Image Retrieval by Fusion of Color Histogram and Edge Orientation Histogram based on wavelet

Abstract

It has very crucial realistic importance to research and research minority gown from the angle of pc imaginative and prescient for minority life-style protection and inheritance. As first exploration in minority get dressed image retrieval, this paper proposed a unique picture characteristic instance technique to offer a reason behind the rich records of minority get dressed photograph. Firstly, the color histogram and detail orientation histogram are calculated for divided sub-blocks of minority get dressed photo. Then, the final function vector for minority dress photo is typical with the resource of effective fusion of color histogram and detail orientation histogram. Finally, the improved Canberra distance is delivered to diploma the similarity amongst query photograph and retrieval photograph. We have evaluated the performances of the proposed algorithm on self-assemble minority robe photograph dataset, and the experimental consequences display that our Approach can efficiently specific the blanketed function of minority dress images, consisting of coloration, texture,  shape and spatial facts. Compared with some conventional techniques, our approach has higher and solid retrieval accuracy.