Detection & Identification of Rice Leaf Diseases using Multiclass SVM and Particle Swarm Optimization Technique
Prabira Kumar Sethy1, Nalini Kanta Barpanda2, Amiya Kumar Rath3

1Prabira Kumar Sethy, Department of Electronics, Sambalpur University, Burla, Sambalpur, India.

2Amiya Kumar Rath, Department of Computer Science & Engineering, Veer Surendra Sai University of Technology, Burla, Sambalpur, India.

3Nalini Kanta Barpanda, Department of Electronics, Sambalpur University, Burla, Sambalpur, India.

Manuscript received on 10 April 2019 | Revised Manuscript received on 17 April 2019 | Manuscript Published on 24 May 2019 | PP: 108-120 | Volume-8 Issue-6S3 April 2019 | Retrieval Number: F22200486S219/19©BEIESP

Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (

Abstract: In India the economic, political and social stability depend directly as well as indirectly on the annual production of rice. The income of hundreds of millions of people depends only on rice production and nothing else. However, as per the report of International Rice Research Institute (IRRI), 37% of the rice yield loss is due to diseases. In this consequence, the farmer can take care of crop on-time with apposite treatment. The disease detection and identification in large field through automatic technique is really useful as it reduces the work, time and cost for observation and evaluation of disease symptoms. This paper reports a novel approach for detection and identification of rice leaf diseases by K-means clustering, multi class SVM and PSO. Gray Level Co-occurrence matrix (GLCM) is used for feature extraction. The disease classification is done using SVM classifier and the detection accuracy is improved by optimizing the data using PSO. The investigational outcomes exhibit the performance of planned methodology in terms of accuracy of disease detection is 97.91%. However, in case of K-Nearest Neighborhood (KNN), Feed Forward neural network (FFNN) and SVM is 77.96%, 85.64% and 90.56% respectively.

Keywords: Disease Detection, FFNN, GLCM, Image processing, KNN, Particle Swarm Optimization, SVM Classifier.
Scope of the Article: Computer Science and Its Applications