Remote sensing extraction method and system of urban built-up areas based on multi-scale spatial nesting

Abstract

The present invention discloses a method and system for extracting urban built-up areas based on multi-scale spatial nesting, wherein the method comprises: constructing an urban index VANUI according to MODIS NDVI data and NPP-VIIRS night light data; adaptively segmenting the urban index VANUI using urban built-up area statistical data to obtain the urban built-up area boundary B1; calculating the morphological building index MBI according to the urban built-up area boundary B1; constructing an adaptive segmentation model, segmenting the morphological building index MBI, and obtaining urban building information O; performing cluster analysis on the urban building information O to obtain regular building information S; calculating the convex hull of the regular building information S to obtain the urban built-up area boundary range. The present invention realizes the nesting of multi-scale space, constrains and continuously optimizes the urban built-up area extraction results, and finally realizes the automatic extraction of urban built-up areas.

Description

Technical Field

[0001] The present invention belongs to the technical field of high-resolution remote sensing image processing and information extraction, and in particular relates to a remote sensing extraction method and system for urban built-up areas based on multi-scale spatial nesting. Background Art

[0002] The area of ​​urban built-up areas is an important criterion for comprehensively measuring the economic development stage of a city and a region, and plays an important role in urban development planning, spatial layout and management optimization. On the one hand, the area of ​​built-up areas reflects the industrial structure and urbanization stage of the corresponding region. On the other hand, the area of ​​built-up areas is also a powerful reflection of regional economic strength. Compared with methods such as field surveying, remote sensing images can quickly and accurately obtain the boundary information of urban built-up areas, and therefore, they are widely studied and applied.

[0003] According to the attribute characteristics of urban built-up areas, there are currently four main methods for extracting urban built-up areas based on remote sensing, including: a) spectral feature extraction based on medium-resolution remote sensing images; b) spatial feature extraction based on high-resolution remote sensing images; c) extraction method based on night light data; and d) a combination of the above methods. However, since the spatial information of the city is an important attribute of the urban built-up area, it is not considered in the urban built-up area index. In addition, the manual selection of thresholds also restricts the efficiency and stability of the urban built-up area information extraction results. Summary of the invention

[0004] The technical problem solved by the present invention is: to overcome the shortcomings of the prior art, to provide a method and system for remote sensing extraction of urban built-up areas based on multi-scale spatial nesting, to realize the nesting of multi-scale space, to constrain and continuously optimize the extraction results of urban built-up areas, and ultimately to realize automatic extraction of urban built-up areas.

[0005] The object of the present invention is achieved through the following technical scheme: a remote sensing extraction method for urban built-up areas based on multi-scale spatial nesting, comprising: constructing an urban index VANUI according to MODIS NDVI data and NPP-VIIRS night light data; adaptively segmenting the urban index VANUI using urban built-up area statistical data to obtain the urban built-up area boundary B1; calculating the morphological building index MBI according to the urban built-up area boundary B1; constructing an adaptive segmentation model to segment the morphological building index MBI to obtain urban building information O; performing cluster analysis on the urban building information O to obtain regular building information S; calculating the convex hull of the regular building information S to obtain the boundary range of the urban built-up area.

[0006] In the above-mentioned urban built-up area remote sensing extraction method based on multi-scale spatial nesting, the urban index VANUI is obtained by the following formula:

[0007] VANUI=DN NPP ×(1-NDVI mean );

[0008] Among them, DN NPP DN value of NPP-VIIRS night light data; NDVI mean is the mean of MODIS NDVI data after filtering; VANUI is the urban index.

[0009] In the above-mentioned urban built-up area remote sensing extraction method based on multi-scale spatial nesting, the urban built-up area boundary B1 is obtained by the following formula:

[0010]

[0011] Among them, B1 is the urban built-up area boundary; DN VANUI is the pixel value of the city index feature image; t is the pixel value interval of the city index feature image; Areat(VANUI) is the area of ​​the city index feature image segmentation result corresponding to the pixel value interval of the city index feature image t.

[0012] In the above-mentioned urban built-up area remote sensing extraction method based on multi-scale spatial nesting, the morphological building index MBI is obtained by the following formula:

[0013]

[0014] Among them, DMP W-TH (d,s)=|MP W-TH (d,(s+Δs))-MP W-TH (d,s)|;

[0015]

[0016] DMP W-TH (d, s) represents the differential morphological profile based on top-hat transformation; D and S represent the number of multi-angles and the number of multi-scales, respectively; MP W-TH (d, s) represent the morphological contours based on top-hat transformation; represents the morphological reconstruction based on b; b represents the maximum brightness of each band of the remote sensing image pixel; d, s and Δs represent the direction, scale and scale increment parameters respectively; k represents the band of the remote sensing image; K represents the maximum value of the applied remote sensing image band; x represents the pixel value of the remote sensing image.

[0017] In the above-mentioned urban built-up area remote sensing extraction method based on multi-scale spatial nesting, the urban building information O is obtained by the following formula:

[0018]

[0019] Where p is the pixel value interval of the morphological building index MBI; Area p (MBI) is the area of ​​the morphological building index segmentation result corresponding to the morphological building index MBI pixel value interval p; Area GUF is the area of ​​global urban footprint data; Mask is the information mask model.

[0020] In the above-mentioned urban built-up area remote sensing extraction method based on multi-scale spatial nesting, the information mask model Mask is:

[0021]

[0022] Among them, GlobeLand30 (water,bare) It represents the "water type" and "bare land type" in the GlobeLand30 land cover product, the "road type" in the Openstreetmap product in OSM, and the "vegetation type" in the Hansen product.

[0023] In the above-mentioned urban built-up area remote sensing extraction method based on multi-scale spatial nesting, the Kmeans clustering algorithm is used to perform cluster analysis on the urban building information O to obtain regular building information S.

[0024] In the above-mentioned urban built-up area remote sensing extraction method based on multi-scale spatial nesting, the regular building information S is obtained by the following formula:

[0025]

[0026] Where k represents the number of clusters; Ci represents the type of cluster; c represents the specific type obtained in the type of cluster; μ i represents the i-th cluster center; F represents the selected feature information; DN F The pixel value obtained for the feature information.

[0027] In the above-mentioned urban built-up area remote sensing extraction method based on multi-scale spatial nesting, the Graham scanning method is used to calculate the convex hull of regular building information S.

[0028] A system for extracting urban built-up areas based on multi-scale spatial nesting includes: a first module for constructing an urban index VANUI based on MODIS NDVI data and NPP-VIIRS night light data; a second module for adaptively segmenting the urban index VANUI using urban built-up area statistical data to obtain the urban built-up area boundary B1; a third module for calculating the morphological building index MBI based on the urban built-up area boundary B1; a fourth module for constructing an adaptive segmentation model, segmenting the morphological building index MBI, and obtaining urban building information O; a fifth module for performing cluster analysis on the urban building information O to obtain regular building information S; and a sixth module for calculating the convex hull of the regular building information S to obtain the urban built-up area boundary range.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] (1) The present invention automatically segments night light data through statistical data of urban built-up areas, obtains rough boundaries of urban built-up areas, and can preliminarily determine the potential area scope of urban built-up areas; within this area, based on high-resolution remote sensing images and medium-resolution thematic information products, extracts information on regular urban buildings, and further refines the potential area scope of urban building boundaries; through Graham scanning, the potential area scope of urban building boundaries is automatically connected to improve the robustness of the method, and finally realizes the automatic information extraction of urban built-up areas;

[0031] (2) The present invention has good robustness for urban built-up areas, especially in medium and large cities, and has the ability to automatically extract in different urban scenes. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0033] Figure 1 It is a flow chart of a method for automatic extraction of urban built-up areas based on high-resolution remote sensing based on multi-scale spatial information nesting provided by an embodiment of the present invention;

[0034] FIG2( a ) is a diagram of NPP-VIIRS night light data (400 meters) in 2016 provided by an embodiment of the present invention;

[0035] FIG. 2( b ) is a diagram of MODIS MOD13Q1 product data (250 meters) in 2016 provided in an embodiment of the present invention;

[0036] FIG2( c ) is a diagram of water body type and bare land type information product (30 meters) of GlobeLand30 in 2010 provided by an embodiment of the present invention;

[0037] FIG. 2( d ) is a 2013 Hansen vegetation type information product diagram provided by an embodiment of the present invention;

[0038] FIG. 2( e ) is a product diagram of Openstreetmap road type information in 2015 provided by an embodiment of the present invention;

[0039] FIG2( f) is a diagram of a 2011 GUF building information product provided by an embodiment of the present invention;

[0040] FIG2( g ) is a high spatial resolution remote sensing image (2 meters) of the Gaofen-1 high spatial resolution remote sensing image on April 9, 2016 provided by an embodiment of the present invention;

[0041] Figure 3 It is a VANUI index characteristic graph corresponding to the 2016 NPP-VIIRS night light data (400 meters) and the 2016 MODISMOD13Q1 product data (250 meters) provided in an embodiment of the present invention;

[0042] Figure 4 The embodiment of the present invention provides a method for adaptively segmenting VANUI using statistical data of urban built-up areas to obtain a result map of the urban built-up area boundary B1;

[0043] Figure 5 It is a result map of the morphological building index MBI calculated from a high-resolution image within the urban built-up area boundary B1 provided by an embodiment of the present invention;

[0044] Figure 6 The embodiment of the present invention provides an adaptive segmentation model based on multi-source mesoscale thematic information products, segmenting the MBI index to obtain an urban building information result map;

[0045] Figure 7 is a schematic diagram of a regular building information result provided by an embodiment of the present invention;

[0046] Figure 8 This is a final extraction result map of the urban built-up area provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0047] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to be able to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0048] In the embodiment of the present invention, the multi-scale spatial data used include: 1) macro-scale remote sensing data: NPP-VIIRS night light data in 2016 (400 meters) (as shown in Figure 2 (a)) and MODIS MOD13Q1 product data in 2016 (250 meters) (as shown in Figure 2 (b)); 2) mesoscale remote sensing data products: water type and bare land type information products of GlobeLand30 in 2010 (30 meters) (as shown in Figure 2 (c)), Hansen vegetation type information products in 2013 (as shown in Figure 2 (d)), Openstreetmap road type information products in 2015 (as shown in Figure 2 (e)) and GUF building information products in 2011 (as shown in Figure 2 (f)); 3) high spatial resolution remote sensing image: GF-1 high spatial resolution remote sensing image of April 9, 2016 (2 meters) (as shown in Figure 2 (g)). Among them, the multispectral and panchromatic resolutions of the Gaofen-1 satellite remote sensing data are 8 meters and 2 meters respectively, with four bands, namely the blue band (0.45-0.52μm), the green band (0.52-0.59μm), the red band (0.63-0.69μm), and the near-infrared band (0.77-0.89μm). The radiation quantization level is 16bit. After the RPC model and Gram-Schmidt fusion processing, the high-resolution orthophoto fusion remote sensing image with a spatial resolution of 2 meters is obtained. The MODIS MOD13Q1 product has a spatial resolution of 250 meters and a temporal resolution of 16 days. Through MRT projection changes, mosaicking and cropping, MODIS NDVI time series data is obtained, and the MODIS NDVI time series is reconstructed through the SG filtering algorithm to eliminate the influence of factors such as atmosphere, sunlight and observation angle, and obtain the reconstructed MODIS NDVI time series product.

[0049] Figure 1 1 is a flow chart of a method for automatic extraction of urban built-up areas based on high-resolution remote sensing based on multi-scale spatial information nesting provided by an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0050] 1) Construct the urban index VANUI based on 250m MODIS NDVI data and 400m NPP-VIIRS night light data;

[0051] VANUI=DN NPP ×(1-NDVI mean )

[0052] Where, DN NPP DN value of NPP-VIIRS night light data; NDVI mean is the mean of the MODIS NDVI time series data after filtering; VANUI is the city index. The results of the VANUI index characteristic graph in this embodiment are shown in Figure 3 .

[0053] 2) Adaptively segment VANUI using statistical data of urban built-up areas to obtain the urban built-up area boundary B1;

[0054]

[0055] Where, B1 is the urban built-up area; DN VANUI is the pixel value of the VANUI feature image; t is the pixel value interval of the VANUI feature, and in this embodiment, the interval of t is [0,1]; Area is the area of ​​the VANUI feature image segmentation result corresponding to the VANUI feature pixel value interval t; N is a parameter, which represents the area in the urban built-up area statistical yearbook of the studied city. According to the China City Statistical Yearbook 2016, the area value of the built-up area of ​​Shanghai in 2016 in this study is 998.6km 2 The result map of the urban built-up area boundary B1 in this embodiment is shown in Figure 4 .

[0056] 3) For the area within the urban built-up area boundary B1, calculate the morphological building index MBI of the high-resolution image;

[0057]

[0058] in, The calculation method is:

[0059] DMP W-TH (d,s)=|MP W-TH (d,(s+Δs))-MP W-TH (d,s)|

[0060] Among them, MP W-TH The calculation method of (d,s) is:

[0061]

[0062] The calculation method of b is:

[0063]

[0064] Where DMP W-TH represents the differential morphological profile based on top-hat transformation; D and S represent the number of multi-angles and the number of multi-scales, which are 4 and 6 respectively through experiments in this study; MP W-TH Represents the morphological contour based on top-hat transformation; Indicates morphological reconstruction based on b; b indicates the maximum brightness of each band of the remote sensing image pixel; d, s and Δs respectively indicate the direction, scale and scale increment parameters, which are set to {0°, 45°, 90° and 135°}, {11, 19, 27, 35, 43, 51, 59} and 8 respectively through experiments in this study; k indicates the band of the remote sensing image; K indicates the maximum value of the applied remote sensing image band, and the high-resolution remote sensing image of GF-1 used in this study has a parameter of 4; x indicates the pixel value of the remote sensing image. The morphological building index MBI result diagram of this embodiment is shown in Figure 5 .

[0065] 4) Building an adaptive segmentation model based on multi-source mesoscale thematic information products to segment the MBI index and obtain urban building information O;

[0066]

[0067] Among them, the calculation method of Mask is:

[0068]

[0069] Where, p is the pixel value interval of the MBI index. In this study, the interval of p is [0, 255]; Area p is the area of ​​the MBI segmentation result when the threshold is p; Area GUF is the area of ​​the thematic information product GUF (Global Urban Footprint, GUF); Mask is the information mask model; GlobeLand30 (water,bare) , OSM and Hansen respectively represent the "water type" and "bare land type" in the GlobeLand30 surface cover product, the "road type" in the Openstreetmap product, and the "vegetation type" in the Hansen product. Figure 6 .

[0070] 5) Using Kmeans clustering algorithm to perform cluster analysis on the extracted urban building results O to obtain regular building information results S;

[0071]

[0072] Among them, μ i The calculation method is:

[0073]

[0074] In the formula, k represents the number of clusters, the parameter is 2; Ci represents the type of cluster, the parameter is 2, indicating two types; c represents the specific type obtained in the type of cluster, namely: regular buildings and irregular buildings; μ i represents the i-th cluster center; F represents the selected feature information. This study uses MBI features, rectangular geometric features and texture features through experiments; DN F The pixel value obtained by the feature information. Figure 7 .

[0075] 6) The Graham scanning method is used to calculate the convex hull of the regular building information S to obtain the final boundary range of the urban built-up area.

[0076] a) First select the point set S (s0, s1, s2, s3, ..., s n ) is denoted as s0. If the y coordinates are the same, then the point with the smallest x coordinate is the same point. It should be understood that the regular building information S is a point set S (s0, s1, s2, s3, ..., s n ), where s0 is the first building information, s1 is the second building information, s2 is the third building information, s3 is the fourth building information, s n The n+1th building information;

[0077] b) Put (s0,s1,s2,s3,……,s n ) Sort by polar angle from small to large (with s0 as the pole), and the points with the same polar angle are sorted by the distance to s0 from small to large;

[0078] c) Push s0, s1, s2 into the stack;

[0079] d) Traverse the remaining points s3, s4, ..., s n By calculating the cross product between two points (s n -s0)×(s n-1 -s0)=(x n -x0)×(y n -y0)-(x n-1 -x0)×(y n-1-y0), when the cross product between the two points is less than 0, it indicates that it is in the counterclockwise direction, and the vertex is retained. When the cross product between the two points is greater than 0, it indicates that it is in the clockwise direction, and the vertex is removed;

[0080] e) By traversing, the final boundary range of the urban built-up area is finally obtained. The final extraction result of the urban built-up area in this embodiment Figure 8 .

[0081] The present embodiment also provides a system for extracting urban built-up areas based on multi-scale spatial nesting, including: a first module, which is used to construct an urban index VANUI based on MODIS NDVI data and NPP-VIIRS night light data; a second module, which is used to adaptively segment the urban index VANUI using urban built-up area statistical data to obtain the urban built-up area boundary B1; a third module, which is used to calculate the morphological building index MBI based on the urban built-up area boundary B1; a fourth module, which is used to construct an adaptive segmentation model, segment the morphological building index MBI, and obtain urban building information O; a fifth module, which is used to perform cluster analysis on the urban building information O to obtain regular building information S; and a sixth module, which is used to calculate the convex hull of the regular building information S to obtain the boundary range of the urban built-up area.

[0082] The present invention gives full play to the thematic information advantages of large-scale night light data, the thematic attribute advantages of medium-scale optical remote sensing, and the spatial information advantages of high-resolution remote sensing images. Focusing on the comprehensive attributes of urban built-up areas, the present invention applies prior information technologies such as unstructured statistical data, clustering methods, and adaptive segmentation to construct an adaptive segmentation model, associate large-medium-small scale spatial information, realize the nesting of multi-scale spaces, constrain and continuously optimize the extraction results of urban built-up areas, and finally realize the automatic extraction of urban built-up areas.

[0083] The present invention automatically segments night light data through statistical data of urban built-up areas, obtains rough boundaries of urban built-up areas, and can preliminarily determine the potential area scope of urban built-up areas; within this area, based on high-resolution remote sensing images and medium-resolution thematic information products, urban regular building information is extracted, and the potential area scope of urban building boundaries is further refined; through Graham scanning, the potential area scope of urban building boundaries is automatically connected to improve the robustness of the method, and finally realize the automatic information extraction of urban built-up areas; the present invention has good robustness for urban built-up areas, especially under the conditions of medium and large cities, and has the ability of automatic extraction in different urban scenes.

[0084] Although the present invention has been disclosed as above in the form of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art may make possible changes and modifications to the technical solution of the present invention by using the methods and technical contents disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the protection scope of the technical solution of the present invention.

Claims

1. A remote sensing extraction method for urban built-up areas based on multi-scale spatial nesting, characterized in that include: Constructing urban index based on MODIS NDVI data and NPP-VIIRS night light data; The city index VANUI is obtained by the following formula: VANUI=DN NPP ×(1-NDVI mean ); Among them, DN NPP DN value of NPP-VIIRS night light data; NDVI mean is the mean value of MODIS NDVI data after filtering; VANUI is the urban index; Adaptively segment the urban index using the statistical data of urban built-up areas to obtain the boundaries of urban built-up areas; Calculate the morphological building index based on the boundaries of the urban built-up area; The morphological building index MBI is obtained by the following formula: Among them, DMP W-TH (d,s)=|MP W-TH (d,(s+Δs))-MP W-TH (d,s)|; DMP W-TH (d, s) represents the differential morphological profile based on top-hat transformation; D and S represent the number of multi-angles and the number of multi-scales, respectively; MP W-TH (d, s) represent the morphological contours based on top-hat transformation; represents the morphological reconstruction based on b; b represents the maximum brightness of each band of the remote sensing image pixel; d, s and Δs represent the direction, scale and scale increment parameters respectively; k represents the band of the remote sensing image; K represents the maximum value of the applied remote sensing image band; x represents the pixel value of the remote sensing image; Construct an adaptive segmentation model to segment the morphological building index and obtain urban building information; The urban building information O is obtained through the following formula: Where p is the pixel value interval of the morphological building index; Area p (MBI) is the area of ​​the morphological building index segmentation result corresponding to the morphological building index MBI pixel value interval; Area GUF is the area of ​​global urban footprint data; Mask is the information mask model; Conduct cluster analysis on urban building information to obtain regular building information; The regular building information S is obtained by the following formula: Where k represents the number of clusters; Ci represents the type of cluster; c represents the specific type obtained in the type of cluster; μ i represents the i-th cluster center; F represents the selected feature information; DN F The pixel value obtained for the feature information; Calculate the convex hull of regular building information and obtain the boundary range of the urban built-up area.

2. The method for extracting urban built-up areas based on remote sensing based on multi-scale spatial nesting according to claim 1, characterized in that: The urban built-up area boundary B1 is obtained by the following formula: Among them, B1 is the boundary of the urban built-up area; DN VANUI is the pixel value of the city index feature image; t is the pixel value interval of the city index feature image; Areat(VANUI) It is the area of ​​the city index feature image segmentation result corresponding to the city index feature image pixel value interval.

3. The method for extracting urban built-up areas based on remote sensing based on multi-scale spatial nesting according to claim 1, characterized in that: The information mask model Mask is: Among them, GlobeLand30 (water,bare) Indicates the "water type" and "bare land type" in the GlobeLand30 land cover product, OSM is the "road type" in the Openstreetmap product, and Hansen is the "vegetation type" in the Hansen product.

4. The method for extracting urban built-up areas based on remote sensing based on multi-scale spatial nesting according to claim 1, characterized in that: The Kmeans clustering algorithm is used to perform cluster analysis on the urban building information O to obtain regular building information S.

5. The method for extracting urban built-up areas based on remote sensing based on multi-scale spatial nesting according to claim 1, characterized in that: The Graham scanning method is used to calculate the convex hull of the regular building information S.

6. A remote sensing extraction system for urban built-up areas based on multi-scale spatial nesting, characterized in that include: The first module is used to construct the urban index based on MODIS NDVI data and NPP-VIIRS night light data; The city index VANUI is obtained by the following formula: VANUI=DN NPP ×(1-NDVI mean ); Among them, DN NPP DN value of NPP-VIIRS night light data; NDVI mean is the mean value of MODIS NDVI data after filtering; VANUI is the urban index; Adaptively segment the urban index using the statistical data of urban built-up areas to obtain the boundaries of urban built-up areas; The second module is used to adaptively segment the urban index VANUI using the statistical data of urban built-up areas to obtain the boundaries of urban built-up areas; The third module is used to calculate the morphological building index based on the boundaries of the urban built-up area; The morphological building index MBI is obtained by the following formula: Among them, DMP W-TH (d,s)=|MP W-TH (d,(s+Δs))-MP W-TH (d,s)|; DMP W-TH (d, s) represents the differential morphological profile based on top-hat transformation; D and S represent the number of multi-angles and the number of multi-scales, respectively; MP W-TH (d, s) represent the morphological contours based on top-hat transformation; represents the morphological reconstruction based on b; b represents the maximum brightness of each band of the remote sensing image pixel; d, s and Δs represent the direction, scale and scale increment parameters respectively; k represents the band of the remote sensing image; K represents the maximum value of the applied remote sensing image band; x represents the pixel value of the remote sensing image; The fourth module is used to build an adaptive segmentation model to segment the morphological building index and obtain urban building information; The urban building information O is obtained through the following formula: Where p is the pixel value interval of the morphological building index; Area p (MBI) is the area of ​​the morphological building index segmentation result corresponding to the morphological building index MBI pixel value interval; Area GUF is the area of ​​global urban footprint data; Mask is the information mask model; The fifth module is used to perform cluster analysis on urban building information and obtain regular building information; The regular building information S is obtained by the following formula: Where k represents the number of clusters; Ci represents the type of cluster; c represents the specific type obtained in the type of cluster; μ i represents the i-th cluster center; F represents the selected feature information; DN F The pixel value obtained for the feature information; The sixth module is used to calculate the convex hull of regular building information and obtain the boundary range of the urban built-up area.

Images