Interpreting a Semantic Segmentation Model for Coastline Detection
This work addresses the need for trust and insight in coastal water body extraction models, though it is incremental as it applies an existing interpretation method to a specific domain.
The study tackled the problem of interpreting a semantic segmentation model for coastline detection to identify important spectral bands, finding that the NIR band is most critical with a 38.12 percentage point accuracy drop when permuted, while some bands like Water Vapour showed unexpected importance.
We interpret a deep-learning semantic segmentation model used to classify coastline satellite images into land and water. This is to build trust in the model and gain new insight into the process of coastal water body extraction. Specifically, we seek to understand which spectral bands are important for predicting segmentation masks. This is done using a permutation importance approach. Results show that the NIR is the most important spectral band. Permuting this band lead to a decrease in accuracy of 38.12 percentage points. This is followed by Water Vapour, SWIR 1, and Blue bands with 2.58, 0.78 and 0.19 respectively. Water Vapour is not typically used in water indices and these results suggest it may be useful for water body extraction. Permuting, the Coastal Aerosol, Green, Red, RE1, RE2, RE3, RE4, and SWIR 2 bands did not decrease accuracy. This suggests they could be excluded from future model builds reducing complexity and computational requirements.