then, an annular segmentation algorithm based on salient points distribution is designed, which takes not only the local image features into account, but also the spatial distribution information of the salient points. Color moments and Gabor features of the regions around the salient points in every annular region were computed as a features vector used for indexing the image. We tested the proposed scheme using a wide range image samples from the Corel Image Library, the experimental results indicating that the method has produced promising results.