Unmanned aerial vehicle (UAV) paired with LiDAR sensor to detect bodies on surface under vegetation cover: Preliminary test

Abstract

The use of unmanned aerial vehicles (UAVs) has become increasingly accessible, enabling their deployment in a diverse range of operational contexts. UAVs have been tested as part of search and rescue missions. Following the successful use of UAVs in the wilderness medicine literature, we questioned their ability to be used in forensic context to search for missing persons or human remains, especially under canopy cover. Subsequently, various sensors were then repurposed from their original applications to address forensic concerns. This preliminary study aimed to evaluate the efficacy of airborne Light Detection and Ranging sensors (LiDAR) in detecting a concealed human body on a surface within a densely vegetated search area. To proceed, two LiDAR sensors were tested with several modalities. A dendrometric method was used to estimate the tree density of the search area, and the Normalized Difference Vegetation Index (NDVI) was used to bring precision in the appreciation of canopy cover density. The results showed that airborne LiDAR sensors can capture body signatures in areas with dense vegetation. The ground point density reached 0.26 % in a high-vegetated area. The study highlighted the importance of refining data processing techniques, including point cloud selection and the implementation of true positive/false positive analysis, to improve detection accuracy. Furthermore, the potential integration of complementary sensors such as thermographic and multispectral sensors was discussed, which may enhance the detection of thermal anomalies and chemical markers associated with decomposition.