U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

Fast Pattern Matching Algorithm for detection of Wild Animal Hairs using SEM Micrographs

NCJ Number
International Journal of Scientific & Engineering Research Volume: 4 Issue: 6 Dated: June 2013 Pages: 1951-1956
Yadav Satendra Kumar; Dahiya Mohinder Singh
Date Published
June 2013
6 pages
This study tested a new algorithm for detecting wild animal hairs using SEM micrographs.
For hair detection of wild endangered animal species of feline family viz. Panthera leo persica, Panthera pradus fusca and Panthera tigris tigris, a fast pattern matching algorithm (FPMA) named as Dahiya and Yadav (DnY-FPMA) has been developed based on normalized cross correlation (NCC) and convolution techniques. The scanning electron microscopic (SEM) images were used for current forensic evaluation. FPMA method applied to recognize and/or locate specific objects in an image using correlation and convolution techniques. To improve the accuracy of the FPMA model, multiple reference templates were used for identification of unknown images. In this model, the best correlations of test SEM images at various magnifications have been made for getting clarity in results. The test images of above species were compared with standard images, which have proved that the correlation method found to be more accurate for evaluation than of the convolution method. FPMA method oriented the images properly and made them noiseless by rotating the inclined images with radon function having reference orientation. The obtained results revealed that the matching pattern for leopard, lion, and tiger were up to a maximum of 99.64 percent,99.50 percent and 99.60 percent with convolution while 99.58 percent, 99.68 percent and 98.87 percent with NCC respectively. (Published Abstract)


No download available