Exponential Fourier Moment-Based CBIR System: A Comparative Study

Content-based image retrieval (CBIR) is a technique for browsing, searching, and retrieving images from a large database of image collections. The availability of large amount of image collections necessitates powerful algorithms for image retrieval. CBIR system extracts image information known as features that are used to retrieve relevant images from image database that best match with query image. A moment-based content-based image retrieval system is explored in this paper. Exponential Fourier moment-based CBIR system, improved exponential Fourier moment-based CBIR system, and accurate and fast exponential Fourier-based CBIR system are developed. Comparative analysis is performed on three moment-based CBIR systems in terms of average precision and retrieval time for the two benchmark databases GT face and COIL-100. Among the three moment-based CBIR systems, it is observed that accurate and fast exponential Fourier-based CBIR system delivers good results.