RSLab at the Top in the 2020 Gaofen Challenge
A team led by Ding Lei (PhD student at RSLab) won the 2nd place in the ‘2020 Gaofen Challenge on Automated High-Resolution Earth Observation Image Interpretation‘ in the subject ‘Automatic Classification of PolSAR Images’.
This competition was held by the China High-Resolution Earth Observation Conference Committee. The contest data have acquired based on the platform of Gaofen (High-Resolution) Earth Observation Project of China. The contest was technically co-sponsored by the Gaofen Project Office, the Aerospace Information Research Institute, Chinese Academy of Sciences, the IEEE Geoscience & Remote Sensing Society (GRSS) and the International Society for Photogrammetry and Remote Sensing (ISPRS).
A total of 701 teams from 253 affiliations with 2023 competitors worldwide have joined the competition. The solution that won the second place is based on a novel CNN based method (MP-ResNet: Multi-path Residual Network for the Semantic Segmentation of PolSAR image).