Glacier Calving Front Segmentation Using Attention U-Net
This work provides an incremental improvement for glaciologists and climate scientists by offering a method for more precise and interpretable glacier calving front segmentation.
This paper addresses the problem of segmenting glacier calving fronts from SAR images to determine tidewater glacier status. The authors propose an Attention U-Net model, which performs comparably to a standard U-Net, achieving a 1.5% better Dice score in the best case, with a glacier front line prediction certainty of up to 237.12 meters.
An essential climate variable to determine the tidewater glacier status is the location of the calving front position and the separation of seasonal variability from long-term trends. Previous studies have proposed deep learning-based methods to semi-automatically delineate the calving fronts of tidewater glaciers. They used U-Net to segment the ice and non-ice regions and extracted the calving fronts in a post-processing step. In this work, we show a method to segment the glacier calving fronts from SAR images in an end-to-end fashion using Attention U-Net. The main objective is to investigate the attention mechanism in this application. Adding attention modules to the state-of-the-art U-Net network lets us analyze the learning process by extracting its attention maps. We use these maps as a tool to search for proper hyperparameters and loss functions in order to generate higher qualitative results. Our proposed attention U-Net performs comparably to the standard U-Net while providing additional insight into those regions on which the network learned to focus more. In the best case, the attention U-Net achieves a 1.5% better Dice score compared to the canonical U-Net with a glacier front line prediction certainty of up to 237.12 meters.