Efficient Processing Methodology for UAV Flight Path Detection

Unmanned Areal Vehicle (UAV) based imagery is an emerging technology that has penetrated numerous verticals such as remote sensing, precision agriculture, land surveying. Various types of sensors are mounted onto UAV, and the images of the area of interest are captured. To get a complete distortionless areal view of the area, an orthomosaic is created using the captured images on which further analysis is done. But the traditional orthomosaic creation techniques are tedious, time-consuming, and also computationally complex. In this paper, a novel algorithm is proposed which speeds up the region of interest (ROI) detection significantly. In this method, the UAV flight path is divided into multiple Sub-Paths, and each path is processed parallelly. This method is universal and drastically improves the processing speed for any set of UAV images. It is observed that the algorithm reduces the computation time by around 75%