Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Brightearth roads: Towards fully automatic road network extraction from satellite imagery

Liuyun Duan, Willard Mapurisa, Maxime Leras, Leigh Lotter, Yuliya Tarabalka

TL;DR

This paper proposes a fully automated pipeline for extracting road networks from very-high-resolution (VHR) satellite imagery that directly generates road line-strings that are seamlessly connected and precisely positioned.

Abstract

The modern road network topology comprises intricately designed structures that introduce complexity when automatically reconstructing road networks. While open resources like OpenStreetMap (OSM) offer road networks with well-defined topology, they may not always be up to date worldwide. In this paper, we propose a fully automated pipeline for extracting road networks from very-high-resolution (VHR) satellite imagery. Our approach directly generates road line-strings that are seamlessly connected and precisely positioned. The process involves three key modules: a CNN-based neural network for road segmentation, a graph optimization algorithm to convert road predictions into vector line-strings, and a machine learning model for classifying road materials. Compared to OSM data, our results demonstrate significant potential for providing the latest road layouts and precise positions of road segments.

Brightearth roads: Towards fully automatic road network extraction from satellite imagery

TL;DR

This paper proposes a fully automated pipeline for extracting road networks from very-high-resolution (VHR) satellite imagery that directly generates road line-strings that are seamlessly connected and precisely positioned.

Abstract

The modern road network topology comprises intricately designed structures that introduce complexity when automatically reconstructing road networks. While open resources like OpenStreetMap (OSM) offer road networks with well-defined topology, they may not always be up to date worldwide. In this paper, we propose a fully automated pipeline for extracting road networks from very-high-resolution (VHR) satellite imagery. Our approach directly generates road line-strings that are seamlessly connected and precisely positioned. The process involves three key modules: a CNN-based neural network for road segmentation, a graph optimization algorithm to convert road predictions into vector line-strings, and a machine learning model for classifying road materials. Compared to OSM data, our results demonstrate significant potential for providing the latest road layouts and precise positions of road segments.
Paper Structure (16 sections, 4 figures, 1 table)

This paper contains 16 sections, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Road Segmentation
  3. Data Preparation
  4. Neural Network Design
  5. Training
  6. Road Network Reconstruction
  7. Road Skeletonization
  8. Linear Roads Recovery
  9. De-Noising
  10. Road Smoothing
  11. Traffic Circle Reconstruction
  12. Road material Classification
  13. Results and Discussion
  14. Quantitative Statistics
  15. Qualitative Results
  16. ...and 1 more sections

Figures (4)

  • Figure 1: The proposed road extraction pipeline from a single input satellite image.
  • Figure 2: The proposed new architecture for road segmentation from a single satellite image.
  • Figure 3: Road materials classification examples. Green for processed roads and red for unprocessed roads (a), pink for gravel roads (b) and orange for sand roads (c).
  • Figure 4: Road networks generated by our pipeline over Timbuktu, Amman and Aden (top to bottom rows), with input satellite images, road segmentation, our extracted road networks, the Ground Truth, and crops (left to right columns). In the crops, Ground Truth are marked in red and our results in yellow.