Spatial And Temporal Development Model Of Capital Cities In Northeast China Based On Multi-source Remote Sensing Image Data

The law of urban space-time development refers to the direct reflection of the landscape structure and its change law in the process of urbanization.The law of time and space develop-ment can provide a reliable basis for urban managers to regulate the relationship between hu-man activities and urbanization.In this process,it is of great significance to accurately delineate the scope of urban built-up areas for analyzing the law of urban space-time development.In this study,the Northeast provincial capital city is taken as the research area,and two methods of urban built-up area extraction,neighborhood extreme value method and boundary pixel fu-sion method,are proposed,which are applied to analyze the spatial and temporal development law of the Northeast provincial capital city from 2000 to 2012.The research work of this paper is as follows:Aiming at the problem of low extraction accuracy caused by light overflow,a new method of building area extraction based on neighborhood extremum is proposed.First of all,a binary quadratic regression model is used to calibrate the night light data.Secondly,a linear filtering algorithm is used to construct a filtering template consisting of the target and eight adjacent pixels with the target as the center.The image is filtered to the extreme value,and a neighbor-hood extreme value image reflecting the gray value change of the image is obtained.Then,the neighborhood extreme value image is obtained,Finally,fractal network segmentation algorithm is used to obtain the image object of the city boundary image,and binary segmentation method is used to obtain the city built-up area.The experimental results show that the average values of kappa coefficient and threshold selection time of neighborhood extremum method are 0.80 and 37 s respectively,which are 0.03,1503s,0.01 and 443s higher than mutation detection method and statistical analysis method,respectively.The spatial coverage area of the con-structed area is close to the reference data,which has better extraction effect and stability.In view of the low resolution of night light data and the problem of virtual detection in the extracted urban built-up area,a method of boundary pixel fusion based on multi-source remote sensing data is proposed to extract the urban built-up area.Firstly,the vector boundary of urban built-up area is extracted by using night light data and Landsat TM data;secondly,the two vector boundaries are gridded respectively;secondly,in view of the missing built-up area in the results of extracting urban built-up area from night light data and Landsat data,the weighted average fusion method is used to fuse the boundary pixel position;finally,the built-up area is extracted.The experimental results show that the overall accuracy and kappa coefficient of the urban built-up area extracted by this method are 1%,0.05,1%and 0.02 higher than that of the night light data and TM data alone,and the spatial coverage of the built-up area extracted by this method is closer to the reference data.In order to analyze the urban spatial and temporal development law of provincial capitals in northeast China,this paper USES built-up areas to demarcate the urban zoning area,and USES the maximum likelihood method to obtain the classification results of urban land cover.In the period 2000-2012 in Shenyang,Changchun and Harbin city as an example,based on the overall urban development trend,fragmentation and polymerization degree of patch category in three aspects:the area(CA),the proportion of landscape patch area(PLAND),the number of patches(NP),patch density(PD)and aggregation index(AI)five landscape index analysis in Shenyang,Changchun and Harbin city the development trend of interior landscape in the city proper.As for the overall development trend of the city,the C A value of urban built-up areas in Shenyang,Changchun and Harbin increased by 17572.88,37208.07 and 15905.05 in 2012 compared with that in 2000.Green space increased by 1,1010.14,1,846.06 and 1,055.39;The water body in-creased by 8793.35,1048.48 and 1433.83,in which the proportion of green space and water body gradually increased,and the overall development of the three cities showed an increasing trend.For the degree of fragmentation,the degree of fragmentation of construction land and green space in Shenyang city decreased,while the degree of fragmentation of water body in-creased.The degree of landscape framentation in Harbin shows an increasing trend.The over-all landscape fragmentation in Changchun city shows a decreasing trend.As for the degree of convergence,the overall degree of convergence of urban landscape in Shenyang and Harbin decreased,the degree of convergence of green space and construction land in Changchun de-creased,and the degree of convergence of water increased.To sum up,it shows that in the process of urban development,Shenyang,Changchun and Harbin have obvious expansion of urban built-up areas,and Shenyang and Harbin have strong efforts in green space and water construction.