HR-SAR-Net: A Deep Neural Network for Urban Scene Segmentation from High-Resolution SAR Data
This enables automated segmentation for SAR data users, reducing reliance on skilled personnel, though it is incremental as it applies existing DNN methods to new high-resolution SAR data.
The paper tackles urban scene segmentation from high-resolution SAR data by developing a deep neural network, achieving 95.19% pixel accuracy and 74.67% mean IoU on 0.15m/px data over a 2.2km² region.
Synthetic aperture radar (SAR) data is becoming increasingly available to a wide range of users through commercial service providers with resolutions reaching 0.5m/px. Segmenting SAR data still requires skilled personnel, limiting the potential for large-scale use. We show that it is possible to automatically and reliably perform urban scene segmentation from next-gen resolution SAR data (0.15m/px) using deep neural networks (DNNs), achieving a pixel accuracy of 95.19% and a mean IoU of 74.67% with data collected over a region of merely 2.2km${}^2$. The presented DNN is not only effective, but is very small with only 63k parameters and computationally simple enough to achieve a throughput of around 500Mpx/s using a single GPU. We further identify that additional SAR receive antennas and data from multiple flights massively improve the segmentation accuracy. We describe a procedure for generating a high-quality segmentation ground truth from multiple inaccurate building and road annotations, which has been crucial to achieving these segmentation results.