Authors:
S. Rubin Bose, J. Angelin Jeba, O. Jeba Singh, R. Regin, P. Velavan, Rahul Panakkal
Addresses:
Department of Computer Science and Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai, Tamil Nadu, India. Department of Electronics and Communication Engineering, S. A. Engineering College, Chennai, Tamil Nadu, India. Department of Academic Research, Alliance University, Bengaluru, Karnataka, India. Department of Computer Science and Engineering, Dhaanish Ahmed College of Engineering, Chennai, Tamil Nadu, India. Department of Computer Science and Engineering, University of Illinois, Champaign, Illinois, United States of America.
Abstract:
Monitoring plant health is essential for improving crop productivity and minimising losses due to diseases. Traditional methods rely on manual inspection, which is often time-consuming, inconsistent, and prone to error. Existing automated systems primarily focus on disease detection but lack features such as severity quantification, causal analysis, treatment recommendations, and continuous monitoring, limiting their effectiveness for practical agricultural use. To address these challenges, this study proposes an integrated leaf health monitoring system that combines disease detection, severity assessment, causal diagnosis, treatment guidance, and pesticide evaluation. The system architecture includes image acquisition of leaves, preprocessing, feature extraction, and classification to determine whether a leaf is healthy or unhealthy. It calculates the percentage of affected areas to quantify disease severity and identifies potential causes, such as fungal, bacterial, or environmental stress factors. Upon detection, real-time alerts are generated, and suitable treatments, including recommended pesticides, are suggested. A temporal tracking module provides daily updates over seven days, enabling farmers to monitor the effectiveness of interventions and adjust treatments accordingly. The system was trained and tested using publicly available datasets of healthy and diseased leaf images, achieving high accuracy in disease detection and reliable quantification of infection severity. Experimental results demonstrate that this approach effectively integrates detection, analysis, and actionable guidance, offering a practical, data-driven solution for precision agriculture. By integrating real-time monitoring with treatment recommendations, the proposed system supports sustainable crop management, reduces crop loss, and enhances overall farming efficiency.
Keywords: Plant Disease Detection; Machine Learning; Precision Agriculture; Treatment Recommendation; Support Vector Machine (SVM); Naïve Bayes; Real-Time Monitoring; Sustainable Farming; Crop Health Management.
Received on: 31/03/2025, Revised on: 22/07/2025, Accepted on: 23/10/2025, Published on: 07/06/2026
DOI: 10.64091/ATIHL.2026.000303
AVE Trends in Intelligent Health Letters, 2026 Vol. 3 No. 2 , Pages: 87-100