Machine Learning, Artificial Intelligence and Crop Saving
Ayantika Ghosh
Division of Biochemistry, Indian Agricultural Research Institute, PUSA, New Delhi (110 012), India
Madhurya Ray*
Dept. of Plant Physiology, Institute of Agricultural Sciences, Banaras Hindu University, Varanasi, Uttar Pradesh (221 005), India
DOI: NIL
Keywords: Artificial Intelligence (AI), Convolutional Neural Network (CNN), Plant diseases, Precision agriculture
Abstract
The food chain cannot exist without plants, yet biotic stress can lead to plant diseases, which can result in significant production losses. These illnesses are prone to errors and are challenging to identify manually. Technologies based on artificial intelligence (AI), in particular machine learning (ML) and deep learning (DL), provide means of early disease identification. This study presents an innovative lightweight Convolutional Neural Network (CNN) model, derived from the VGG-19 architecture, designed for precise classification of bacteriosis in peach leaves. A thorough dataset comprising both bacteriosis-affected and healthy leaves was assembled and preprocessed to optimize classification performance. The newly proposed LWNet model was trained and evaluated against four other CNN models: LeNet, AlexNet, VGG-16 and the standard VGG-19. The LWNet model demonstrated remarkable performance with a 99% accuracy rate, surpassing the other models. These findings underscore the model's effectiveness in detecting bacteriosis, thereby supporting precision agriculture and enhancing crop health management.
Downloads
not found
Reference
Akbar, M., Ullah, M., Shah, B., Khan, R.U., Hussain, T., Ali, F., Alenezi, F., Syed, I., Kwak, K.S., 2022. An effective deep learning approach for the classification of Bacteriosis in peach leave. Frontiers in Plant Science 13, 1064854. DOI: https://doi.org/10.3389/fpls.2022.1064854.
Son, N., Chen, C.R., Syu, C.H., 2024. Towards artificial intelligence applications in precision and sustainable agriculture. Agronomy 14(2), 239. DOI: https://doi.org/10.3390/agronomy14020239.