An Efficient Re-Ranking of Web Images Using Incremental Learning and Hashing Approach

Objective: The main objective of this method is to retrieve the image contents from the web server accurately as per the user requirement. This is done by introducing the web page re-ranking mechanism which aims to retrieve the contents content with the knowledge of semantic meaning along with textual information

Method: A novel image re-ranking framework that applies incremental SVM learning algorithm and locality sensitive hashing algorithm is proposed in this work to overcome the problem resides in the existing methodology and retrieve the accurate contents as per the user requirement. This approach is based on the knowledge of universal visual semantic space for highly diverse images.

Results: The performance evaluation of the proposed methodology is done by comparing it with the existing approaches to prove the improvement of proposed approach. The comparison is done by using the performance metrics called the precision, accuracy and recall from which it is proved that the proposed method provides better result than the existing approach.

Conclusion: The finding of this work concludes that the novel re-ranking mechanism retrieves the accurate results than the existing approach with higher accuracy, precision and recall rate.