Content Based Image Retrieval Using Automatic Image Annotation

Generally images are stored in the database by its name and search operation is performed by its name only. As the database contain large no. of images and other data, it is difficult to find the required image. The performance of existing method is inefficient and less accurate. Hence an approach is proposed in which image is identified by its contents. First the image is stored in the database and then while extracting the features it is considered as a single segment. Different features are extracted from the image. If there is only one content in the image then whole image is considered as single unit. Then each content belong to a particular image is tagged using a tag file. Precision and Recall values will be used while evaluating the performance of CBIR systems. User gives the query as input to the system which is actually the content in the image. To retrieve the image it tag each segment with other, and search image in database, if match found then required image will be given as output. As images take more time and space while transmission, this system sends tag file through network to reduce time and space. If images are similar then differentiate using another feature like color, texture, version or format.