Road Extraction from High Resolution Image with Deep Convolution Network—A Case Study of GF-2 Image

Abstract

Recently, with the development of remote sensing and computer techniques, automatic and accurate road extraction is becoming feasible for practical usage. Nowadays, accurate extraction of road information from satellite data has become one of the most popular topics in both remote sensing and transportation fields. It is very useful for applying this technique to fast map updating, construction supervision, and so on. However, as there is usually a huge volume of information provided by remote sensing data, an efficient method to refine the big volume of data is important in corresponding applications. We apply deep convolution network to perform an image segmentation approach, as a solution for extracting road networks from high resolution images. In order to take advantage of deep learning, we study the methods of generating representative training and testing datasets, and develop semi-supervised leaning skills to enhance the data scale. The extraction of the satellite images that are affected by color distortion is also studied, in order to make the method more robust for more applicational fields. The GF-2 satellite data is used for experiments, as its images may show optical distortion in small pieces. Experiments in this paper showed that, the proposed solution successfully identifies road networks from complex situations with a total accuracy of more than 80% in discriminable areas.

HsgNet: A Road Extraction Network Based on Global Perception of High-Order Spatial Information

“…RoadTracer [21], proposed by Bastani, uses an iterative search process based on CNN decision function to output the network directly from CNN. Xia et al [22] also directly used deep convolutional neural network (DCNN) to extract roads and tested it on GaoFen-2 satellite (GF-2) images. Some scholars have considered using road topological features to improve the accuracy of road extraction [23] and initially attempted to generate topologically connected road networks using constrained models.…”

An Ensemble Wasserstein Generative Adversarial Network Method for Road Extraction From High Resolution Remote Sensing Images in Rural Areas

“…However, despite numerous studies, road extraction from HR-RS images is still far from large-scale practical application. The spectral heterogeneity in HR-RS images [6,13] (e.g. large intra-class or small inter-class spectral variability) and the presence of occlusions (e.g.…”

“…Nevertheless, although numerous methods have been proposed over the decades, road extraction from HR-RS images is still far from large-scale practical application. The spectral heterogeneity in HR-RS images [6,13] and the occlusions by trees, shadows and other non-road objects on roads are the key factors that make the segmented road networks messy and unclean [14,15]. A more general and robust road segmentation method is needed to improve the performance of road extraction from HR-RS images.…”

Towards Automatic Extraction and Updating of VGI-Based Road Networks Using Deep Learning

“…Recently, Bastani et al [28] used an iterative search process guided by a CNN-based decision function to derive the road network graph directly from the output of the CNN. [29] employs post-processing to make the extracted road region more realistic, several morphological algorithms are typically deployed to fill the holes, smooth the edges, connect the road segment, and then achieve the coarse center lines of road segments.…”