A Content based Image Retrieval Framework Using Color Descriptor

Abstract

In this work, we propose an image search method by visual content (CBIR), which is based on the color descriptor. The proposed method take account the spatial distribution of colors and make the signature partially invariant under rotation. The basic idea of our method is to use circular shift (clockwise or anticlockwise direction) and mirror (horizontal direction and vertical direction respectively) matching scheme to measure the distance between signatures. Through some experiments, we show that this approach leads to a significant improvement in the quality of results.

Image Retrieval Using DCT/KWT and D4/KWT in Distributed System

“…The CBIR techniques are tested on the image database [14] taken 200 variable size images spread across 4 categories of Buses, Dinosaurs, Rose and Elephants. The categories and the distribution of the images are shown in table (1). The database is exists in distributed system where clients/ server as a star topology.…”

“…In this section we propose an Algorithm for CBIR using DCT/KWT and D4/KWT, Figure (1) shows this procedure in flow chart. 7) and save the result.…”

“…The classification step matches the features extracted from a query image with the features of the database images and groups' images according to their similarity. Out of the two steps, the extraction of features is considered most critical because the particular features made available for discrimination directly influence the efficacy of the classification task [1].…”