A Comprehensive Survey of Modern Content Based Image Retrieval Techniques

Abstract-The content based image retrieval (CBIR) is one of the most popular and heart favorite research areas of the digital image processing. All the existing image search tools, like Google, Bingo, Yahoo are based on textual annotation of images. In these tools, all the images are manually annotated with keywords and then retrieved using text-based search methods. Therefore performances of these systems are not satisfactory. The basic objective of content based image retrieval is to extract visual content of an image automatically, like color, texture, or shape. The CBIR technology can be used in several applications such as forensic laboratories, crime detection, and image searching sites. One most probable application is matching a forensic sketch to a set of previously available set of images. The research area of retrieving images which are based on the visual content of the query picture intensified recently. Also it demands on the quite wide methodology spectrum on the area of the image processing. In this paper, we will propose a more comprehensive survey of popular content based image retrieval techniques. We will discuss pros and cons of each CBIR methodology. I