Content Based Image Retrieval (Cbir) Using Hierarchical Nested Dynamic Clusters

Abstract

In Content-Based Image Retrieval (Cbir) System, Query Results Of Images Are Searched And Sorted From The Image Database By Feature Similarities Based On The Given Query. Many Of The Image Retrieval Systems Extract Specific Features From A Query Image And Compare These Features With The Corresponding Pre-Computed Features Of All The Images In The Database. Still, Growing Size Of The Database Results In Long Search Which Delays That May Be Unacceptable In Many Practical Situations. Our Approach To Solving This Problem Is To Create Image Representation, An Image Indexing Scheme And  By Grouping The Images Beforehand Based On The Content, So That At The Time Of The Query, Only The Relevant Set Of Clusters Need To Be Examined. The Clustering Must Be Performed In Such A Way That The Retrieval Accuracy Is Not Sacrificed In This Process