Problem
Research questions and friction points this paper is trying to address.
Automatic Recognition
Snow Thickness Measurement
Radar Imagery
Skip-WaveNet: a wavelet based multi-scale architecture to trace snow layers in radar echograms
🤖 AI Summary
To address the low accuracy and poor generalizability of automatic snow layer thickness estimation in polar ice-sheet radar imagery, this paper proposes a novel WaveNet variant integrating wavelet-based multi-scale analysis and skip connections. Methodologically, it innovatively combines wavelet feature decomposition, multi-scale dilated convolutions, and end-to-end joint training—simultaneously optimizing deep regression for thickness prediction and semantic segmentation for boundary delineation—thereby enhancing robustness to cross-regional snow layer boundaries and improving physical interpretability. Evaluated on real airborne radar data, the model achieves a mean absolute error of 3.31 pixels and an average segmentation accuracy of 94.3%. Crucially, estimated snow thicknesses exhibit strong agreement with in-situ ice-core measurements (R² > 0.96), substantially outperforming existing approaches. This work provides a reliable remote sensing foundation for high-precision sea-level rise modeling.
📝 Abstract
Airborne radar sensors capture the profile of snow layers present on top of an ice sheet. Accurate tracking of these layers is essential to calculate their thicknesses, which are required to investigate the contribution of polar ice cap melt to sea-level rise. However, automatically processing the radar echograms to detect the underlying snow layers is a challenging problem. In our work, we develop wavelet-based multi-scale deep learning architectures for these radar echograms to improve snow layer detection. These architectures estimate the layer depths with a mean absolute error of 3.31 pixels and 94.3% average precision, achieving higher generalizability as compared to state-of-the-art snow layer detection networks. These depth estimates also agree well with physically drilled stake measurements. Such robust architectures can be used on echograms from future missions to efficiently trace snow layers, estimate their individual thicknesses, and thus support sea-level rise projection models.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Wavelet Method
Deep Learning
Ice Sheet Snow Depth Estimation
🔎 Similar Papers
No similar papers found.
