NCTM - SLIIT 2009 >

Please use this identifier to cite or link to this item:

Title: A Simple Method for Classification of Endemic Plants in Sri Lanka by Herbarium Images
Authors: Wijesingha, D.
Marikar, F.M.M.T.
Keywords: Endemic plants
Image processing
Probabilistic Neural Network
Issue Date: Dec-2009
Publisher: SLIIT
Citation: PSRS2009
Series/Report no.: SLIIT/LIB/;
Abstract: In this paper, we describe an automatic detection system for plant classification for endemic plants in Sri Lanka. Stemonoporus a genus of Dipterocarpaceae family which has about 17 species of plants had been selected for the proposed system. Images of National Herbarium specimen samples were used. Digital pictures of leaves were enhanced, segmented, and a set of features were extracted. Features were then used as inputs to a Probabilistic Neural Network which is used in MATLAB classifier and tests were performed to identify the best classification model. Several classification models were assessed via cross-validation. The results of this study suggested that: leaf width, length, perimeter and area related features might be used as factors for prediction; and that machine vision systems might lead to the successful prediction of targets when fed with appropriate information. The overall classification accuracy utilizing the proposed technique is 85%, whereas using feature obtained 95% accuracy. The results obtained showed that this proposed technique is effective in performing plant species classification by herbarium samples.
ISSN: 1800-3591
Appears in Collections:NCTM - SLIIT 2009

Files in This Item:

File Description SizeFormat
Page 17-20.pdf106.2 kBAdobe PDFView/Open
View Statistics

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.


Valid XHTML 1.0! DSpace Software Copyright © 2002-2010  Duraspace - Feedback