Urban and rural statistical population spatialization method based on two-dimensional and three-dimensional land function structure

Abstract

The invention relates to the field of remote sensing technology and geographic information systems, and discloses an urban and rural statistical population spatialization method based on a two-dimensional and three-dimensional land function structure. According to the method, a unified geographic framework is constructed, and spatial gridding processing is carried out on multi-source data such as geographic space big data and high-resolution remote sensing data; constructing a population spatialization model and evaluation index system considering the building three-dimensional structure based on the multi-source data; urban and rural partitioning is carried out through town boundary data, the weight of each index factor of a town area and the weight of each index factor of a rural area are calculated respectively, and multi-source data fusion is carried out in combination with population statistical data to obtain a population spatialization result. On the basis of gridding processing of multi-source heterogeneous geographic space data, a population spatialization model is optimized through digital surface model (DSM) data and point of interest (POI) data, population spatialization is carried out regionally under the condition that influences of various factors are comprehensively considered, and a new thought and method are provided for population spatialization research.

Description

Technical Field

[0001] The invention relates to the fields of remote sensing technology and geographic information system, and discloses a method for spatialization of urban and rural statistical population based on two-dimensional and three-dimensional land use functional structures. Background Art

[0002] The spatial distribution pattern of population is key information in the fields of national land space planning, urban development research, natural resource development, disaster risk assessment and rescue. Traditional population data is usually obtained by statistical means such as population census based on administrative districts. The statistical method has rigorous theoretical support and authoritative normative processes, but there are problems such as long update cycle, difficulty in data visualization and spatial analysis, and data sources are too dependent on administrative district boundaries, resulting in difficulty in scale conversion of statistical data. Population spatialization can break through the administrative area restrictions of population data and simulate the spatial distribution of population at the grid scale, thereby more objectively and truly reflecting the population density differentiation in the designated study area.

[0003] Most of the existing population spatialization methods are based on population spatialization models established based on land use types, with a single modeling factor, or they select several factors such as land use, terrain, and night light intensity for modeling. Most of these methods only consider two-dimensional scale information and cannot fully reflect the impact of multidimensional information such as the three-dimensional structure and building attributes of urban areas on population distribution; at the same time, they lack consideration of the different population distribution influencing laws between urban and rural areas. Summary of the invention

[0004] The present invention discloses a method for spatialization of urban and rural statistical population based on two-dimensional and three-dimensional land use functional structures, and provides relevant solutions to the above problems.

[0005] A method for spatialization of urban and rural statistical population based on two-dimensional and three-dimensional land use functional structure, using geospatial big data and high-resolution remote sensing data multi-source data, comprehensively considering the impact of building three-dimensional structure and other population spatial distribution influencing factors on the spatialization of population in urban and rural areas, including the following steps: S1, obtain high-resolution remote sensing image data, night light data, terrain data, river road network data, demographic data, DSM data, points of interest (POI), administrative division data, urban development boundary multi-source data, build a unified geographic grid framework, and perform spatial grid processing on various multi-source heterogeneous data; S2, based on multi-source data, constructs a population spatialization index system for urban and rural areas respectively. The modeling factors mainly include land use, night light intensity, terrain slope, river buffer zone, road buffer zone, and the increase of building height and point of interest (POI) factors in urban areas. The weight of each factor is calculated by analyzing the contribution rate of various modeling factors to the evaluation index system; S3, based on the weights of each factor and the statistical data of urban and rural populations in each district and county, a population spatialization model is constructed to spatialize the population.

[0006] Step S1 specifically includes the following steps: S11, obtain regional high-resolution satellite remote sensing image data, use computer automatic classification and manual visual interpretation to extract the current land use information and the spatial distribution of rivers and road networks in the region, and perform buffer processing on river and road network vectors; obtain regional DEM terrain data and extract slope spatial distribution information; S12, obtain regional nighttime light data, use SG filtering (Savitzky Golay Filter) to smooth out the abnormal values ​​of the light data, and obtain the spatial distribution of nighttime light intensity; S13, obtain DSM data, demographic data, administrative division data and point of interest (POI) data, organize and clean the point of interest (POI) data, and obtain the spatial distribution of POI data of catering service density, company enterprise density, shopping service density, transportation facility density, science and education culture density, commercial residential density, life service density, medical care density, and government agency density; S14, build a unified geographic grid framework, and perform spatial grid processing on all data: for slope and night light intensity raster type data, calculate the average value of all rasters inside the grid as the grid value; for vector data of various POI point types, count the number of points in each grid according to the spatial position, and assign the total number of points to the constructed geographic grid; for river road network buffer vector data, superimpose the vector and the grid, and if the two intersect and the intersection area is greater than 50% of the grid area, the grid is assigned a value of 1, and the rest are 0; for land use vector data, count the area of ​​different types of patches in each grid according to the spatial position, and assign the land type attribute of the patch with the largest area to the grid; for demographic data, first associate it with the administrative division vector, and then assign the demographic data to each grid according to the spatial position.

[0007] Step S2 specifically includes the following steps: S21, land use factor assignment, assigning different levels of 1 to 10 to land use types with population distribution, the higher the level, the higher the correlation with population, and 0 when there is no population distribution; land use types are mainly divided into 6 types: cultivated land, forest land, grassland, water area, artificial surface, and bare land, and the values ​​are 3, 2, 1, 0, 9, and 1 respectively; S22, slope factor assignment, the slope is assigned different levels from 0 to 10, the higher the level, the higher the correlation with the population, the slope of 0-5° is assigned to 10, 5-10° is assigned to 7, 10-15° is assigned to 5, 15-30° is assigned to 3, and greater than 30° is assigned to 1; S23, modeling normalization processing, except for land use factor and slope factor, other modeling factors are normalized, with a unified value range of 0~10. After normalization, the grid data is the maximum value of each factor, is the minimum value of each factor, Indicates the grid number. The higher the factor value, the more population is allocated to the area in theory. S24, construct a rural area population index system, calculate the weights of five factors: land use, night light intensity, slope, river buffer zone, and road buffer zone; use the entropy weight method to calculate the weight of each index and construct an evaluation matrix Standardize the raw data of various indicators, and for positive indicators such as land use, night light intensity, river and road buffer zones Slope is negative indicator is the maximum value of the original data of the indicator, is the minimum value of the original data of the indicator, is the standardized value; the information entropy of each indicator The weight of each indicator is is the number of grids, is the number of indicators; S25, calculating the building height in the urban area, extracting the elevation information of the building outline and the highest point within the outline based on the DSM data, making a 10m buffer based on the highest point, and taking the road elevation within the buffer as the ground height, the difference between the highest point elevation of the building and the road elevation is the building height; S26, construct a population index system for urban areas. In addition to factors such as land use, night light intensity, slope, river buffer zone, and road buffer zone, add building height factors and building attribute factors such as catering service density, company density, shopping service density, transportation facility density, science, education and culture density, commercial residential density, living service density, medical care density, and government agency density; use the entropy weight method in step S24 to calculate the weight of each factor.

[0008] Step S3 specifically includes the following steps: S31, using urban development boundary data to divide the region into two areas, urban and rural, and spatializing the population separately; S32, use mask to remove grids with water land use type, and then spatialize the population of urban and rural areas according to the zoning population statistics and the weights of each indicator. For the County The population of each grid, For the The total population of the county, For the The weight of each indicator; For the Grid The index value, For the The grid is located in County The total value of each indicator is used to obtain the population spatial distribution grid result. Beneficial Effects

[0009] The present invention discloses a method for spatialization of urban and rural statistical population based on two-dimensional and three-dimensional land use functional structure; the method constructs a unified geographic framework, and processes multi-source data such as geographic spatial big data and high-resolution remote sensing data into spatial grids; constructs a population spatialization model and evaluation index system that considers the three-dimensional structure of buildings based on multi-source data; divides urban and rural areas into urban and rural areas through urban boundary data, calculates the weights of various indicator factors in urban and rural areas respectively, and combines demographic data to perform multi-source data fusion to obtain population spatialization results. Based on the grid processing of multi-source heterogeneous geographic spatial data, the present invention optimizes the population spatialization model through building height data and point of interest (POI) data, and performs population spatialization by region under comprehensive consideration of the influence of various factors, providing new ideas and methods for population spatialization research. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A flow chart of a spatialization method for urban and rural statistical population based on two-dimensional and three-dimensional land use functional structure provided for this application; Figure 2 A flowchart of another embodiment of a method for spatializing urban and rural statistical population based on two- and three-dimensional land use functional structure provided in this application. DETAILED DESCRIPTION

[0011] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, so that the processes, methods, systems, products or devices of a series of units are not necessarily limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or devices.

[0012] Reference Figure 1 and Figure 2The method of spatialization of urban and rural statistical population based on two-dimensional and three-dimensional land use functional structure uses multi-source data such as geospatial big data and high-resolution remote sensing data, and comprehensively considers the impact of building three-dimensional structure and other population spatial distribution factors on the spatialization of urban and rural population, such as Figure 1 As shown, the following steps are included: In step S1, high-resolution remote sensing image data, night light data, terrain data, river network data, demographic data, DSM data, point of interest (POI) administrative division data, urban development boundaries and other multi-source data are obtained, and a unified geographic grid framework is constructed to perform spatial grid processing on various multi-source heterogeneous data, which specifically includes the following steps: S11, obtain regional high-resolution satellite remote sensing image data, use computer automatic classification and manual visual interpretation to extract the current land use information and the spatial distribution of rivers and road networks in the region, and perform buffer processing on river and road network vectors; obtain regional DEM terrain data and extract slope spatial distribution information; S12, obtain regional nighttime light data, use SG filtering (Savitzky Golay Filter) to smooth out the abnormal values ​​of the light data, and obtain the spatial distribution of nighttime light intensity; S13, obtain DSM data, demographic data, administrative division data and point of interest (POI) data, organize and clean the point of interest (POI) data, and obtain the spatial distribution of POI data such as catering service density, company enterprise density, shopping service density, transportation facility density, science and education culture density, commercial residential density, life service density, medical care density, and government agency density; S14, build a unified geographic grid framework, and perform spatial grid processing on all data: for raster type data such as slope and night light intensity, calculate the average value of all rasters within the grid as the grid value; for vector data of various POI point types, count the number of points in each grid according to the spatial position, and assign the total number of points to the constructed geographic grid; for vector data such as river road network buffer, superimpose the vector and the grid, and if the two intersect and the intersection area is greater than 50% of the grid area, the grid is assigned a value of 1, and the rest are 0; for land use vector data, count the area of ​​different types of patches in each grid according to the spatial position, and assign the land type attribute of the patch with the largest area to the grid; for demographic data, first associate it with the administrative division vector, and then assign the demographic data to each grid according to the spatial position.

[0013] In step S2, based on multi-source data, spatial index systems of urban and rural populations are constructed respectively. Modeling factors mainly include land use, night light intensity, terrain slope, river buffer zone, road buffer zone, etc. Urban areas add factors such as building height and POI. The weight of each factor is calculated by analyzing the contribution rate of various modeling factors to the evaluation index system. Specifically, the following steps are included: S21, land use factor assignment, assigning different levels of 1 to 10 to land use types with population distribution, the higher the level, the higher the correlation with population, and 0 when there is no population distribution; land use types are mainly divided into 6 types: cultivated land, forest land, grassland, water area, artificial surface, and bare land, and the values ​​are 3, 2, 1, 0, 9, and 1 respectively; S22, slope factor assignment, the slope is assigned different levels from 0 to 10, the higher the level, the higher the correlation with the population, the slope of 0-5° is assigned to 10, 5-10° is assigned to 7, 10-15° is assigned to 5, 15-30° is assigned to 3, and greater than 30° is assigned to 1; S23, modeling normalization processing, except for land use factor and slope factor, other modeling factors are normalized, with a unified value range of 0~10. After normalization, the grid data is the maximum value of each factor, is the minimum value of each factor, Indicates the grid number. The higher the factor value, the more population is allocated to the area in theory. S24, construct a rural regional population index system, calculate the weights of five factors including land use, night light intensity, slope, river buffer zone, and road buffer zone; use the entropy weight method to calculate the weight of each indicator and construct an evaluation matrix Standardize the raw data of various indicators, and for positive indicators such as land use, night light intensity, river and road buffer zones Slope is negative indicator is the maximum value of the original data of the indicator, is the minimum value of the original data of the indicator, is the standardized value; the information entropy of each indicator The weight of each indicator is is the number of grids, is the number of indicators; the indicator weight calculation results are: land use 0.32, night light intensity 0.45, slope 0.14, river buffer 0.06, road buffer 0.03.

[0014] S25, calculating the building height in the urban area, extracting the elevation information of the building outline and the highest point within the outline based on the DSM data, making a 10m buffer based on the highest point, and taking the road elevation within the buffer as the ground height, the difference between the highest point elevation of the building and the road elevation is the building height; S26, construct a population index system for urban areas, in addition to factors such as land use, night light intensity, slope, river buffer zone, road buffer zone, etc., add building height factors and building attribute factors such as catering service density, company enterprise density, shopping service density, transportation facility density, science and education culture density, business residential density, life service density, medical care density, and government agency density; calculate the weight of each factor according to the principal component analysis method in step S24. The calculation results are land use 0.13, night light intensity 0.16, slope 0.02, river buffer zone 0.03, road buffer zone 0.03, building height 0.18, catering service density 0.07, company enterprise density 0.04, shopping service density 0.04, transportation facility density 0.03, science and education culture density 0.02, business residential density 0.08, life service density 0.06, medical care density 0.09, and government agency density 0.02.

[0015] Step S3 specifically includes the following steps: S31, using urban development boundary data to divide the region into two areas, urban and rural, and spatializing the population separately; S32, use mask to remove grids with water land use type, and then spatialize the population of urban and rural areas according to the zoning population statistics and the weights of each indicator. For the County The population of each grid, For the The total population of the county, For the The weight of each indicator; For the Grid The index value, For the The grid is located in County The total value of each indicator is used to obtain the grid result of Qingdao’s population spatial distribution.

[0016] It should be noted that most of the existing population spatialization methods are based on population spatialization models established based on land use types, with a single modeling factor, or they select several factors such as land use, terrain, and night light intensity for modeling. Most of these methods only consider two-dimensional scale information and cannot fully reflect the impact of multidimensional information such as the three-dimensional structure of buildings and building attributes in urban areas on population distribution; at the same time, there is a lack of consideration of the different population distribution influencing laws in urban and rural areas.

[0017] The present invention is based on satellite remote sensing and geographic information technology, and comprehensively considers natural and human geographical factors such as land use, terrain conditions, river and road network distribution, and night light intensity. On the basis of the previous population spatialization model, it adds influencing factors such as building height and POI information to optimize the population spatialization model, providing new ideas and methods for population spatialization research.

Claims

1. A method for spatialization of urban and rural statistical population based on two-dimensional and three-dimensional land use functional structure, characterized in that: Using geospatial big data and high-resolution remote sensing data, the paper comprehensively considers the impact of building three-dimensional structure and other population spatial distribution influencing factors on the spatialization of population in urban and rural areas, including the following steps: S1, obtain high-resolution remote sensing image data, night light data, terrain data, river road network data, demographic data, DSM data, points of interest, administrative division data, urban development boundary multi-source data, build a unified geographic grid framework, and perform spatial grid processing on various multi-source heterogeneous data; S2, based on the multi-source data, respectively construct the population spatialization index system of urban and rural areas; the modeling factors mainly include: land use, night light intensity, terrain slope, river buffer zone, road buffer zone, urban area increase building height, point of interest factor, and calculate the weight of each factor by analyzing the contribution rate of each type of modeling factor to the evaluation index system; S3, based on the weights of each factor and the statistical data of urban and rural populations in each district and county, a population spatialization model is constructed to spatialize the population.

2. The method for spatialization of urban and rural statistical population based on two-dimensional and three-dimensional land use functional structure according to claim 1 is characterized in that: The step S1 specifically includes the following steps: S11, obtain regional high-resolution satellite remote sensing image data, use computer automatic classification and manual visual interpretation to extract the current land use information and the spatial distribution of rivers and road networks in the region, and perform buffer processing on river and road network vectors; obtain regional DEM terrain data and extract slope spatial distribution information; S12, obtaining regional nighttime light data, using the SG filtering method to smooth out the abnormal values ​​of the light data, and obtaining the spatial distribution of nighttime light intensity; S13, obtaining the DSM data, the demographic data, the administrative division data and the POI data, sorting and cleaning the POI data, and obtaining the spatial distribution of POI data of catering service density, company enterprise density, shopping service density, transportation facility density, science and education culture density, commercial residential density, life service density, medical care density, and government agency density; S14, build a unified geographic grid framework, and perform spatial grid processing on all data: for slope and night light intensity raster type data, calculate the average value of all rasters inside the grid as the grid value; for vector data of various POI point types, count the number of points in each grid according to the spatial position, and assign the total number of points to the constructed geographic grid; for river road network buffer vector data, superimpose the vector and the grid, and if the two intersect and the intersection area is greater than 50% of the grid area, the grid is assigned a value of 1, and the rest are 0; for land use vector data, count the area of ​​different types of patches in each grid according to the spatial position, and assign the land type attribute of the patch with the largest area to the grid; for demographic data, first associate it with the administrative division vector, and then assign the demographic data to each grid according to the spatial position.

3. The method for spatialization of urban and rural statistical population based on two-dimensional and three-dimensional land use functional structure according to claim 2 is characterized in that: The step S2 specifically includes the following steps: S21, land use factor assignment, assigning different levels of 1 to 10 to land use types with population distribution, the higher the level, the higher the correlation with population, and 0 when there is no population distribution; land use types are mainly divided into 6 types: cultivated land, forest land, grassland, water area, artificial surface, and bare land, and the values ​​are 3, 2, 1, 0, 9, and 1 respectively; S22, slope factor assignment, the slope is assigned different levels from 0 to 10, the higher the level, the higher the correlation with the population, the slope of 0-5° is assigned to 10, 5-10° is assigned to 7, 10-15° is assigned to 5, 15-30° is assigned to 3, and greater than 30° is assigned to 1; S23, modeling normalization processing, except for land use factor and slope factor, other modeling factors are normalized, with a unified value range of 0~10. After normalization, the grid data is the maximum value of each factor, is the minimum value of each factor, Indicates the grid number. The higher the factor value, the more population is allocated to the area in theory. S24, construct a rural area population index system and calculate the weights of five factors: land use, night light intensity, slope, river buffer zone, and road buffer zone; Use the entropy weight method to calculate the weight of each indicator and construct an evaluation matrix Standardize the raw data of various indicators, and for positive indicators such as land use, night light intensity, river and road buffer zones Slope is negative indicator is the maximum value of the original data of the indicator, is the minimum value of the original data of the indicator, Y is the standardized value; the information entropy of each indicator The weight of each indicator is is the number of grids, is the number of indicators; S25, calculating the building height in the urban area, extracting the elevation information of the building outline and the highest point within the outline based on the DSM data, making a 10m buffer based on the highest point, and taking the road elevation within the buffer as the ground height, the difference between the highest point elevation of the building and the road elevation is the building height; S26, constructing a population index system for urban areas, in addition to factors such as land use, night light intensity, slope, river buffer zone, road buffer zone, etc., adding building height factors and building attribute factors such as catering service density, company enterprise density, shopping service density, transportation facility density, science, education and culture density, commercial residential density, life service density, medical care density, and government agency density; calculating the weight of each factor according to the entropy weight method in step S24.

4. The method for spatialization of urban and rural statistical population based on two-dimensional and three-dimensional land use functional structure according to claim 3 is characterized in that: The step S3 specifically includes the following steps: S31, using urban development boundary data to divide the region into two areas, urban and rural, and spatializing the population separately; S32, use mask to remove grids with water land use type, and then spatialize the population of urban and rural areas according to the zoning population statistics and the weights of each indicator. For the County The population of each grid, For the The total population of the county, For the The weight of each indicator; For the Grid The index value, For the The grid is located in County The total value of each indicator is used to obtain the population spatial distribution grid result.

Images