Application of machine learning techniques in rice leaf disease detection. (2022)

• Record Type: Journal Article
• Authors: Pallathadka, Harikumar; Ravipati, Pavankumar; Sekhar Sajja, Guna; Phasinam, Khongdet; Kassanuk, Thanwamas; Sanchez, Domenic T.; Prabhu, P.
• Is Part Of: Materials today. Volume 51:Part 8(2022)
• Page Start: 2277
• Page End: 2280
• Publication Date: 2022
• Subjects: Machine learning -- Precision agriculture -- Classification -- Detection -- Leaf disease -- Preprocessing -- Feature extraction; Materials science -- Congresses -- Periodicals; 620.1
• DOI: 10.1016/j.matpr.2021.11.398
• Languages: English
• ISSNs: 2214-7853

Abstract

Abstract: The automated leaf disease diagnosis system is a precision agriculture system that predicts sickness by analyzing images of infected leaf disease with Computer Vision, Image Processing, and Machine Learning algorithms. Thanks to automated disease detection technology, which speeds up the diagnosis procedure, the farmer can make an informed decision about a plant sickness. Previously, the farmer had to submit the infected leaf to a pathology lab, where the pathologist confirmed the disease, a time-consuming procedure. As a result of the delayed reaction, crop productivity declines. As a result, it is important to automate the disease detection system in order to increase crop yield. This article presents a machine learning based framework for classification and detection of leaf disease. SVM, Naïve Bayes and CNN are used in framework. Preprocessing is done using histogram equalization. For feature extraction, PCA algorithm is used.