Research On Retrieving Images Based On The Color Characteristic Of Images

In recent years, with the rapid development of Internet, the multimedia information is booming. All this information is mostly images. In the face of the immensity of information, people are eager to acquire information needed rapidly and efficiently. Effective recognizing, management and searching all these images have been an emergent problem.Most of The present image retrieval systems only concern the key word of the image description, neglecting another important element of image -the content of image. Content-Based Image Retrieval (CBIR) has been one of most active researches in multimedia technology and many research institutes have made great progress in this field. Content Based Image Retrieval (CBIR) means retrieving images based on the image content description features directly. The present research on CBIR focuses on the similarity-matching of lower-features, such as color, texture and shape.Within the image retrieval based on the content, the color, being important vision information of image, has been used broadly. Color is basal element of image content and is one of the main sensing features for recognizing images. Color histogram is used widely for retrieving images, and the processing mostly concerns the choice and quantization of color space, the definition of color histogram and the distance of two points in color histogram space.This paper have discussed three key questions on how to make use of the characteristic of color, which include expressing color, obtaining the characteristic of color and measuring the likeness based on color.Aimed at static image retrieval based color histogram, an algorithm of image retrieval method based on color apperception is presented in this article. To show color characteristic of image, the method of the HSV color space, which is suitable to the visual characteristic of human, is utilized. Taking advance of human's feeling to color, it quantifies color sector with unequal interval, and get characteristic vector. It matched the similarity of image with the distance function. Users can express their searching demand through the example mode. A new method of image indexing based on generalized color histogram is presented here. The paper combines the original image with its smoothed image to form a binary set. It calls this binary set as a "generalized image", and the histogram of the generalized image is a generalizedhistogram. To express accurately the information of color distributing of an image, in this paper, dominant color matrix is proposed. It can index an image by its dominant matrix with the basic principle of the classic histogram method.Experiment results show that the new method is very efficient, and the precision of the indexing output is improved largely compared with the classic histogram method.