Fast and effective image retrieval using colour and texture features with self-organising map

Abstract

Content-based image retrieval (CBIR) is an emerging research area from the last two decades. Most of CBIR methods are still incapable of providing better retrieval results in less searching time. In this paper, we introduce self-organising map (SOM) clustering approach with fusion of features. Using SOM, system performances are improved by the learning and searching capability of neural network. Here, first we extract colour moment, colour histogram, local binary pattern, colour percentile, and wavelet transform-based colour and texture features. All these features are computationally light weighted, speedup the process of image indexing. Hereafter, all these features are fused together, and fed to SOM which generates clusters of images, having similar visual content. SOM produces different clusters with their centres, and query image content are matched with all cluster representative to find closest cluster. Finally, images are retrieved from this closest cluster using similarity measure. So, at the searching time the query image is searched only in small subset depending upon cluster size and is not compared with all the images in the database, reflects a superior response time with good retrieval performances. Experiments on benchmark database confirm the effectiveness of this work.