Year: 2024
Type: article
Abstract: This research aims to develop a deep learning model for detecting and classifying four major sugarcane diseases found in Thailand: sugarcane mosaic disease, red rot, rust, and leaf scorch. The study utilizes sugarcane leaf images as input data. The development process includes data preparation, dataset diversification through data augmentation, and training models using EfficientNetB0 and MobileNetV2 to compare performance. The experimental results show that EfficientNetB0 achieved the highest accuracy with an F1-Score of 0.95. Furthermore, an analysis of the effects of image brightness and blur on model performance reveals that these factors do not impact prediction accuracy, demonstrating the model's robustness under varying environmental conditions. This research highlights the potential of applying the developed model for effective sugarcane disease detection and classification, supporting disease management in the agricultural sector.
Institutions: Uttaradit Rajabhat University, Khon Kaen University
Language: Thai
Topic: Smart Agriculture and AI
Subfield: Plant Science
Domain: Life Sciences
Open Access status: diamond
OpenAlex ID: W7220626697