Developing an image processing pipeline to improve the position accuracy of single UAV images

Abstract

Unmanned aerial vehicle (UAV) based remote sensing has been extensively used in precision agriculture applications, such as vegetation growth and health monitoring, yield estimation, and irrigation management. Conventional procedures for UAV data collection and processing require collecting highly overlapped images, stitching images to generate an orthomosaic, and using ground control points (GCPs) in the field or UAV onboard real-time-kinematic (RTK) global navigation satellite system (GNSS) data to improve position accuracy. For improving efficiency, a previous study developed a framework to process individual UAV images for mapping cotton emergence. The current study aimed to build a near-real time image processing pipeline to further improve the positioning accuracy of single UAV images. The improved image processing pipeline comprised feature detection and matching, false matches removal, geometric transformation matrix calculation, crop row alignment, image position assignment, and mapping. The developed pipeline was tested for mapping in both cotton and corn fields. Results showed that the position accuracies for measuring the distance between GCPs were 0.32 ± 0.21 m and 0.57 ± 0.28 m in a cotton and a corn field, respectively, when compared to ground truth data collected with an RTK-GNSS. The developed pipeline did not require GCPs in the field or image post-processing steps, such as image mosaicking and feature extraction, which allowed processing in near-real time and may possibly be implemented in real-time using an onboard edge computing system. The pipeline was used to map emergence parameters for cotton and corn fields, including stand count, canopy area, mean days to imaging after emergence, and plant spacing standard deviation. These maps demonstrated the success of the developed methods in providing a low-cost near real-time tool (8.6 and 3.6 s/image for the cotton and corn fields, respectively) for mapping emergence parameters at field-scale for use in both research and agricultural production.