Query Processing for Content Based Image Retrieval
J. Sreedhar1, S. Viswanadha Raju2, A. Vinaya Babu3
1J. Sreedhar, Associate Professor, Department of CSE, DVRCET, Hyderabad, India.
2Dr. S. Viswanadha Raju, Professor, Department of CSE, SIT, JNT University, Hyderabad, India.
3Dr. A. Vinaya Babu, Professor, Department of CSE, Director of Admissions, JNT University, Hyderabad, India.
Manuscript received on October 04, 2011. | Revised Manuscript received on October 19, 2011. | Manuscript published on November 05, 2011. | PP: 122-131 | Volume-1 Issue-5, November 2011. | Retrieval Number: E0165091511/2011©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In this paper, we investigated image retrieval based on image content, Content Based Image Retrieval (CBIR) and proposed a framework to characterize the image content and similarity between the images. Our paper discusses the CBIR problem and the solution. Due to the enormous increase in image database sizes, the need for the development of CBIR systems arose. Firstly, this paper outlines feature extraction methods for color and texture. The extracted features for each image in the database used as the basis for similarity between the images. We built user interface based on Java, in which user can easily select query image and view top ten retrieved images based on decreasing order. We extended our approach to sub image retrieval also. Our results report that HSV based color features and contrast based texture features outperform than RGB based features and results reported in the paper are convincing.
Keywords: Colour, Texture, Query, CBIR.