Image Retrieval from WWW using Attributes in HTML TAGs

Web pages contain large number of images and it is difficult to map the semantics of the images using text present in the HTML documents. Retrieval system is designed for ranking and retrieving images using ranking mechanism. These ranking mechanism use the text present in the HTML document and this alone may not be sufficient. In this paper, the text present in the TAG is analysed for retrieving relevant images from WWW. Each attribute in the TAG is carefully considered and categorized into four levels. A suitable weight is assigned to each level such that importance of each level is considered and accordingly the weight is assigned. The top level attributes in the TAG is assigned higher weight and lower weight is assigned to the lower level attributes. From the experimental result, we have observed that the performance of the proposed approach is encouraging compared to the recently proposed approach.