An Image Retrieval using combined approach Wavelets and Local Binary Pattern

Abstract

With the invent of Internet and the availability of efficient image capturing devices such as image scanners, digital cameras and high capacity public networks, cheap storage; the volume of digital images is increasing exponentially. This created a need of image searching, retrieval and browsing tool for users from various domains including fashion, crime prevention, publishing, remote sensing, architecture, medicine etc. Content Based Image Retrieval (CBIR) provides a solution for above said issues. Content based image retrieval is the utilization of computer vision techniques to the issue of digital image searching in large databases. The basic principle is the representation of image as a feature vector and to measure the similarities between the query image and feature vectors of images in the database using image processing techniques. Determining correct features to represent the images and the similarity metric that groups visually similar images together are the two main milestones in construction of any CBIR system. In CBIR, the images are sorted/indexed based on visual features, such as color, texture, shape, motion, structure or combining above different features.