Professor Kořistka Award 2026

The Professor Karel Kořistka Award, presented annually by the Faculty of Civil Engineering at CTU Prague for the best master's thesis in Geodesy and Cartography, was awarded for 2026 to Matěj Klimeš. His thesis addressed transfer learning for cloud semantic segmentation in satellite imagery — the challenge of teaching a model to recognise clouds using data from one satellite, then transferring that ability to imagery from an entirely different sensor.

Klimeš built on a U-Net based model, first trained to detect clouds using Sentinel-2 data. He then adapted it to VENuS satellite data using several transfer learning techniques. The results showed that the adapted model achieved performance comparable to, or even better than, a model trained directly on the target data — a finding with practical implications for deploying similar models on new sensors without the need to collect large, sensor-specific training datasets.