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