Road Segmentation in SAR Satellite Images with Deep Fully-Convolutional Neural Networks

Abstract

Remote sensing is extensively used in cartography. As transportation networks grow and change, extracting roads automatically from satellite images is crucial to keep maps up-to-date. Synthetic Aperture Radar satellites can provide high resolution topographical maps. However roads are difficult to identify in these data as they look visually similar to targets such as rivers and railways. Most road extraction methods on Synthetic Aperture Radar images still rely on a prior segmentation performed by classical computer vision algorithms. Few works study the potential of deep learning techniques, despite their successful applications to optical imagery. This letter presents an evaluation of Fully-Convolutional Neural Networks for road segmentation in SAR images. We study the relative performance of early and state-of-the-art networks after carefully enhancing their sensitivity towards thin objects by adding spatial tolerance rules. Our models shows promising results, successfully extracting most of the roads in our test dataset. This shows that, although Fully-Convolutional Neural Networks natively lack efficiency for road segmentation, they are capable of good results if properly tuned. As the segmentation quality does not scale well with the increasing depth of the networks, the design of specialized architectures for roads extraction should yield better performances.

arXiv comment5 pages, accepted for publication in IEEE Geoscience and Remote Sensing Letters

References in corpus (4)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Road Extraction by Deep Residual U-Net
• Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials
• Fully Convolutional Networks for Semantic Segmentation

Cited by in corpus (11)

• Road Segmentation for Remote Sensing Images using Adversarial Spatial Pyramid Networks
• Automatic extraction of road intersection points from USGS historical map series using deep convolutional neural networks
• Robust Iris Segmentation Based on Fully Convolutional Networks and Generative Adversarial Networks
• Fully Convolutional Networks and Generative Adversarial Networks Applied to Sclera Segmentation
• HR-SAR-Net: A Deep Neural Network for Urban Scene Segmentation from High-Resolution SAR Data
• Image-Conditioned Graph Generation for Road Network Extraction
• SemiCurv: Semi-Supervised Curvilinear Structure Segmentation
• Physics-Guided Generative Adversarial Networks for Sea Subsurface Temperature Prediction
• Classifying Bicycle Infrastructure Using On-Bike Street-Level Images
• Weakly-supervised land classification for coastal zone based on deep convolutional neural networks by incorporating dual-polarimetric characteristics into training dataset
• Self-Labeling of Fully Mediating Representations by Graph Alignment