A content-based web image retrieval system for person identification structured on the SSIM, PSNR ve SNR

Image searching is one of the most important services that need to be supported. In general, 2 different approaches have been applied to allow searching of image collections: one is based on image textual Meta data and the other is based on image content information. Querying by is called content-based image retrieval (CBIR). CBIR systems involve research on databases and image processing, handling problems that vary from storage issues to friendly user interfaces. In this study, a computer vision method, which is a content-based web image retrieval system, is proposed for automatically identifying person. We have developed an image retrieval system named as the (PICBIR=Person identification with content based image retrieval), which uses structural similarity (SSIM) as the method for retrieving person images, similar to a given set of reference same images in a database. PSNR and SNR are among the methods that have been widely used to study images for the purpose of comparison. The FEI database was used to test the particular methods used in this study. The current database consists of 2800 images that were taken at different positions and exposures of 200 people. It has been evident that the method proposed in the inquiries that was made for different male and female samples is successful in face recognition.