Heuristic Features for Color Correlogram for Image Retrieval

Content based image retrieval (CBIR) usually uses color features for representing images. The most common features of images are color histogram and color correlogram. The color correlogram was proposed to characterize not only the color distribution of pixels as in the case of color histogram, but also the spatial correlation of pairs of colors. Simple RGB values and their normal sequence r,g,b of pixels are used for color correlogram. In this paper, different statistical and heuristic operations are applied on the RGB values of pixels, and presented as the features of images. Simulation results show better results with the proposed features as compared to normal RGB features.