Integrated Color Image Retrieval EHD Features

In1992, T. Kato first proposed a novel image retrieval technology-content-based image retrieval technology (Content-Based Image Retrieval, referred CBIR). CBIR technology has been made great progress in the feature.In this article, we focus on image color features and the edge histogram retrieval (Edge Histogram Detection referred EHD) technology. In the beginning of the paper, the background and application of the CBIR are proposed. At the same time, the worldwide status of researching of CBIR and the key technology are discussed, and then the contrast is discussed between some classic CBIR systems. This paper is listed the used CBIR system which is based on the analysis of its strengths and weaknesses, and proposed a comprehensive color and EHD features image retrieval algorithm on the basis of color and shape features.The study focused on the following aspects:First, extracting the color features can be improved. Then, it is transformed from the RGB color space into HSV color space which is more suitable for human vision than the RGB color space. In order to reduce the complexity of the algorithm, the color space is quantized to256dimensions and the paper is proposed the Sub-block Dominant Color. Second, a new edge histogram (EHD) is proposed, which is used the spatial relations filter noise reduction and an adaptive threshold to extract edge. Using Matlab, the validity of this method is verified by simulation experiments. Image retrieval algorithm which is based on improved edge detection and color feature is feasibility and effectiveness. Finally, the paper is proposed the algorithm in the deep research and analysis, combined with the development trend of CBIR system, the paper is proposed the algorithm to improve.