AnInteractive Method for Content-Based Image Retrieval based on Relevance Feedback and Multiple GLCM features

Abstract

The manifold increase in theuse of digital media on the Internet has resulted in the increasing emphasis on the field of image search or retrieval.By ignoring visual content as a basis for image retrieval and adapting methods with text search techniques the search engines suffer inconsistency between the text words and visual content. Content-based image retrieval makes use of the representation of the visual content to identify relevant images and has attracted well-deserved attention in the last few years. CBIR combines the contents or features of images like color, texture, and edges instead of keywords, labels or metadata related to an image. The earlier CBIR methods were not able to bridge the gap between high-level concepts and low-level features. This work reviews the previous works carried out in this area. An approach ispresented for retrieval of images based on acombination of multiple GLCM features and relevance feedback based interactive approach. Performance evaluation of this method is done based on retrieval score and accuracy.