Synthesizing location semantics from street view images to improve urban land-use classification

ABSTRACT

Land-use maps are instrumental to inform urban planning and environmental research. Street view images (SVIs) have shown great potential for automated land-use classification for land-use mapping. However, previous studies overlooked SVI-derived location contextual information that may help improve land-use classification. This study proposes a novel land-use classification method that synthesizes location semantics from SVIs to account for contextual information from SVIs, land parcels and roads around the SVIs. The proposed method first generates land-use scene images (LUSIs) by using an SVI-derived straightforward algorithm. The LUSIs are then relocated to land parcels by using a displacement strategy and classified into land-use types by using a deep learning network. This study determines the land-use types of land parcels with classified LUSIs. Two case studies, consisting of LUSIs for five land-use types, show that introducing location semantics of SVIs can remarkably improve the classification accuracy of land-use types.

KEYWORDS:

• Land-use map
• street view image
• location semantics
• land-use classification

Acknowledgments

We would like to thank Kaishun Wu, Yanling Chen and Wenze Kan for assistant with the experiments. We also sincerely acknowledge Dr. May Yuan, Dr. Christophe Claramunt and the anonymous referees for their insightful comments.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data and codes availability statement

The data and code that support the findings of this study are available in the Figshare repository with the identifier(s) at the link: https://doi.org/10.6084/m9.figshare.12444266.

Additional information

Funding

This research was supported by the National Earth Observation Data Center [NODAOP2020015], the National Natural Science Foundation of China [41801378, 42071382] and the Open Fund of Key Laboratory of Urban Land Resources Monitoring and Simulation, Ministry of Natural Resources [KF-2019-04-033].

Notes on contributors

Fang Fang

Fang Fang, received a B.S. degree in Computer Science and Technology in 1998 and a Ph.D. degree in Management Science and Engineering in 2012 from the China Uni- versity of Geosciences, Wuhan, China. She is currently an Associate Professor at China University of Geosciences. Her research interests include spatial data mining, machine learning, and GIS applications. She is a member of CCF and IEEE.

Yafang Yu

Yafang Yu, received a B.S. degree in Information and Computing Science in 2017 and a MA.Eng. degree in Software engineering from the China University of Geosciences (Wuhan), in 2020. She is currently a deep learning algorithm engineer. Her main interests include deep learning, remote sensing, and GIS applications.

Shengwen Li

Shengwen Li, received the B.Sc. degrees in computer science and technology from China University of Geosciences (Wuhan), in 2000, and the Ph.D. degree in Cartography and Geographic information Engineering from China University of Geosciences (Wuhan), in 2010. He is currently an associate professor in School of Geography and Information Engineering, China University of Geosciences (Wuhan). His main interests include spatial statistics, geospatial Artificial Intelligence, natural language processing.

Zejun Zuo

Zejun Zuo, received a B.S. degree in Geographic Information System in 2002 and a Ph.D. degree in Cartography and Geographic Information Engineering in 2012 from the China University of Geosciences, Wuhan, China. He is currently a lecturer in China University of Geosciences (Wuhan). His research interests include spatial data modeling, application of 3D modeling and visualization analysis.

Yuanyuan Liu

Yuanyuan Liu, received B.E. degree from NanChang University, NanChang, China, in 2005, M.E. degree from Huazhong University of Science and Technology, Wuhan, China, in 2007, and Ph.D. degree from Central China Normal University. She is currently an Associate Professor in China University of Geosciences (Wuhan). Her research interests include image processing, computer vision and pattern recognition.

Bo Wan

Bo Wan, received a B.S. degree in Computer Science and Technology in 1998 and a Ph.D. degree in Cartography and Geographic information Engineering in 2007 from the China University of Geosciences, Wuhan, China. He is currently a Professor at China University of Geosciences. His research interests include GIS platform development, GIS & RS applications, and 3D geology modeling.

Zhongwen Luo

Zhongwen Luo, received the B.S. degrees in Mathematics from Wuhan University, China, in 1985, and the M.E. degree in Mechanics from Institute of Mechanics, Academia Sinica, China, in 2009. He is currently a Professor at China University of Geosciences. His research interests include AI, robotics, High Performance Computation and network.