Precision Crop Farming Framework for Small-Scale Rainfed Agriculture Using UAV RGB High-Resolution Imagery
ABSTRACT
This paper presents a precision crop farming framework developed for small-scale rainfed agriculture using unmanned aerial vehicle (UAV) red, green, and blue (RGB) high-resolution imagery. The aim is to enhance farm management by providing precise spatial and temporal information in heterogeneous farming systems in Botswana's semi-arid regions. The precision crop farming framework integrates UAVs and Global Navigation Satellite System (GNSS) data, introducing new vegetation indices and employing machine learning algorithms for high-accuracy crop and land use analysis. The framework comprises four components: data collection, applications, data processing, and users. Methods included UAV data acquisition, global navigation satellite system geo-referencing, and machine learning classification. Results demonstrated high spatial resolution and classification accuracy, providing actionable insights into crop conditions, planting patterns, and farm variability. The precision crop farming framework is a tool for improving agricultural productivity and sustainability, providing a foundation for efficient, data-driven farm management practices.
KEYWORDS
Unmanned aerial vehicle imagery, small-scale rainfed architecture, geospatial information system, machine learning, remote sensing.