Unmanned Aerial Vehicle (UAV) for Crop Health Mapping and Estimation.
Abstract:
Zero hunger, the second goal of the sustainable development goals (SDGs), faces significant challenges worldwide due to crop pests and diseases leading to major yield losses. Detecting and treating crop health issues promptly and accurately is essential for achieving zero hunger, especially in regions like Africa, where hunger and scarcity have reached alarming levels. Precision agriculture can play a crucial role in improving economic growth and agricultural system sustainability. This research aims to develop a low-cost geospatial automated system for assessing crop health, leveraging technological tools like visible light unmanned aerial vehicles (UAVs).
The study area spans approximately 21 hectares between the Federal University of Technology main campus and Garatu village. Images with 1.0cm resolution were acquired using a visible light camera (RGB) after site plan design, pre-marking, and establishment of Ground Control Points (GCPs). Image processing involved three main stages: Initial processing, point cloud mesh generation, and creation of digital terrain models (DSMs), orthophotos, and index maps. The algorithms employed for these stages were Scale Invariant Feature Transformation (SIFT), Bundle Block Adjustment (BBA), and Structure from Motion (SFM), respectively.
Various vegetation index maps (NDVI, VDVI, NGRDI, NExG maps) were developed from the orthophotos to estimate the healthiness of crop samples collected from the field. The extracted values of the sample crops showed different ranges for each vegetation index. For instance, NDVI values ranged from 0.32 to 0.96, VDVI from -0.25 to 0.63, NGRDI from -0.23 to 0.43, NExG from -0.19 to 0.42, and VARI from -0.82 to 0.86. The crop samples exhibited relatively fair healthy characteristics in the vegetation indices, with NDVI and VDVI standing out as the most optimal and reliable among them.
Furthermore, statistical analysis was performed to correlate the NDVI value of each crop with its respective tissue test result. The analysis demonstrated that results from photogrammetry could be more reliable in estimating the health of crops such as cassava and rice, among others, with a 100% chance of obtaining the same results. This research contributes to the advancement of precision agriculture and supports the quest for achieving zero hunger by ensuring better crop health assessment and management.
Unmanned Aerial Vehicle (UAV) for Crop Health Mapping and Estimation. GET MORE, ACTUARIAL SCIENCE PROJECT TOPICS AND MATERIALS