Masoumeh Bourjandi - Department of Computer, Aliabad katoul Branch, Islamic Azad University, Aliabad katoul, Iran

Image retrieval based on content, includes a group of methods, in which low level features like shape, texture, color, or high level semantic featuresare applied for image retrievals. One of the strong tools that today is widely used in CBIR systems, is fuzzy color histogram. Since fuzzy logic has been usedwidely used in determination of image effect, in this paper, we try to determine the color of each pixel in color fuzzy histogram more effectively, by use of etenfuzzy color histogram, suitable fuzzy rules, and color weight vector, and by use of place location feature for each color, increase the system's efficiency in order to extract good images from database. The results presented in this paper, show the high efficiency andaccuracy of this method in suitable image retrieval from database.

fuzzy color histogram, eten fuzzy color histogram, color weight vector, place location