Safe Pro AI Completes United Nations Development Programme’s Purchase Order for Artificial Intelligence Minefield Analysis in Ukraine

In This Article:

SpotlightAI?, Powered by AWS, Selected by UNDP for Hyperscale AI Analysis of Drone Imagery Addressing the Unprecedented Scope of the Landmine Crisis in Ukraine

AVENTURA, FL / ACCESSWIRE / September 5, 2024 / Safe Pro Group Inc. (Nasdaq:SPAI) ("Safe Pro" or the "Company"), a premier provider of artificial intelligence (AI) solutions specializing in drone imagery processing for humanitarian mine-clearing efforts, today announced that its Safe Pro AI subsidiary has completed a Phase III drone survey operation in Ukraine under a purchase order (PO) awarded by the United Nations Development Programme (UNDP). The Company analyzed land in Kharkiv Oblast, Ukraine, and provided the UNDP with detailed reports and survey data of items of interest as detected by its artificial intelligence (AI) and machine learning (ML) models.

Phase III is the latest milestone in a multi-year/multi-phase technology evaluation program being undertaken by the UNDP, and the activity in Kharkiv Oblast was conducted in partnership with the World Food Programme (WFP) and the Food and Agriculture Organization (FAO). Conducted in collaboration with Ukraine's Ministry of the Economy (MoE), and supported by Croatia, France, Malta, the Netherlands and Sweden, this Mine Action Project was designed to showcase innovative tools and methodologies utilizing Unmanned Aerial Vehicles / Unmanned Ground Vehicles (UAV/UGV) to enhance land release efficiency in agricultural areas.

The latest Phase III operation is the result of over a year of testing under UNDP's technology evaluation program including:

Powered by Amazon Web Services (AWS), SpotlightAI? is the Company's hyper scalable, cloud-based ecosystem of tools for the analysis of imagery collected by Commercial-off-the-Shelf (COTS) drones. It utilizes advanced machine learning models for small object detection and data reporting built on an extensive proprietary dataset capable of identifying 150+ types of land mines, unexploded ordnance (UXO) and explosive remnants of ware (ERW). It can rapidly identify, GPS-locate, and produce high-resolution orthomosaic maps of detected landmines and UXO from drone-based imagery, significantly reducing the time and costs required in human-based image analysis. This rapid, highly-scalable analysis capability is critically needed in Ukraine where an area estimated to exceed 139,300 km2 (Source: Ministry of Economy of Ukraine as of August 2024) (roughly equivalent to the area of the State of North Carolina) is potentially contaminated by UXO.