Integrated Color and Intensity MSF Scheme for Image Retrieval Systems

Markov Stationary Features (MSF) based on homogeneous chain model, for content-based image analysis, is getting popularity nowadays. It not only considers the distribution of colors, just as histogram method does, but also characterizes spatial co-occurrence patterns. However, handling a large-scale database images with degree heterogeneity, simple MSF sufficient to discriminate one requires. In this paper, an Integrated Color and Intensity (ICI-MSF) non-homogeneous Chain proposed overcome shortcoming. By incorporating intensities information colors exploiting time inhomogeneous concept, it possible improve certain aspects existing methods. Without compromising effectiveness robustness, in paper keeps feature level simplicity. Widely recognized databases namely WANG1000 Corel10800, are used evaluate compare performance algorithm The experimental results justify method.