Next Article in Journal
Modeling Spatio-Temporal Evolution of Urban Crowd Flows
Previous Article in Journal
Quality Control of “As Built” BIM Datasets Using the ISO 19157 Framework and a Multiple Hypothesis Testing Method Based on Proportions
Open AccessArticle

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

by Yan Xie 1,2, Fang Miao 3, Kai Zhou 4,5,* and Jing Peng 5
1
College of Geophysics, Chengdu University of Technology, Chengdu 610059, China
2
Geological Team 103, Guizhou Bureau of Geology Mineral Exploration Development, Tongren 554300, China
3
Big Data Research Institute, Chengdu University, Chengdu 610106, China
4
College of Computer science, Sichuan University, Chengdu 610065, China
5
Science and Technology Information Department, Sichuan Provincial Department of Public Security, Chengdu 610041, China
*
Author to whom correspondence should be addressed.
ISPRS Int. J. Geo-Inf. 2019, 8(12), 571; https://doi.org/10.3390/ijgi8120571
Received: 24 October 2019 / Revised: 30 November 2019 / Accepted: 9 December 2019 / Published: 10 December 2019
Road extraction is a unique and difficult problem in the field of semantic segmentation because roads have attributes such as slenderness, long span, complexity, and topological connectivity, etc. Therefore, we propose a novel road extraction network, abbreviated HsgNet, based on high-order spatial information global perception network using bilinear pooling. HsgNet, taking the efficient LinkNet as its basic architecture, embeds a Middle Block between the Encoder and Decoder. The Middle Block learns to preserve global-context semantic information, long-distance spatial information and relationships, and different feature channels’ information and dependencies. It is different from other road segmentation methods which lose spatial information, such as those using dilated convolution and multiscale feature fusion to record local-context semantic information. The Middle Block consists of three important steps: (1) forming a feature resource pool to gather high-order global spatial information; (2) selecting a feature weight distribution, enabling each pixel position to obtain complementary features according to its own needs; and (3) inversely mapping the intermediate output feature encoding to the size of the input image by expanding the number of channels of the intermediate output feature. We compared multiple road extraction methods on two open datasets, SpaceNet and DeepGlobe. The results show that compared to the efficient road extraction model D-LinkNet, our model has fewer parameters and better performance: we achieved higher mean intersection over union (71.1%), and the model parameters were reduced in number by about 1/4. View Full-Text
Keywords: road extraction; remote sensing; CNN; high-order spatial information; global feature road extraction; remote sensing; CNN; high-order spatial information; global feature
Show Figures

Figure 1

MDPI and ACS Style

Xie, Y.; Miao, F.; Zhou, K.; Peng, J. HsgNet: A Road Extraction Network Based on Global Perception of High-Order Spatial Information. ISPRS Int. J. Geo-Inf. 2019, 8, 571.

Show more citation formats Show less citations formats
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop