Survey of Deep Learning Models for Image-Based Disease Detection in Plants

Crop diseases pose a severe risk to global food supplies, but accurate and timely diagnosis is hampered by a lack of resources in many regions. As the number of people with access to smartphones continues to rise throughout the world, and as recent advancements in computer vision allowed by deep learning pave the way, we may soon be able to use them to aid in the detection of diseases. Acquiring pictures, preprocessing, segmenting, and extracting features are all part of the image processing pipeline for disease detection in plants. In this review study, we take a look back at some of the most popular methods for classifying plant diseases: “Convolutional Neural Network (CNN)”, “Support Vector Machine (SVM)”, “K-Nearest Neighbors”, and “Artificial Neural Network (ANN)”. The results of the study show that the convolutional neural network method outperforms the more conventional techniques in terms of accuracy. This review study discusses the use of DL models to depict different plant diseases and gives a detailed description of their operation. Furthermore, several research gaps are noted from which better clarity may be obtained in regard with early detection of plant diseases.