Comparing UAV RGB, Thermal and LiDAR Cameras for the Optimal Detection of Brown Hares (Lepus europaeus) on Rosemains Farm, Pathhead

This research paper tests three Unmanned Aerial Vehicle (UAV) mountable camera’s ability to detect the Brown Hare (Lepus europaeus) on the agricultural site Rosemains Farm, Pathhead. The three cameras are RGB, thermal, and LiDAR. Detection technologies have largely focused on larger mammals and struggled to detect and identify smaller mammals. RGB and thermal images were visually analysed for hare detections and hare pixel temperature and band values extracted for statistical analysis, testing for any statistically significant differences between hare values and surrounding vegetation (wildflower and crop) values. LiDAR point clouds were visually analysed. The thermal proved most successful at detecting brown hares, with the RGB often needing the thermal heat signature as location guidance and the LiDAR failing to detect mammals. However, though the thermal was successful in terms of detection, it struggled in terms of identification. Further research is needed into a methodology of lower altitude surveys for small mammal species detection and identification, as well as to increase chances of successful semi-automated detection for small mammals.