International Journal of Applied Engineering Research
ISSN 0973-4562 Volume 9, Number 24 (2014) pp. 28533-28546
© Research India Publications
http://www.ripublication.com
Paper code: 29956 IJAER
An Image Retrieval by Content of the Image using
Color, Texture and Shape
Purohit Shrinivasacharya
1*
and Dr. M. V. Sudhamani
2
1*
Department of Information Science and Engineering
Siddaganga Institute of Technology,
Tumkur, 572103, India
Email: purohitsn@gmail.com
2
Department of Information Science and Engineering
R. N. S. Institute of Technology, Bangalore, 560098, India
Email: mvsudha_raj@hotmail.com
Abstract
As per the recent survey, it is noticed that image storage and retrieval on the
World Wide Web (WWW) has made significant progress. The present
searching systems have still confronts for us while searching the images from
the web. In this paper, a new idea to extract the images from the image
database is presented to achieve better performance. The proposed CBIR
system is implemented by considering the low level visual features of an
image. The low level features are obtained by converting the color image into
a grayscale image. The color features are obtained using modified Block
Truncation Coding (BTC) method. The four statistical properties (energy,
contrast, entropy and uniformity) of the texture extracted as a second low
level features. The third visual property of image is extracted as shape feature
using shape descriptor. Finally, an experiment is conducted on the ground
truth image database that has more than 1000 images of different categories.
In our previous work a middle range of block size has given best performance
as color features. A block size of 32 x 32 is chosen for the experiment to
enhance the performance of the system. The proposed system performance
has been improved by combining these three features.
Keywords: World Wide Web, BTC, Color, Texture, Shape, Database.
Introduction
The image data in the web is growing very fast in the present day because of social