An integrated approach to Content Based Image Retrieval

Content based image retrieval, in the last few years has received a wide attention. Content Based Image Retrieval (CBIR) basically is a technique to perform retrieval of the images from a large database which are similar to image given as query. CBIR is closer to human semantics, in the context of image retrieval process. CBIR technique has its application in different domains such as crime prevention, medical images, weather forecasting, surveillance, historical research and remote sensing Here content refers to the visual information of images such as texture, shape and color. Contents of image are richer in information for an efficient retrieval in comparison to text based image retrieval. In this paper, we have proposed a content based image retrieval integrated technique which extracts both the color and texture feature. To extract the color feature, color moment (CM) is used on color images and to extract the texture feature, local binary pattern (LBP) is performed on the grayscale image. Then both color and texture feature of image are combined to form a single feature vector. In the end similarity matching is performed by Euclidian distance which compares feature vector of database images with query images. LBP mainly used for face recognition. But we are going to use LBP for natural images. This combined approach provides accurate, efficient, less complex retrieval system.