An Intelligent Localization Method for Underground Space Targets Based on the Fusion of Camera and LiDAR

The search and positioning of underground space targets are of significant importance in various fields. However, existing technologies still exhibit low efficiency and accuracy in complex underground scenarios. This paper proposes an intelligent positioning method for underground space targets based on the fusion of cameras and LiDAR. Firstly, the real-time detection of underground targets via a camera is performed, then it combines LiDAR point cloud maps to calculate the position of each element in the current image within the reference frame of the constructed point cloud map. Finally, a fusion and deduplication strategy is employed to suppress duplicate targets. The resulting target positioning distribution map provides an effective and intuitive representation of the reconnaissance results for underground targets.