Uncertainty evaluation of segmentation models for Earth observation
It addresses the practical need for reliable uncertainty measures in Earth observation applications, though it is incremental as it benchmarks existing methods rather than proposing new ones.
This paper benchmarks existing uncertainty estimation methods for semantic segmentation of satellite imagery, evaluating their ability to identify prediction errors and noise-corrupted regions on two remote sensing datasets (PASTIS and ForTy).
This paper investigates methods for estimating uncertainty in semantic segmentation predictions derived from satellite imagery. Estimating uncertainty for segmentation presents unique challenges compared to standard image classification, requiring scalable methods producing per-pixel estimates. While most research on this topic has focused on scene understanding or medical imaging, this work benchmarks existing methods specifically for remote sensing and Earth observation applications. Our evaluation focuses on the practical utility of uncertainty measures, testing their ability to identify prediction errors and noise-corrupted input image regions. Experiments are conducted on two remote sensing datasets, PASTIS and ForTy, selected for their differences in scale, geographic coverage, and label confidence. We perform an extensive evaluation featuring several models, such as Stochastic Segmentation Networks and ensembles, in combination with a number of neural architectures and uncertainty metrics. We make a number of practical recommendations based on our findings.