A hybrid approach for image retrieval using visual descriptors

Abstract

Retrieval of images using visual features is a hot research field in image processing used in various utilizations like in the business field, medical image, geographical images etc. In this research work, we propose and implement a fused approach for image retrieval technique using HSV Histogram color feature, LBP and SFTA texture feature of an image. Standardized Euclidean distance is operated as similarity check method. Wang image repository is used having 1000 images categorized into 10 classes of images for experimental evaluation. Experimental outcomes clear that the proposed system gives improved result of precision than the other conventional methods which is also presented in this paper.