Implementation of Deep Learning Technique for Citrus Leaf Blotch Disease Severity Detection

Abstract

Utilising computer vision methods, there has been an ongoing study in recognizing and categorizing plant diseases. Citrus is a member of the plant family and is highly susceptible to disease, there hasn't been much research done on citrus disease detection. Citrus leaf blotch (CLB) disease can be detected and categorized based on how severe the illness is through a model for citrus leaf disease detection and classification. To categorize 8000 real-phase images of citrus leaves which include healthy and CLB-infected images, A deep learning (DL) model based on convolutional neural networks (CNN) has been presented.. This ranking accuracy of the CLB disease is 97.81% for binary classification and 98.81% for multi-classification, respectively. Additionally, cutting-edge pre-trained models have been compared., showing that it outperforms them in terms of multiple classifications of CLB sickness.