Friday 23rd of February 2018

Object Identification Using Raster Models:A Comparative Analysis

Akbar Khan and L.Pratap Reddy

Shape is most important visual feature of an image. Most of the region based shape descriptors are complex in nature and computationally intensive. In this paper, we analyzed various raster models like circular, square by exploring exterior contour information with the help of grid based approaches. Spiral models are introduced in this paper with a comparative study. We compute a one dimensional shape vector from two dimensional shape region. The retrieval performance of these raster models are evaluated over 400 test images.

Keywords: shape description, shape vector, spiral raster, object identification

Download Full-Text


Akbar Khan
Akbar Khan received B.Tech(ECE) from Nagarjuna University in 1990, M.E from SGSITS, Indore in 1994. Currently he is working as Assoc. Professor at Nimra Institute of Science & Technology, Vijayawada, in Dept., ECE. He is pursuing PhD at JNTUK, Kakinada under the supervision of Dr L. Pratap Reddy. His research interest includes Image Processing and Pattern Recognition

L.Pratap Reddy
L. Pratap Reddy received the B.E. degree from Andhra University (INDIA) in ECE in 1985, the M.Tech degree in Electronic Instrumentation from Regional Engineering College (Warangal) in 1988 and the Ph.D. degree from Jawaharlal Nehru Technological University (Hyderabad) in 2001. From 1988 to 1990 he was lecturer in ECE Department of Bangalore Institute of Technology (Bangalore), from 1991 to 2005 he was faculty member in JNTUCE (Kakinada). Since 2006 he is with ECE Dept., at JNTU, Hyderabad. His current activity in research and development includes, apart from telecommunication engineering subjects, Image Processing, Pattern Recognition and Linguistic processing of Telugu language. He published 25 technical papers, articles and reports. He is active member in professional bodies like ISTE, IE, IETE & CSI.

IJCSI Published Papers Indexed By:





IJCSI is a refereed open access international journal for scientific papers dealing in all areas of computer science research...

Learn more »
Join Us

Read the most frequently asked questions about IJCSI.

Frequently Asked Questions (FAQs) »
Get in touch

Phone: +230 911 5482

More contact details »