Automatic Sugarcane Disease Detection and Outbreak Alert System using Deep Learning

Abstract

Agriculture Agriculture is a vital sector in Pakistan, with sugarcane being the second-largest crop. However, the industry faces challenges due to crop diseases, prompting the need for innovative solutions. This research introduces the Automatic Sugarcane Crop Diseases Detection and Outbreak Alert System (ASCD-OAS), a deep learning-based system to revolutionize disease monitoring in sugarcane fields. The ASCD-OAS leverages computer vision and deep learning to automatically detect diseases like red rot, mosaic, yellow leaf, and rust with high accuracy. It employs a multi-stage architecture, including convolutional neural networks and advanced image processing, to analyze field images captured by mobile devices. This enables the early and precise identification of diseased plants. The system also integrates a real-time alert mechanism to promptly notify stakeholders of potential outbreaks. This proactive approach empowers farmers and authorities to take immediate action, mitigating the impact of diseases and improving productivity and sustainability. The research presents a comprehensive evaluation of the ASCD-OAS, demonstrating its effectiveness. The system achieved an impressive 94.04% accuracy in classifying sugarcane diseases using an ASCD-OAS model, showcasing its potential as a reliable solution for the agricultural sector in Pakistan. The successful implementation of the ASCD-OAS can transform the sugarcane industry, leading to increased yields, improved crop quality, and reduced economic losses. This innovative approach paves the way for the adoption of advanced technologies, fostering sustainable development and enhancing food security in the region.

Keywords

• Sugarcane
• Crop Diseases
• Deep Learning
• Computer Vision
• Convolutional Neural Networks
• Disease Monitoring