Updating Road Maps at City Scale With Remote Sensed Images and Existing Vector Maps

Currently, many countries have built geoinformation databases and gathered large amounts of geographic data. However, with the extensive construction of infrastructure and rapid expansion of cities, the road updating process is imperative to maintain the high quality of current basic geographic information. Currently, road extraction and change detection are two commonly used methods to solve road updating problems. Most of the existing methods rely on a large number of accurate road labels to generate road information while ignoring the use of quantities of available but incomplete road maps. In our work, we proposed a semisupervised road extraction method specifically for road-updating applications [semi-supervised road updating network (SRUNet)]. In this approach, historical road maps are fused with the latest remote sensing images, and the state of the roads is updated directly. A multibranch network is the core of the method, which consists of three noteworthy parts: the map encoding branch (MEB) proposed for representation learning, the boundary enhancement module (BEM) for improving the accuracy of boundary prediction, and the residual refinement module (RRM) for further optimizing the prediction results. We applied our method to two datasets: the DeepGlobe public dataset and our self-constructed dataset from Zhengzhou and Nanjing. Experimental results show that our method achieves an improvement of 14.37% over the baseline approach. Notably, the addition of historical maps improved the model's performance by 12.4%. Promising results were obtained on the two cities' large-scale road networks. With reliable prediction results and improved performance, we believe that SRUNet is meaningful for a wide range of road renewal applications.