arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1912.04441cs.CVeess.IV
收藏

HR-SAR-Net:一种用于高分辨率SAR数据城市场景分割的深度神经网络

HR-SAR-Net: A Deep Neural Network for Urban Scene Segmentation from High-Resolution SAR Data

ETH Zürich(苏黎世联邦理工学院);University of Bologna(博洛尼亚大学)

Xiaying Wang, Lukas Cavigelli, Manuel Eggimann, Michele Magno, Luca Benini

摘要:

越来越多的用户可以通过商业服务提供商获得分辨率达到0.5米/像素的合成孔径雷达数据。对合成孔径雷达数据进行分段仍然需要熟练的人员,这限制了大规模使用的潜力。我们表明,使用深度神经网络(DNN)可以从下一代分辨率SAR数据(0.15 m/px)自动可靠地执行城市场景分割,实现95.19%的像素准确度和74.67%的平均IoU,数据收集区域仅为2.2 km ${}^2$。所提出的DNN不仅有效,而且非常小,只有63 k个参数,计算简单到足以使用单个GPU实现约500 Mpx/s的吞吐量。我们进一步确定,额外的SAR接收天线和多个航班的数据大大提高了分割精度。我们描述了一个程序,用于从多个不准确的建筑物和道路注释生成高质量的分割地面实况,这对实现这些分割结果至关重要。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑