Search papers, labs, and topics across Lattice.
This study evaluates the effectiveness of various hyperspectral imaging techniques for detecting PFM-1 landmines using UAVs, focusing on the balance between false alarms and target discovery. By comparing ground-measured signatures, fully informed in-scene signatures, and a novel human-in-the-loop signature bootstrapping approach, the authors reveal significant differences in inspection efficiency and accuracy among the methods. The human-in-the-loop approach notably reduces the number of candidate reviews needed for target confirmation, achieving results comparable to fully informed signatures with fewer inspections required for certain algorithms like ACE.
Human-in-the-loop bootstrapping can dramatically cut down the inspection effort needed for accurate landmine detection in UAV hyperspectral imaging.
Hyperspectral imaging (HSI) is useful for material discrimination, but operational mine screening also depends on how many false alarms must be inspected before targets are found. This paper studies PFM-1 landmine detection in unmanned aerial vehicle (UAV) visible and near-infrared (VNIR) HSI using spectral angle mapper (SAM), matched filter (MF), adaptive coherence estimator (ACE), and constrained energy minimization (CEM). We compare a ground-measured SVC signature, a fully informed in-scene core-pixel signature, and a simulated human-in-the-loop signature bootstrap. Besides receiver operating characteristic area under the curve and average precision, we report target-discovery curves and spatial candidate-review counts. Full-review bootstrapping reaches the fully informed in-scene signature case after all seven target regions are verified, but the required inspection effort varies strongly: ACE confirms all regions in two rounds and nine candidate inspections, whereas the SAM variants need thousands of candidate reviews for their final target locations. Code is available at https://github.com/SagarLekhak/IEEE_WHISPERS_2026_UAV_HSI_PFM1.
