A Review of Machine Learning Approaches in Plant Leaf Disease Detection and Classification

Abstract

To identify the recent advancements in the development of plant disease detection and classification system based on Machine Learning (ML) and Deep Learning (DL) models. In this study, we have collected more than 45 papers published during the year 2017-2020 from the peer-reviewed journals of different databases such as Scopus and Web of Science analogous to the keywords such as plant disease identification, recognition, and classification using ML and DL algorithms. An organized way of analysis of various plant disease classification models has been shown in well-formed tables. In this paper, we have conducted a systematic literature study on the applications of the state-of-the-art ML and DL algorithms such as Support Vector Machine (SVM), Neural Network (NN), K-Nearest Neighbor (KNN), Naïve Bayes (NB), other few popular ML algorithms and AlexNet, GoogLeNet, VGGNet, and other few popular DL algorithms respectively for plant disease categorization. Each stated algorithm is characterized through the corresponding processing methods such as image segmentation, feature extraction, along with the standardized experimental-setup metrics such as total number of training/testing dataset employed, number of diseases under considerations, type of classifier utilized, and the percentage of classification accuracy. This work will be a beneficial resource for researchers to recognize any particular type of plant diseases through data-driven approaches. The development of mobile-based applications using the studied ML/DL approaches will surely increase agricultural productivity.