Authors: Basavaraj S. Anami; N.M. Naveen; N.G. Hanamaratti
Addresses: Department of Computer Science and Engineering, K.L.E. Institute of Technology, Hubli – 580030, India ' Department of Information Science and Engineering, K.L.E. Institute of Technology, Hubli – 580030, India ' Department of Genetics and Plant Breeding, University of Agricultural Sciences, Dharwad – 580005, India
Abstract: The paper presents a methodology for recognition of varieties from bulk paddy sample images based on colour features extracted from different colour models such as RGB, HSV and YCbCr. The colour features used in the work are mean, range and variance. Feature set reduction is carried out based on the range of feature values and a reduced feature set consisting of seven significant colour features is adopted. A feed-forward neural network is used as classifier. The average recognition accuracy of 94.33% is achieved using the reduced seven colour features. The work finds application in developing a machine vision system in agriculture sciences wherein automation of recognition and classification of bulk food grains becomes possible.
Keywords: colour features; variety recognition; bulk paddy images; neural networks; image recognition; agriculture; machine vision; bulk food grains; classification; bulk rice grains; image acquisition; feature extraction; feature selection; paddy rice.
International Journal of Advanced Intelligence Paradigms, 2015 Vol.7 No.2, pp.187 - 205
Received: 17 Sep 2014
Accepted: 16 Mar 2015
Published online: 24 Jul 2015 *