In the traditional image retrieval methods based on wavelet transform, the wavelet transform and the importance of the different color components of HSV color space are not considered. For this question, this paper proposes an new image retrieval method using wavelet transform in HSV color space. Firstly, each image would be transformed from RGB color space to HSV color space. Then, each color component(H,S,V) of the image is performed by wavelet transform to extract the texture features of the image, and the feature vector of the image is also obtained. Finally, calculating the similarity between the retrieved image and each image in database by assigning the weights of three color components reasonably with the importance, and get the retrieval results. The results of experiments show that the algorithm given in this paper can get better retrieval performance.