UAV High-Resolution Hyperspectral Imaging for Monitoring Phomopsis Stem Canker in Sunflowers

This study investigates the potential of high-resolution hyperspectral imaging from unmanned aerial vehicles (UAVs) for the early detection and monitoring of Phomopsis stem canker in sunflower (Helianthus annuus L.). Field experiments were conducted using two sunflower cultivars with differing susceptibility to the pathogen, under controlled conditions involving pathogen inoculation and biostimulant treatments. Hyperspectral data were acquired using a Headwall Nano-Hyperspec sensor mounted on a DJI Matrice 300 RTK drone, and processed to generate reflectance orthomosaics. A suite of vegetation indices sensitive to pigment content and physiological stress was computed and analyzed using multifactor ANOVA. The results indicate that while early-stage detection remains challenging, indices such as NIRv and MTVI2 showed moderate sensitivity three weeks post-inoculation. At five weeks, several indices, particularly those related to chlorophyll and anthocyanins (e.g., EVI, ARI2), exhibited strong responses to pathogen presence. The biostimulant treatment did not demonstrate a clear protective effect. These findings highlight the potential of UAV-based hyperspectral imaging for non-destructive disease monitoring in precision agriculture.