Fully Convolutional Network for Automatic Road Extraction from Satellite Imagery

TL;DR

This paper introduces a fully convolutional neural network based on ResNet-34 and U-Net for automatic, accurate, and efficient road extraction from high-resolution satellite imagery, outperforming previous methods.

Contribution

The paper presents a novel combination of ResNet-34 and U-Net architecture for improved road extraction from satellite images, with efficient training and high accuracy.

Findings

Superior results on DEEPGLOBE challenge

Requires moderate GPU memory for training

Fast prediction capability

Abstract

Analysis of high-resolution satellite images has been an important research topic for traffic management, city planning, and road monitoring. One of the problems here is automatic and precise road extraction. From an original image, it is difficult and computationally expensive to extract roads due to presences of other road-like features with straight edges. In this paper, we propose an approach for automatic road extraction based on a fully convolutional neural network of U-net family. This network consists of ResNet-34 pre-trained on ImageNet and decoder adapted from vanilla U-Net. Based on validation results, leaderboard and our own experience this network shows superior results for the DEEPGLOBE - CVPR 2018 road extraction sub-challenge. Moreover, this network uses moderate memory that allows using just one GTX 1080 or 1080ti video cards to perform whole training and makes pretty fast predictions.