Snow Avalanche Segmentation in SAR Images With Fully Convolutional Neural Networks

Abstract

Knowledge about frequency and location of snow avalanche activity is essential for forecasting and mapping of snow avalanche hazard. Traditional field monitoring of avalanche activity has limitations, especially when surveying large and remote areas. In recent years, avalanche detection in Sentinel-1 radar satellite imagery has been developed to improve monitoring. However, the current state-of-the-art detection algorithms, based on radar signal processing techniques, are still much less accurate than human experts. To reduce this gap, we propose a deep learning architecture for detecting avalanches in Sentinel-1 radar images. We trained a neural network on 6345 manually labeled avalanches from 117 Sentinel-1 images, each one consisting of six channels that include backscatter and topographical information. Then, we tested our trained model on a new synthetic aperture radar image. Comparing to the manual labeling (the gold standard), we achieved an F1 score above 66%, whereas the state-of-the-art detection algorithm sits at an F1 score of only 38%. A visual inspection of the results generated by our deep learning model shows that only small avalanches are undetected, whereas some avalanches that were originally not labeled by the human expert are discovered.

Keywords

• Convolutional neural networks (CNNs)
• deep learning
• saliency segmentation
• Sentinel-1 (S1)
• snow avalanches
• synthetic aperture radar (SAR)