UNIT.City — місце, де люди працюють... КРАЩЕ! Обирай свій простір просто зараз 👉
Валентин ШнайдерAI Eng
30 July 2025, 09:38
2025-07-30
IT teams in Kyiv created AI models to detect mines in drone photos
In Kyiv, over 50 Ukrainian IT professionals joined forces at the AI Data Jam hackathon to train artificial intelligence to recognize explosives in aerial photographs. The initiative was a continuation of the spring AI Labeling Sprint and aims to accelerate humanitarian demining of Ukraine’s territories.
In Kyiv, over 50 Ukrainian IT professionals joined forces at the AI Data Jam hackathon to train artificial intelligence to recognize explosives in aerial photographs. The initiative was a continuation of the spring AI Labeling Sprint and aims to accelerate humanitarian demining of Ukraine’s territories.
According to the Ministry of Economy, Environment and Agriculture, 13 teams took part in the event. Over the course of two days, they worked with over 30,000 images (8,000 training and 23,000 test), most of which had been annotated during a previous hackathon. The participants had to create AI models that could accurately recognize mines and other ERW in drone images.
The teams worked both offline in Unit.City and remotely. Mentoring support was provided by experts from UADamage, The HALO Trust and Dropla Tech. The most successful team was MineWatch AI, which received a certificate for $2000 from UNDP and the Government of Luxembourg. The winners also received an invitation to continue developing the solution together with demining experts.
«This is no longer an experiment. We are creating a unique tool that will make Ukraine safer, meter by meter, without risking life,» emphasized Deputy Minister of Economy Ihor Bezkaravayny.
Benjamin Lark, Head of UNDP’s Mine Action Programme in Ukraine, also noted the importance of the event. According to him, the hackathon brought together representatives of the state, technology businesses, civil society and international partners around a real problem.
All models and data created within the initiative will form the basis for further improvement of the UXO recognition systems. After the testing phase, the technologies will begin to be implemented in real humanitarian demining processes. They can be applied not only in Ukraine, but also in regions affected by the consequences of war around the world.
We previously wrote about how Dmytro Titov, CEO of the Ukrainian startup Ailand System, which produces UAVs for mine clearance, spoke about plans to export drones when the relevant restrictions are lifted.
Kyiv schoolchildren created a useful AR application for teaching first aid. How Firselp, which won gold at the Infomatrix 2025 international hackathon, works