The detection and identification of explosive ordnance objects (both surface and sub-surface) is going through a period of technological innovation, driven by both the great need and the arrival of new types of field-deployable sensors that can be integrated into drones and hand-held devices.

Thanks to support from the European Union and partnerships with emergency mine action and state EOD actors, Tech 4 Tracing and our partner VFRAME are leading the use of computer vision to create detection and identification algorithms for a range of surface objects, including cluster sub-munitions, scatterable mines, grenades, mortars and other unexploded ordnance.

Our workflow uses 3D photogrammetry, photorealistic 3D models and custom-made 3D-printed replicas to generate synthetic data to train AI detection algorithms for objects in a range of conditions and landscapes.

Two field missions to Ukraine in 2023 provided critical benchmark data and feedback from mine action actors about the deployment modalities of these detectors. Field-testing of the first set of detector algorithms, including for the PFM-1, PTM-1, 9N235/N210, will begin in January 2024 in Ukraine.