Feature Extraction using Overlap Blocks for Content based Image Retrieval

Abstract

The paper presents an extension of content based image retrieval (CBIR) techniques based on Overlap Modified Block Truncation Coding (BTC) and Overlap Gabor Magnitude. Modified Block truncation coding based features is one of the CBIR methods proposed using color features of image. The approach basically considers red, green and blue planes of an image to compute feature vector. This MBTC based CBIR can be extended as Overlap Modified BTC for performance improvement in image retrieval. The Overlap Gabor Magnitude technique is used for texture features of image. The CBIR techniques like MBTC, Gabor and Fusion of MBTC and Gabor are further modified using overlap blocs technique to analyze and compare their performances. The proposed CBIR technique is implemented on a database having 1000 images spread across 11 categories and COIL image database having 1080 images spread across 15 categories. For each proposed CBIR technique 55 queries (5 per category) are fired on the database and net average precision and recall are computed. The results have shown performance improvement (higher precision and recall values) with Overlap (Fusion technique , Gabor Magnitude and Modified BTC compared to Non-Overlap (Fusion technique , Gabor Magnitude and Modified BTC).In all Overlap technique based CBIR system gives best performance.. General Terms Algorithm, Performance Keywords Content based image retrieval, Modified BTC, Gabor Magnitude, color, texture.