Automatic Target Recognition using Invariant Target Features and Statistical Classifier

Nitin, V Kamath and Sudesh, K Kashyap and Shantha Kumar, N Automatic Target Recognition using Invariant Target Features and Statistical Classifier. In: International conference on innovative science and engineering technology, 8-9 April 2011, VVP Engineering College, Rajkot, Gujarat, India.

[img] PDF
ICISET-2011.pdf - Accepted Version
Restricted to Registered users only

Download (356kB) | Request a copy
[img] Indexer Terms (Generate index codes conversion from application/pdf to indexcodes)
indexcodes.txt
Restricted to Registered users only

Download (5kB) | Request a copy

Abstract

Automatic Target Recognition refers to the autonomous or aided target detection and identification by processing of information extracted from the various sensors. In this paper a novel concept is proposed to detect multiple targets and perform pre-classification to categorize the targets based on their aspect ratio. Features are computed using Zernike moments to yield target feature set which are invariant to shift, size and orientation. A novel approach for classification of the target features by Weighted Minimum Mean Distance is also introduced to reduce the false classifications. The proposed approach has been successfully verified on various target images

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Automatic Target Recognition, Canny Edge Detection, Zernike moments, Weighted Minimum Mean Distance, Target Aspect Ratio
Subjects: ENGINEERING > Engineering (General)
Depositing User: Dr Sudesh Kumar Kashyap
Date Deposited: 30 May 2011 10:02
Last Modified: 08 Jun 2011 08:25
URI: http://nal-ir.nal.res.in/id/eprint/9180

Actions (login required)

View Item View Item