Leaf disease detection using machine learning and deep learning: Review and challenges

• Published in: Applied soft computing Vol. 145; p. 110534
• Main Authors: Sarkar, Chittabarni, Gupta, Deepak, Gupta, Umesh, Hazarika, Barenya Bikash
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.09.2023
• Subjects: Deep learning; Leaf species; Machine learning; Plant disease; Recognition models; Deep learning; Recognition models; Leaf species; Plant disease; Machine learning
• ISSN: 1568-4946; 1872-9681
• DOI: 10.1016/j.asoc.2023.110534

Identification of leaf disorder plays an important role in the economic prosperity of any country. Many parts of a plant can be infected by a virus, fungal, bacteria, and other infectious organisms but here we mainly considered the detection of leaf disease of a plant as a research topic. We have performed an in-depth study of this topic from 2010 to 2022 and found that many researchers use multispectral or hyperspectral imaging to study crop diseases. Machine learning (ML) and deep learning (DL) models are used to classify different types of leaf diseases. We made a workflow mechanism to help researchers in this field. Support vector machine (SVM), Random Forest, and multiple twin SVM (MTSVM) are popular ML models for predicting leaf disease, while convolutional neural networks (CNN), visual geometry group (VGG), ResNet (RNet), GoogLeNet, deep CNN (DCNN), back propagation neural networks (BPNN), DenseNet (DNet), LeafNet (LN), and LeNet are common deep learning models used for detecting leaf disease. Among these deep learning models, it is evident that models like CNN, VGG, and ResNet are highly capable at finding diseases in leaves. The performance of the algorithms is generally evaluated using F1 score, precision, accuracy and others. This review will be helpful for the researchers who are working in this area and looking for various efficient ML and DL-based classifiers for leaf disease detection. •A study on leaf disease (LD) detection is conducted using Machine Learning and Deep Learning from 2010 to 2022.•Various feature extraction techniques are discussed for LD detection.•Illustrations of various plant diseases are discussed with their images and sources.•Applications, challenges, limitations as well as some future works are also discussed.