Efficient and Secure Image Reranking for TBIR

Abstract

Tag based image search plays an vital role to find images shared by clients in social networks. However, how to make the top ranked result relevant and with diversity. In previous papers these systems search image by using tags as query. However search results are weakly relevant tags and noisy tags. In this work we are using secure and efficient image re-ranking for TBIR. It overcomes the drawbacks that are present in the image re-ranking based on topic diversity.1) Tag mismatch 2) Query Ambiguity The main advantage of AES algorithm is to protect image data from unauthorized users. To detect and find unauthorized users for tag based image retrieval. AES is a type of symmetric key block cipher based on 10 rounds and key length is 128 bits .during encryption each round performs four transformations .sub bytes, shift rows, mix columns and add round key. AES is essential to secure the image data .