Extraction And Spatial Pattern Analysis Of Changsha Built-up Area Based On Multi-source Night-light Remote Sensing Data

Rapid urbanization inevitably brings about a large expansion of urban built-up areas,which also brings about a large number of problems in terms of population,urban resources and urban environment,hindering the development of cities.The extraction and direction control of urban built-up areas can help to familiarize with the internal spatial structure of cities and understand the development process of cities,which can help to reduce the negative effects brought by rapid urbanization expansion and provide a basis for the real-time development of urbanization.In this paper,we use 2013 DMSP/OLS,2015 and 2018 NPP/VIIRS,2019 Luojia1-01 night light remote sensing data combined with landsat8 OLI remote sensing data as the study data,and take Changsha city as the study area,and perform noise reduction processing,built-up area extraction,and urban spatial pattern analysis on four night light remote sensing data.The results of the study are as follows.(1)The dual-threshold combined with the mask method has the best noise reduction effect on the night-light remote sensing data.Three methods,the dual-threshold method,the mask method,and the dual-threshold combined with mask method,were used to conduct accuracy comparison tests using monthly synthetic data and annual standard data for December of 2015 and 2016.The study shows that the dual-threshold combined with mask method has the best noise reduction effect and the total amount of lights is closest to the annual standard data.The noise reduction process for the 2018 NPP/VIIRS annual synthetic data and the 2019 Luojia1-01 night light remote sensing data using the double threshold and mask method can obtain the remote sensing image of night lights with the least noise.(2)The extraction accuracy of urban built-up areas combining multi-source data night light remote sensing images will be improved with the increase of spatial resolution of night light remote sensing data,and the extraction accuracy of Luojia1-01 night light remote sensing data has the best effect.Mutation detection method,statistical data comparison method,and SVM supervised classification method combining multi-source data are used to evaluate the accuracy by establishing confusion matrix.The results show that Luojia1-01 night-light remote sensing data is not suitable for the traditional threshold method for urban built-up area extraction,and the kappa coefficient is 0.743.The SVM supervised classification method combined with multi-source data can fully reflect the contour of urban built-up area,and the kappa coefficient is 0.856.With the improvement of the quality of night-light remote sensing data,the kappa coefficient is gradually improved,and the accuracy is also gradually improved.In terms of the accuracy of night light remote sensing image built-up area extraction combining multiple sources data,Luojia1-01 data is smaller than DMSP/OLS and NPP/VIIRS data in terms of light spillover effect and has the best effect.(3)The four aspects of urban lighting scale,urban development trajectory,urban landscape pattern analysis and urban spatial pattern change drivers in Changsha City are analyzed.The urban lighting scale is analyzed by four indicators: total lighting,nighttime lighting average,nighttime lighting increment,and nighttime lighting increase rate,which show that the urban lighting developed rapidly during 2013-2019,and the urban built-up area expanded rapidly;the urban development trajectory is analyzed by standard ellipse difference,and the center of gravity of the city shifts to the southeast,and the expansion of the urban built-up area is faster in the north-south direction than in the east-west direction;the urban The analysis of the landscape pattern shows that Changsha City focuses on internal spatial optimization,and the analysis of the driving force of the change of urban spatial pattern shows that Changsha City is influenced by natural geography,social economy and urban traffic,and the city develops in the north-south direction.Changsha City,Xiangtan and Zhuzhou create an integrated urban cluster to drive the economic development of the central and southern regions.