Sunflower Lodging Detection and Monitoring Through UAV-Based Multispectral Data

Sunflower lodging significantly hampers agricultural productivity, resulting in notable yield reductions and increased harvesting challenges. This study investigates the effectiveness of UAV-based multispectral imagery in monitoring sunflower lodging across various sunflower varieties. High-resolution multispectral data is collected from 117 sunflower plots in Islamabad, Pakistan, from March to July 2023. We develop a comprehensive methodology to detect and measure lodging incidents using this data. The UAV is equipped with a Senterra Double 4K camera, capturing images in five spectral bands, including visible and near-infrared wavelengths. Advanced image processing techniques are employed to identify and map lodging areas, utilizing spectral indices sensitive to vegetation structure and health, like the Normalized Difference Vegetation Index (NDVI) and the Modified Soil-Adjusted Vegetation Index (MSAVI). The findings indicate that UAV multispectral imagery accurately and promptly detects sunflower lodging, showing a strong negative correlation (r = -0.83) between NDVI<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">mean</inf> and lodging severity. Additionally, the analysis highlights varietal differences in susceptibility to lodging. This study offers valuable information for breeding programs aimed at developing more resilient sunflower cultivars by identifying specific varietal responses to lodging. Moreover, this approach provides a scalable solution for monitoring lodging in extensive agricultural fields. The results emphasize the potential of UAV-based remote sensing technologies to enhance crop management, optimize yield outcomes, and mitigate the adverse effects of lodging and other detrimental events on sunflower production.