Title:	TEXTURE BASED CLASSIFICATION OF DISEASED RICE LEAVES
Author(s):	Santanu Phadikar, Jaya Sil
ISBN:	978-972-8939-22-9
Editors:	Yingcai Xiao, Tomaz Amon and Piet Kommers
Year:	2010
Edition:	Single
Keywords:	Rice leaves, Run length texture, Modified Green, Bayes’ Classifier.
Type:	Poster / Demonstration
First Page:	523
Last Page:	525
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	Image texture is powerful information to analysis various types of images. In this paper, texture of rice leaves images are extracted and used to classify the normal leaf and diseased leaf of a rice plant. A new color plane (MG) has been formulated for feature extraction, based on which different statistical texture features are calculated to construct the co-occurrence and run length matrix. Finally, Bayes’ Classifier is employed for classification of rice leaves. Thresholding operation is applied for computation of run length value and the results of classification show that using the MG plane with threshold 3 gives accuracy of 91.1% better than the gray level and the hue plane under similar situation.
	Go Back