A content-based image retrieval algorithm is proposed after researching feature extraction of image texture, shape feature extraction and relevance feedback algorithm. Fourier transform is used in the feature extraction of the texture. Boundary moments to detect the image boundaries is used in the feature extraction of the shape, similarity measuring function is used in image similarity match. And the introduction of relevance feedback algorithm in order to allow users to interact with the query system for image retrieval system to increase the adaptability features.