A partnership between H2i Balkans (Sarajevo, Bosnia-Herzegovina) and AQUA PROCON (Brno, Czech Republic) has won support of the UNDP Challenge Fund for the project on quantifying real time rainfall intensity based on computer vision and machine learning. This grant is a part of the Czech-UNDP Partnership for Sustainable Development Goals (CFCS) project, which is aimed to facilitate transfer of technology and solutions that will help address the region’s development challenges.

The project “Quantifying real time rainfall intensity based on computer vision and machine learning” focuses on development of a technology enabling creation of supplementary network of rainfall-observing instruments. The technology will process video feeds from CCTV cameras or smartphones to infer rainfall intensity using machine learning, big data and computer vision.

According to the Federal Hydro Meteorological Institute, there were more than 500 rainfall stations before year 1992, while now there are less than 100 which are operational. Thus there is an urgent need to explore solutions to increase the spatial density of monitored rainfall.

The results of the project will be presented on a web-portal which will enable the authorities to be better prepared for meteorologically and hydrologically based hazards.

Project duration is 12 months and it will be implemented in Sarajevo, Bosnia and Herzegovina.