A Review on Plant Leaf Disease Classification Using Deep Learning

Zahraa Y.  Younes; Elham Mohammed Thabit A.  Alsaadi

doi:10.24996/ijs.2024.65.11.43

Authors

Zahraa Y. Younes Department of Computer Science, College of Computer Science and Information Technology, University of Kerbala, Karbala, Iraq https://orcid.org/0009-0007-4557-2386
Elham Mohammed Thabit A. Alsaadi Department of Information Technology, College of Computer Science and Information Technology, University of Kerbala, Karbala, Iraq

DOI:

https://doi.org/10.24996/ijs.2024.65.11.43

Keywords:

Deep learning, DenseNet, ResNet, VGG, EfficientNet, Plant disease, classification, KNN, CNN, SVM

Abstract

The Food and Agriculture Organization (FAO) claims that plants provide more than 80% of the world's food supply. The report (2023), found that plant diseases are responsible for an estimated 40% of global crop losses each year. Therefore, it was necessary to find a solution that detects and classifies plant diseases early in the plant growth stage, using deep learning techniques. In this review, we will discuss the techniques for detecting plant diseases using leaves that have been studied in previous years. An overview of the articles cited in this review reveals that EfficientNet outperforms all other CNN models for plant leaf disease classification. It achieves an impressive accuracy rate that surpasses even the best-performing VGG, ResNet, and DenseNet models. DenseNet is also a good option, especially when computing resources are limited.