Deep Tiered Image Segmentation For Detecting Internal Ice Layers in Radar Imagery
This work addresses the slow and laborious manual labeling of internal ice layers in radar data, which is important for understanding polar ice structure and climate modeling, representing a domain-specific advancement.
The paper tackles the problem of automatically detecting internal ice layers in polar ice radar imagery, which is challenging due to varying numbers of layers and merging/splitting boundaries, by proposing a novel deep neural network for tiered segmentation and applying it to a large-scale dataset with human-labeled ground truth, achieving evaluation on this dataset.
Understanding the structure of Earth's polar ice sheets is important for modeling how global warming will impact polar ice and, in turn, the Earth's climate. Ground-penetrating radar is able to collect observations of the internal structure of snow and ice, but the process of manually labeling these observations is slow and laborious. Recent work has developed automatic techniques for finding the boundaries between the ice and the bedrock, but finding internal layers - the subtle boundaries that indicate where one year's ice accumulation ended and the next began - is much more challenging because the number of layers varies and the boundaries often merge and split. In this paper, we propose a novel deep neural network for solving a general class of tiered segmentation problems. We then apply it to detecting internal layers in polar ice, evaluating on a large-scale dataset of polar ice radar data with human-labeled annotations as ground truth.