Water Stress Diagnosis in Rainfed Wheat Through UAV Multispectral Imagery and IoT Data

Wheat, being a staple crop, plays a primary role in meeting nutritional needs globally and serves as a backbone of agricultural economies. To meet the food security demands of rapidly growing population under limited land and water resources, optimizing water use and crop yield through plant water stress identification at various growth stages is crucial, particularly in rain-fed areas. Recently, Unmanned Aerial Vehicles (UAV s) have become a powerful tool for monitoring crop health and environmental conditions. In this study, UAV borne multispectral imagery is captured for spring wheat from mid-season to the harvest. Soil Moisture (SM) content values are also recorded using an Internet of Things (IoT) Sensor. After photogrammetric preprocessing and registration, three vegetation indices (VIs) - Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), Normalized Difference Red Index (NDRI), and Excess Green (ExG) are extracted. To understand the relationship between soil moisture and remote sensing data, correlation of soil moisture with VIs is calculated. Moreover, machine learning models are build to estimate soil moisture values and subsequently, inverted water stress maps are generated. Spatial distribution maps of soil moisture are also created and the plots exhibiting significant water stress due to moisture deficiency are identified.