Answers are grounded in this paper and cite their sources.
Unmanned aerial vehicle UAV-based RGB imaging is increasingly used to track crop diseases and assess fungicide performance. However, standard disease severity measures may miss an important factor: fungicide-induced changes in canopy structure that modulate disease observability. Here we analyse a multi-site dataset of wheat yellow and leaf rust, integrating UAV-derived canopy cover and disease severity, and rater-based disease severity assessments. Results indicate that between visual and UAV imagery-based disease severity was high but fundamentally non-stationary, governed by canopy cover and epidemic intensity. Measurement error followed structured, canopy-dependent regimes. The canopy-normalized disease area, a metric that harmonizes severity with structural and spatial context, revealed ranking shifts in fungicide efficacy that are not captured through traditional metrics. This suggests that severity-based comparisons may combine true disease suppression with treatment-related changes in how visible symptoms are. Framing plant disease quantification as a measurement challenge, rather than only a detection, may support a more accurate assessment of treatment effects. Modelling the influence of canopy structure on symptom expression could provide a scalable and robust basis for disease monitoring and decision-making in precision agriculture.
No takes yet. Share an insight, caveat, or question.
Jarroudi et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: