Content based Image Retrieval from Forensic Image Databases

Due to the proliferation of video and image data in digital form, Content based Image Retrieval has become a\ud prominent research topic. In forensic sciences, digital data have been widely used such as criminal images,\ud fingerprints, scene images and so on. Therefore, the arrangement of such large image data becomes a big issue\ud such as how to get an interested image fast. There is a great need for developing an efficient technique for\ud finding the images. In order to find an image, image has to be represented with certain features. Color, texture\ud and shape are three important visual features of an image. Searching for images using color, texture and shape\ud features has attracted much attention. There are many content based image retrieval techniques in the literature.\ud This paper gives the overview of different existing methods used for content based image retrieval and also\ud suggests an efficient image retrieval method for digital image database of criminal photos, using dynamic\ud dominant color, texture and shape features of an image which will give an effective retrieval result