Monitoring And Analysis Of Urban Land Expanded In The Central Region Of China Using DMSP-OLS Data

City is an important place for human life and the city combines various elements of social and economy,population and ecological environment effectively.The rapid development of industrialization in China has greatly promoted the development of urbanization in our country since Reform and Opening in1978,especially with the deepening of central region strategy rising in the last dozen years or so,the speed of urbanization in the central region of China to be further improved,and the development speed and scale of construction land is increasing and expanding,which is also combined with speed up city expansion rapidly.This paper using central region of China as the study area by using DMSP/OLS data,Landsat TM data and China Statistics Yearbook as the main data source,Extraction of urban land-use information in Central China by using support vector machine algorithm,combined with the classification results of Landsat TM data and China Statistics Yearbook as the city land information extraction accuracy verification basis.The dynamic characteristics of urban land use during the four periods of 2003-2012 were analyzed,identified in Central China city expansion of the main social and economic driving factors and mechanism in recent ten years using grey relational analysis and Back Propagation Neural Network method.Through the exploration on the extraction and analysis methods of macro scale urban areas from remote sensing image,the characteristics of urban sprawl were obtained and evaluated.The study has important practical and theoretical significance to promote the sustainable development of social economy of coastal area and even the whole country.The main conclusions are illustrated as follows:1.The feasibility and reliability of using support vector method to extract urban-land information from DMSP/OLS night light data is verified.The urban construction classification results of Landsat TM data and China Statistics Yearbook of the study area,This paper uses the classification results of Landsat TM data and statistical yearbook data to verify the extraction results of urban construction land in the study area,two important provinces and six provincial capitals by using support vector machine method.2.It is proved that the method of correlation analysis in Grey Relational Analysis is feasible in the analysis of the driving force of urban expansion.Comparing the relationship between long time sequence number expansion of the city economic and social impact ofsome atypical features of the data,determine the correlation degree of each factor with the expansion of the city,the relationship between the quantitative research of urban expansion and social and economic factors.3.The paper built a model for urban expansion based on the city of central China.Then it combined with the economic and social factors in the study period,which was used to forecast the urban land area in 2015.The simulated values and the actual values are of the fitting result can be very good.All of this can be the conclusion as dynamic model constructing is suitable for city models in central area.The relative gap predictions by neural model are relatively high because of low ratio of city availability but it can cater demands for scale of city in a big extent.