Development of a Novel Method for Disease Severity Driven Variable Rate Chemical Application Based on Plant Morphological Indicators
Tushar Dhar *
Division of Agricultural Engineering, ICAR-Indian Agricultural Research Institute, New Delhi, 110012, India.
Roaf Ahmad Parray
Division of Agricultural Engineering, ICAR-Indian Agricultural Research Institute, New Delhi, 110012, India.
Bishnu Maya Bashyal
Division of Plant Pathology, ICAR-Indian Agricultural Research Institute, New Delhi, 110012, India.
Tapan Kumar Khura
Division of Agricultural Engineering, ICAR-Indian Agricultural Research Institute, New Delhi, 110012, India.
*Author to whom correspondence should be addressed.
Abstract
Early blight, caused by Alternaria alternata, poses a critical challenge to tomato (Solanum lycopersicum L.) production, causing significant yield losses worldwide. Accurate quantification of plant disease severity is essential for the development of intelligent, site-specific crop protection systems. This study investigates the morphological responses of tomato plants following incidence of early blight disease across different stages of disease progression, with the objective of establishing biologically meaningful indicators for imaging-based disease severity classification. Key plant morphological parameters, including plant height, total leaf area, and diseased leaf area, were monitored over time and compared with healthy plants. Analysis of variance revealed a statistically significant difference in plant height between healthy and diseased plants after inoculation of disease, with average plant height after 90 days of growth were 94.86 and 81.81 cm respectively, indicating the impact of disease on overall plant growth. Temporal analysis of leaf area and diseased area exhibited distinct disease progression patterns, comprising an initial latent phase, a rapid symptom expansion phase, and a terminal phase characterized by tissue degradation. Disease severity was quantified using an area-based severity percentage derived from the ratio of diseased area to total leaf area, providing a normalized and scalable metric of infection intensity. The observed morphological and spatial disease characteristics closely correspond to features that can be extracted using machine vision techniques, such as changes in canopy geometry and lesion extent. The findings highlight the potential and the importance of severity based assessment of disease for variable-rate site-specific spraying systems, demonstrating clear advantages over conventional target-specific approaches in reducing chemical application, improving disease control efficiency, and supporting sustainable crop protection practices.
Keywords: Disease severity, early blight, machine vision, variable rate spraying