Region Based Features for Image Analysis and Retrieval Using Local Histogram Refinement

Content based image retrieval is highly used in computer vision in order to solve image retrieval problems. The proposed method is based on the information provided by the histogram analysis of intensity/grayscale values of images. Some additional properties are also calculated and used that are based upon regional characteristics of various objects in the image. The need to retrieve the additional regional properties arises due to the fact that the standard histograms are insensitive to small changes. Many images of different appearances can have similar histograms, because, histograms provide coarse characterization of an image, which is the main disadvantage of histograms. The research is based on the concept of Histogram Refinement [1]. Distributing the grayscale image intensities by splitting the pixels by their intensity values into several classes just like histogram refinement method can provide an estimate of object characteristics present in an image. The classes are all related to colors and are based on color coherence vectors (CCVs). After the calculation of clusters using color refinement method, inherent features of each of the cluster is calculated based on the regional properties of the clusters. These additional region based features expound some structural information of the image. Finally, all of these features are used for image retrieval.