Identification Research Of Urban Built-up Areas Based On POI And LJ 1-01

Although the deep accumulation has been made by domestic and foreign scholars in the field of urban built-up areas,there are still shortcomings in the innovative application of data and methods in the extracted urban built-up areas.In this study,POI data,LJ 1-01 data and LJ & POI fusion data are used to identify the built-up areas in the main urban area of Kunming,and the scientific problem of how to improve the extraction accuracy of urban built-up areas is explored based on the existing data and methods.As the only carrier of urban function,the location of urban built-up area determines the development of all urban construction and urban planning.The built-up area of a city even has a direct bearing on the current land and space planning.Using POI data,LJ 1-01 data,LJ & POI fusion data,this study identifies the built-up area of Kunming main urban area,and makes the innovation of the following two points: on the method,the research method of Density-Graph and the OSTU algorithm of GGM function are put forward in this study.On the data,the fusion data LJ & POI is put forward,which all to some extent make up for the lack of innovation of data and methods in previous studies on urban built-up areas,and propose a new perspective and research paradigm for the extraction of urban built-up areas.The research process of this study is mainly divided into three steps: Firstly,the Kernel Density Curve of POI is analyzed.In this process,the critical value of the Kernel Density Curve is selected by using the Density-Graph algorithm,and then the range of urban built-up area is extracted.Then the LJ 1-01 image data is analyzed,and the extraction threshold of night light is determined by the OSTU threshold segmentation algorithm based on the GGM function.Finally,the LJ&POI data is obtained by fusing the POI data with the LJ1-01 data by using the geometric mean value,and then the range of urban built-up area is extracted by using the Density Graph methosd and the OSTU algorithm of GGM function respectively,and the accuracy is compared.The following important conclusions are obtained by using the Density Graph method and the OSTU algorithm of GGM function to identify urban built-up areas:(1)The urban built-up area based on POI identification has both advantages and disadvantages.Although the precision of POI recognition is more than 80%,the overall recognition precision is better,the scope of built-up area extracted by POI Kernel Density analysis is smaller,and the edge details of built-up area extracted are more smooth,because the number of POI data in the lower urbanization area is smaller,which also makes the built-up area extracted is only a general range,so theoretically,it is feasible to identify the built-up area based on POI,but it needs to further eliminate the impact of urbanization on POI data.(2)As for the threshold segmentation calculation of LJ1-01 data by using OSTU algorithm of GGM function,the accuracy of the built-up area extracted also reaches more than 80%,and the extraction effect of the built-up area is roughly the same as that of the built-up area extracted by POI.However,LJ 1-01 data is seriously affected by light overflow,which leads to the complexity of edge details and internal details of the builtup area and the serious plaque of the built-up area.Therefore,it is feasible to apply the OSTU algorithm based on GGM function to the threshold extraction of night light,but this method does not significantly eliminate the impact of light data itself.(3)Using the LJ & POI data from the fusion of POI data and LJ 1-01 data for the analysis of Density Graph and the OSTU algorithm of GGM function shows that the accuracy of the built-up area extracted by the two methods reaches more than 93%.Moreover,the extracted built-up areas are improved in terms of smoothness,complexity of details and degree of fragmentation,and the extracted built-up areas are more complete,indicating that LJ&POI data can achieve better extraction effect than POI and LJ1-01 data.(4)By comparing the extraction accuracy of three kinds of data,the order from high to low should be LJ&POI(OSTU)>LJ&POI(Density-Graph)>LJ1-01(OSTU)>POI(Density-Graph),In addition,LJ&POI(OSTU)and LJ&POI(DensityGraph),as well as LJ1-01(OSTU)and POI(Density-Graph)are roughly the same in terms of the accuracy of the extracted built-up areas.Therefore it can be considered that data fusion plays a greater role in the extraction of urban built-up areas,and the idea of LJ&POI data will also guide the future study of urban built-up areas.The accurate identification of urban built-up areas plays an important practical role in the field of urban planning.The LJ&POI data obtained through the fusion of POI data and LJ1-01 data can accurately identify the built-up areas of cities,which can play a positive guiding role in the future urban planning and urban construction.