Title: Markov random field classification technique for plant leaf disease detection

Authors: Anusha Rao; Shrinivas B. Kulkarni

Addresses: Department of ISE, Dayananda Sagar Academy of Technology and Management, Bangalore, India ' Department of Computer Science and Engineering, SDM College of Engineering and Technology, Dharwad, India; Affiliated to: Visvesvaraya Technological University, Belagavi, India

Abstract: In recent era of technology, computer vision technique has grown attraction of the researchers. This technique helps to identify and classify the objects according to the application requirement. These techniques are widely used for plant leaf detection and helping to develop an automated process for plant leaf disease detection. A new approach is developed in this work for plant leaf disease detection using Markov random classification technique. MRF-based problem is formulated for disease detection. In the next stage, the general stages of computer vision classification model i.e., pre-processing and feature extraction is applied. For pre-processing, noise removal and image enhancement models are applied and feature extraction is combination of statistical features. Neighborhood pixel modeling and MRF classification models are applied to obtain the classification of input data. Performance of three classification models is compared. Study shows that proposed approach gives robust performance for plant leaf disease detection and classification.

Keywords: plant leaf; plant disease; computer vision; Markov random field; MRF.

DOI: 10.1504/IJCAET.2020.10022049

International Journal of Computer Aided Engineering and Technology, 2020 Vol.12 No.3, pp.336 - 354

Received: 29 Sep 2017
Accepted: 22 Jan 2018

Published online: 02 Apr 2020 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article