Thirteen teams of IT specialists and engineers worked on creating artificial intelligence models capable of automatically detecting explosive devices (ED) in drone images.
The event took place at the innovation park Unit.City and was a continuation of the AI Labeling Sprint held earlier on May 5 in Kyiv. The goal of such hackathons is to introduce innovative technologies into demining to speed up the clearance of Ukrainian territories contaminated with mines and explosive devices.
Using over 8,000 training and 23,000 testing images—which included aerial photos annotated by participants of the first hackathon—the participants of the second stage, AI Data Jam, spent two days developing highly efficient AI algorithms for automatic recognition of mines and other explosive devices.
Participants worked both offline and remotely, receiving mentorship from experts from organizations such as UADamage, The HALO Trust, and Dropla Tech.
The winner was the MineWatch AI team, which trained the best AI model. The team members received a $2,000 certificate from UNDP Ukraine and the Luxembourg government. They also gained the opportunity to further develop their product in collaboration with specialists actively involved in demining.
“What started as an experiment has grown into a full-fledged project with international support and real results. Thanks to everyone who found the time and opportunity to invest in this important cause. The motivation is huge: together we aim to create a unique product that will help clear Ukraine of mine debris, making it safer meter by meter—without risking lives and with maximum use of technology,” emphasized Igor Bezkaravaynyi, Deputy Minister of Economy, Environment, and Agriculture of Ukraine.
“Ukraine today is demonstrating an impressive level of innovation in humanitarian demining. What we see is the future being created right now. Our goal is to make innovations accessible to all who work in this field. For UNDP, it is important that this hackathon unites all key players—the government, veterans, tech teams, civil society, and international partners,” said Benjamin Lark, Head of UNDP’s Mine Action Programme in Ukraine.
The data collected during the hackathons and the AI models developed will form the basis for further training and refinement of artificial intelligence algorithms. After testing, these technologies will gradually be integrated into humanitarian demining processes. This will significantly accelerate the clearance of Ukrainian lands from mines and other explosive devices, making demining more effective and safer. The solutions developed have potential for application not only in Ukraine but also internationally.
The hackathon was organized by the Ministry of Economy, Environment and Agriculture of Ukraine, the Demine Ukraine initiative, the Center for Humanitarian Demining, and the Ukrainian company UADamage, with support from the United Nations Development Programme (UNDP) in Ukraine and funding from the Luxembourg government.
