Publication Date: 2023/07/15
Abstract: Sugarcane is an important cash crop worldwide, and disease outbreaks can cause significant economic losses. Early detection and timely management of sugarcane diseases are crucial for maintaining crop yield and quality. In this project, we propose a deep learning-based approach for sugarcane disease prediction using transfer learning and Adam optimizer. We utilized pre-trained models, including AlexNet, VGG16, VGG19, MobileNet, DenseNet, ResNet50, and EfficientNet, to train our model on sugarcane disease images, including Red Rust, Red Rot, Bacterial Blight, and Healthy Crops. The proposed model achieved high accuracy and performance, with ResNet101 outperforming other models. The project demonstrates the effectiveness of transfer learning in improving the classification of sugarcane diseases. The developed system can assist farmers in early detection and timely treatment of sugarcane diseases, thus increasing crop yield and reducing losses.
Keywords: Deep Learning, CNN, Transfer Learning, Alexnet, VGG 16, Adam’s Optimizer, epoch.
DOI: https://doi.org/10.5281/zenodo.8149167
PDF: https://ijirst.demo4.arinfotech.co/assets/upload/files/IJISRT23JUN2073.pdf
REFERENCES