A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Quick detecting of urban built-up areas using the neural network model based on the nighttime light data
The accurate and quick detection of urban built-up regions is crucial for better city management and planning, as well as for cognition of urbanization. Nighttime light images and their features provide a reliable means to quantify urban regions. However, only a limited number of studies have delved into the specifics of how light variations across diverse land use types. Therefore, this study made a new attempt using the neural network model method based on VIIRS nighttime light data to detect the spatiotemporal evolution of urban built-up regions in Zhejiang province spanning from 2012 to 2021. The results revealed that the neural network model-based method used in this study had noticeable performance in identifying urban built-up areas, achieving an overall precision of 93.27% and a kappa coefficient of 0.83. Furthermore, there was a significant increase in urban built-up areas and a slight decrease in rural built-up areas in Zhejiang, with more than 35% of built-up regions converting from non-urban regions. The detailed features of urban built-up region dynamics were also detected using a city-level analysis. Moreover, this study extracted urban built-up regions based on the light variation characteristics of land use itself through the neural network method, which provides a new insight into the applications of nighttime light data in urban studies.
Quick detecting of urban built-up areas using the neural network model based on the nighttime light data
The accurate and quick detection of urban built-up regions is crucial for better city management and planning, as well as for cognition of urbanization. Nighttime light images and their features provide a reliable means to quantify urban regions. However, only a limited number of studies have delved into the specifics of how light variations across diverse land use types. Therefore, this study made a new attempt using the neural network model method based on VIIRS nighttime light data to detect the spatiotemporal evolution of urban built-up regions in Zhejiang province spanning from 2012 to 2021. The results revealed that the neural network model-based method used in this study had noticeable performance in identifying urban built-up areas, achieving an overall precision of 93.27% and a kappa coefficient of 0.83. Furthermore, there was a significant increase in urban built-up areas and a slight decrease in rural built-up areas in Zhejiang, with more than 35% of built-up regions converting from non-urban regions. The detailed features of urban built-up region dynamics were also detected using a city-level analysis. Moreover, this study extracted urban built-up regions based on the light variation characteristics of land use itself through the neural network method, which provides a new insight into the applications of nighttime light data in urban studies.
Quick detecting of urban built-up areas using the neural network model based on the nighttime light data
Zou, Bin (editor) / Cui, Yaoping (editor) / Xu, Pengfei (author) / Chen, Xinyi (author) / Jin, Pingbin (author) / Zhao, Qiuhao (author) / Zhou, Guangyao (author)
International Conference on Remote Sensing, Mapping, and Geographic Information Systems (RSMG 2024) ; 2024 ; Zhengzhou, China
Proc. SPIE ; 13402
2024-11-27
Conference paper
Electronic Resource
English
Wider urban zones: use of topology and nighttime satellite images for delimiting urban areas
| BASE | 2022