CN116310853A - Multi-source data-based extraction method for edge regions of medium and small cities - Google Patents

Multi-source data-based extraction method for edge regions of medium and small cities Download PDF

Info

Publication number
CN116310853A
CN116310853A CN202211565899.2A CN202211565899A CN116310853A CN 116310853 A CN116310853 A CN 116310853A CN 202211565899 A CN202211565899 A CN 202211565899A CN 116310853 A CN116310853 A CN 116310853A
Authority
CN
China
Prior art keywords
poi
data
nuclear density
cities
landscape
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202211565899.2A
Other languages
Chinese (zh)
Inventor
李健锋
彭飚
刘思琪
魏雨露
张璐璐
王璐瑶
郭超
谢潇
齐丽
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xian Jiaotong University
Shaanxi Land Engineering Technology Research Institute Co Ltd
Original Assignee
Xian Jiaotong University
Shaanxi Land Engineering Technology Research Institute Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Xian Jiaotong University, Shaanxi Land Engineering Technology Research Institute Co Ltd filed Critical Xian Jiaotong University
Priority to CN202211565899.2A priority Critical patent/CN116310853A/en
Publication of CN116310853A publication Critical patent/CN116310853A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G06V20/13—Satellite images
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A30/00—Adapting or protecting infrastructure or their operation
    • Y02A30/60—Planning or developing urban green infrastructure

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Software Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • Medical Informatics (AREA)
  • Health & Medical Sciences (AREA)
  • Databases & Information Systems (AREA)
  • Remote Sensing (AREA)
  • General Health & Medical Sciences (AREA)
  • Astronomy & Astrophysics (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Image Analysis (AREA)

Abstract

The application discloses a method for extracting a border region of a medium and small city based on multi-source data, which comprises the following steps: acquiring remote sensing images, POI (PointOfInterest) data, lamplight images and population data of cities; analyzing the remote sensing image to obtain landscape disorder degree; evaluating the nuclear density according to the POI data to obtain the POI nuclear density; determining night light intensity according to the light image; determining a single interpretation effort degree of the view disorder degree, the POI nuclear density and the night light intensity on population data; determining the weight of the landscape disorder degree, the POI nuclear density and the night light intensity; determining the comprehensive index of each region according to the weight; the regions are divided into a plurality of categories according to the comprehensive index, including urban edge areas. The model provided by the application is obviously stronger than a landscape disorder degree threshold method and a POI nuclear density breakpoint analysis method in the aspects of precision, detail and integrity, has higher universality and can meet the research requirements of the border areas of small and medium cities.

Description

Multi-source data-based extraction method for edge regions of medium and small cities
Technical Field
The application relates to the technical field of image processing, in particular to a method for extracting a border region of a medium and small city based on multi-source data.
Background
The urban border area is used as a connecting tie between cities and villages, is an area with the fastest land utilization and space structure change in the urban expansion process, and has the characteristics of diversity, dynamics and transition. The method for quickly and accurately identifying the urban border area has important practical significance for optimizing urban space layout, controlling urban unlimited expansion and protecting land resources.
The existing urban and rural border area identification methods mainly comprise an urban and rural gradient view method, a threshold method, a mutation point/fracture point analysis method and the like. Urban and rural gradient view methods are mainly used for identifying urban and urban border areas according to spatial gradient changes of regional land utilization, socioeconomic performance, population density and other factors. However, urban and rural gradient view methods have difficulty overcoming subjectivity in determining demarcation points in discrete areas of the landscape structure. The threshold method is to determine the urban border area according to the threshold range of the indexes such as distance from the built-up area, population density, building proportion, information entropy and the like. The threshold method has the characteristics of simplicity and practicability, but the determination of the normal threshold is obtained through repeated experiments, and the problems of low efficiency, discontinuous results, poor universality and the like exist. The mutation point/breaking point analysis method is to calculate mutation/breaking values of single or comprehensive indexes such as night light intensity, impervious surface index, landscape disorder degree and the like in different directions through a model to determine urban edge areas, and is a current mainstream method.
However, the recognition research for urban edges is mainly focused on large cities, and no extraction model suitable for the edge areas of small and medium cities exists. Different from large cities, the range of the edge areas of the small and medium cities is often smaller, and the spatial resolution of the data (economic, population and lamplight images) related to urban development is lower, so that the difficulty of accurately identifying the edge areas of the cities is increased.
Disclosure of Invention
The embodiment of the application provides a method for extracting a border region of a medium and small city based on multi-source data, which is used for solving the problems of high difficulty, low precision, poor generality and discontinuous result existing in the extraction of the border region of the medium and small city in the prior art.
In one aspect, an embodiment of the present application provides a method for extracting a border region of a medium and small city based on multi-source data, including:
acquiring remote sensing images, POI data, lamplight images and population data of cities;
land utilization classification is carried out on the remote sensing images, and the landscape disorder degree of the city is obtained;
evaluating the nuclear density of the city according to the POI data to obtain the POI nuclear density of the city;
determining the night light intensity of the city according to the light image;
determining single explanatory power degrees of the view disorder degree, the POI nuclear density and the night light intensity on population data respectively;
determining the weight of the landscape disorder degree, the POI nuclear density and the night lamplight intensity according to the proportion of the single interpretation effort degree to the sum of the interpretation effort degrees;
determining the comprehensive index of each region in the city according to the weight;
the regions of the city are divided into a plurality of categories according to the composite index, including city edge regions.
The extraction method for the edge region of the medium and small cities based on the multi-source data has the following advantages:
based on GF-2 image, POI data, NPP/VIIRS image and Woldpop multisource data, a city boundary extraction model of the medium and small city is constructed by taking the view disorder degree, POI nuclear density and night lamplight intensity as city characteristic factors. Through the test discovery of three small and medium cities of a Hantai area, a Shangzhou area and a Hanjingjingjingji area, the model provided by the application is obviously stronger than a landscape turbulence threshold method and a POI nuclear density breakpoint analysis method in the aspects of precision, detail and integrity, has higher universality and can meet the research requirements of the border area of the small and medium cities.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for extracting a border region of a medium-small city based on multi-source data according to an embodiment of the present application;
FIG. 2 is a schematic view of a landscape disorder of a Han dynasty region according to an embodiment of the present application;
FIG. 3 is a schematic diagram of core density at different bandwidths of a Hamming region according to an embodiment of the present disclosure;
fig. 4 is a schematic diagram of an urban border area extraction result of a han-tai area according to an embodiment of the present application;
fig. 5 is a schematic diagram of urban border area extraction results of a business state area and a han shou area according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
Fig. 1 is a flowchart of a method for extracting a border region of a medium-small city based on multi-source data according to an embodiment of the present application. The embodiment of the application provides a method for extracting a border region of a medium and small city based on multi-source data, which comprises the following steps:
s100, acquiring remote sensing images, POI data, lamplight images and population data of cities.
Illustratively, the data used in this application includes GF-2, POI, woldpop, NPP/VIIRS, and five types of administrative boundaries, with Table 1 showing specific information for the data. The remote sensing image obtained by the GF-2 satellite realizes sub-meter spatial resolution and multispectral comprehensive remote sensing data acquisition, and has the characteristics of high positioning accuracy, high spatial resolution, high time resolution and the like. The application adopts No. 4 GF-2 remote sensing images of 7 months of 2020 to cover a research area.
POIs are geographic objects that can be abstracted into points, especially some geographic entities that are closely related to people's lives. POI data in this application is derived from the Goldmap (https:// lbs. Amp. Com /).
The lamplight image is obtained by adopting NPP/VIIRS data, the NPP/VIIRS data is derived from the national geophysical data center (National Geophysical Data Center, NGDC) of the United states, and the NPP data is detected by a Suomi NPP satellite carried visible infrared imaging radiation instrument. The application selects the month data lamplight product of 7 months in 2020 provided by NGDC, and the spatial resolution is 500m.
The population data is Woldpop data. The application selects Woldpop data with 100m resolution in 7 months in 2020.
TABLE 1 data specific information
Figure SMS_1
S110, land utilization classification is carried out on the remote sensing images, and the landscape disorder degree of the city is obtained.
Illustratively, S110 specifically includes: performing land utilization classification on the remote sensing images by using an object-oriented SVM (Supported Vector Machine) classification method; and determining the landscape disorder degree according to the land utilization type.
The application adopts an object-oriented SVM classification method. The object-oriented classification method breaks through the limitation that a single pixel is used as a basic classification and processing unit in the traditional classification method, classifies images from an object level, and reduces the loss rate of semantic information contained in the traditional pixel-based classification method. The SVM is a classification algorithm based on the VC dimension theory of statistical learning theory and the structural risk minimization principle. Compared with a neural network or a traditional classification method based on statistics, the SVM controls the complexity of the model through the number of vectors, and the complexity of the model is not required to be controlled through reducing characteristic variables through dimension reduction processing, so that the SVM classifier does not lose characteristic information of ground object targets in the classification process, and the occurrence of some overfitting phenomena is reduced.
The object-oriented SVM classification method has the advantages of object-oriented multi-scale segmentation and SVM, firstly, multi-scale segmentation is carried out according to the properties of an object region on an image, not only the spectrum information of the image is considered, but also the characteristics such as textures, geometric shapes, space topological relations and the like are added, and then the training sample is utilized for carrying out support vector machine classification. The object-oriented SVM classification method has obvious advantages in the aspects of precision, generalization, high-dimensional data processing and the like, and is widely applied to remote sensing image classification application.
Further, the remote sensing image is preprocessed before land utilization classification is performed on the remote sensing image by using an object-oriented SVM classification method. The preprocessing comprises the operations of atmosphere correction, fusion, mosaic and cutting of the remote sensing image.
After the preprocessing of the remote sensing image is completed, land utilization classification can be carried out on the remote sensing image, and then the landscape disorder degree is determined. In particular, the landscape turbulence level may represent the degree of disruption and dispersion of urban landscapes, reflecting the heterogeneity and homogeneity of landscape space. The higher the heterogeneity of land utilization patches per unit area, the greater the degree of landscape turbulence. The urban and rural areas are single in land type, most of the areas are connected with construction lands or agricultural lands, and the landscape disorder degree is low. The urban border area is an active expansion band between urban landscapes and agricultural land, the land utilization types are various, and the landscape disorder degree is high. Thus, the extent of the urban border zone can be determined by the difference in landscape turbulence. The formula of the landscape disorder degree is as follows:
Figure SMS_2
wherein W is a landscape disorder value and X n Representing the ratio of the n-th type of land to the unit area in the unit area; n represents the total number of land utilization types per unit area.
And S120, evaluating the nuclear density of the city according to the POI data to obtain the POI nuclear density of the city.
Illustratively, S120 specifically includes: and evaluating the POI data by using a nuclear density evaluation tool to obtain the POI nuclear density.
The nuclear density analysis is commonly used for evaluating density values of point or line element neighborhood, simulates the spatial distribution situation of elements, and is widely applied to geospatial analysis research. The main principle is that in a certain bandwidth range, the estimated density of the element is reduced along with the increase of the distance, the core density of the element at the center position is the highest, and the bandwidth edge is 0. The nuclear density analysis follows the law of spatial correlation, and the closer the distance is, the greater the correlation of the ground features, the POI data also conform to the law. The formula of POI kernel density is as follows:
Figure SMS_3
wherein lambda is (s) Calculating the nuclear density of the POI in the s-th area, wherein r is the bandwidth set by the nuclear density function, and n' is the total of all elements participating in calculationQuantity d ls Is the distance value at POI points i and s,
Figure SMS_4
is a weight of the distance.
Further, the POI data is also preprocessed prior to evaluation of the POI data with the nuclear density evaluation tool. The preprocessing includes filtering and re-projecting the POI data.
S130, determining the night light intensity of the city according to the light images.
Illustratively, S130 specifically includes: and determining DN (Digital Number) value of each image in the lamplight images to obtain the night lamplight intensity.
Similarly, prior to determining DN values for each of the light images, the light images are also pre-processed, including re-projecting, denoising, and cropping the NPP/VIIRS data.
And S140, determining single explanatory power degrees of the view disorder degree, the POI nuclear density and the night light intensity on the population data respectively.
Illustratively, S140 specifically includes: and determining the single explanatory power degree of the landscape disorder degree, the POI nuclear density and the night lamplight intensity on the population data respectively by utilizing the difference and factor detection of the geographic detector.
The geographic detector is a set of statistical methods for detecting spatial dissimilarity, explaining the driving force behind it, including dissimilarity and factor detection, interaction detection, risk zone detection and ecological detection. The main principle of the geographic detector is that the research area is divided into a plurality of subareas, and if the sum of variances of the subareas is smaller than the total variance of the subareas, the spatial diversity exists; if the spatial distribution of the two variables tends to agree, there is a statistical correlation between the two. The geographic detector can evaluate the interaction relationship among space diversity, detection interpretation factors and analysis variables, and is widely applied to the fields of nature, environmental science, human health and the like. The dissimilarity and factor detection is to detect the spatial dissimilarity of the attribute Y and the interpretation power of a certain factor X to the attribute Y, and the q value is used for measurement. Population density in population data is closely related to urban development, and the application determines each factor weight according to landscape disorder degree, POI nuclear density and interpretation degree of night light intensity on the population data. The formula of q is:
Figure SMS_5
Figure SMS_6
wherein: l is layering of the data Y of the people mouth or the disturbance degree of the landscape, the nuclear density of POIs and the intensity X of night light, namely classification or partition; n (N) h And N' is the number of units of the h layer and the full region, sigma h 2 Sum sigma 2 The variances of the population data Y of the h layer and the whole region are respectively, and SSW and SST are respectively the sum of the variances in the layer and the sum of the variances of the whole region.
And S150, determining the weight of the landscape disorder degree, the POI nuclear density and the night light intensity according to the proportion of the single interpretation effort degree to the sum of the interpretation effort degrees.
S160, determining the comprehensive index of each area in the city according to the weight.
For example, a weighted sum of the landscape irregularity, POI kernel density, night light intensity, and corresponding weights may be employed as the composite index.
S170, dividing each area of the city into a plurality of categories according to the comprehensive index, wherein the categories comprise city edge areas.
For example, the natural breakpoint method may be used to divide each region of a city into a plurality of categories, including three categories, namely, a city core region, a city edge region, and a rural area abdominal region.
The natural breakpoint method is a map ranking algorithm that divides data into two groups, namely binary natural breakpoints. The natural breakpoint method considers that natural turning points and breakpoints exist between any number columns, and the study objects can be divided into groups with similar properties through the turning points and the breakpoints. The main principle of the algorithm is to cluster all the numerical values, and the condition of the clustering is that the inter-group variance is maximum and the intra-group variance is minimum. The process implemented by the natural breakpoint method is divided into three steps:
(1) Calculating the deviation square sum SDAM of the array average value:
Figure SMS_7
SDAM is the sum of squares of the deviations of the array average,
Figure SMS_8
mean values of the groups.
(2) For each region within the range, the sum of squares deviation SDCM of the class mean is calculated and the smallest one is found.
(3) Calculating a variance fitting goodness GVF:
GVF＝(SDAM-SDCM)/SDAM
the GVF has a value between 0 and 1, 1 indicating excellent fitting, and 0 indicating poor fitting. The variance fitting goodness is used for verifying whether the classification result achieves the grouping purpose of minimum intra-class difference and maximum inter-class difference.
After each area of the city is divided into a plurality of categories according to the comprehensive index, grid conversion vector and smoothing processing are carried out on the grid images obtained after category division, and a city edge area range is obtained.
Description of the experiment
Study area overview. The Han dynasty area is located in the center of basin in the south of the west, south of the Shaanxi province, in the south of the Ehan river, in the north of the England, in the Qinling province. The land features of the Han dynasty are high, low, and the north part belongs to the Qin Ling south slope mountain land, the altitude is 700-2000 meters, and the total area is 34%; the middle part is a hilly area with the altitude of 541-700 meters and the total area of 28 percent; the south is the Han river alluvial plain, accounting for 38 percent of the total area. The Han dynasty area is the biggest commodity distribution area in Shaanxi, is the core area of Qinling and Bashan, and has important economic and ecological values. Since town construction of the Han dynasty area is mainly concentrated in the south, the application forms a research area with administrative boundaries of 8 towns in the south.
Results and analysis. Landscape disturbance degree thresholding results: based on GF-2 remote sensing images pretreated in a research area, land utilization is divided into vegetation by utilizing an object-oriented SVM classification method(cultivated land, woodland, grassland), construction land, water and unused land, fig. 2 (a). As can be seen from the figure, the construction land is mainly distributed in the middle and south of the investigation region, the water body is distributed along the border of the south, the unused land is mainly distributed in the north, and the vegetation is mainly distributed in the northwest, northeast and southeast. Through statistics, the vegetation area of the research area is 80.34km 2 The construction land area is 60.83km 2 The water body area is 6.95km 2 The area of unused land is 3.31km 2 。
In order to highlight the ring layer structure of the landscape disorder degree of the research area, a 100m multiplied by 100m grid is constructed as a space calculation scale unit, the occupied area ratio of vegetation, construction land, water and unused land in the unit grid is calculated by using ArcGIS10.3 software, and finally the landscape disorder degree of the research area is calculated, and in fig. 2 (b). As can be seen from the figure, the landscape structure features of the urban core area are prominent, the landscape disorder degree is low, and a low-value area of the centralized connection is present. Through repeated experiments, the threshold value is smaller than 0.46 and is used as a mark for identifying the urban core area. However, when a large-scale green land exists in the urban core area, the landscape disorder is high, and the complete urban core area cannot be identified by the landscape disorder threshold method. The difference between the urban border area and the rural landscape disorder degree is not obvious, and the landscape disorder degree is high. Different from large cities, small and medium-sized cities have fewer population and smaller villages, and the villages are gathered in the Hanzhong plain at the places of the study areas and are scattered, so that the landscape disorder between the urban border areas and the rural areas is high. Although the landscape disorder degree threshold method is widely applied to the identification of the border region of the large city, the defects of discontinuous results, insufficient detail, poor universality and the like commonly existing in the single-factor threshold method are obviously amplified in the process of identifying the border region of the research region city. Therefore, the landscape disturbance degree thresholding method is not suitable for the identification of the border areas of small and medium cities.
Results of POI nuclear density breakpoint analysis: based on the POI data after pretreatment of the study area, POI nuclei Density was calculated using the Kernel Density tool of arcgis 10.3. The bandwidth in the nuclear density analysis has a critical effect on the results, and 500m, 1000m and 1500m are selected for the nuclear density analysis respectively after referring to the existing research results, and the results are shown in (a) - (c) in fig. 3. As can be seen from the graph, when the bandwidth is 500m, the nuclear density analysis result is relatively fine and discontinuous, and the overall distribution situation of the urban POI is not obvious; when the bandwidth is 1500m, the local characteristics of the urban POI overall distribution situation are difficult to develop, and the detail is insufficient; when the bandwidth is 1000m, the nuclear density analysis result has better stability, and meanwhile, the overall distribution situation is obvious, so that the analysis requirement of the border area of the research area city can be met.
In fig. 3, (d) is the result of classifying the POI core density of 1000m in the investigation region bandwidth into three types by using the natural breakpoint method. As can be seen from fig. 3 (b) and (d), the research area shows obvious ring layer structure distribution, the urban core area is distributed with large-area continuous high-density areas, the urban edge area has lower density, and the density of most rural areas is close to 0. Compared with the landscape disorder degree threshold method, the analysis result of the POI nuclear density can clearly and completely identify the urban core area, but the analysis result of the POI nuclear density is not ideal for the urban edge area. The POI nuclear density breakpoint analysis method can identify part of villages with better development and more POIs as urban border areas, meanwhile, the urban border areas are in a development stage, POI data are less, the updating speed is low, and therefore larger errors exist between results and actual urban border areas. Unlike major cities, the POI data integrity of the border areas of the middle and small cities is low and the updating is slow, and the single-factor POI nuclear density breakpoint analysis method is difficult to realize the accurate extraction of the border areas of the research areas.
Model results are extracted from the border areas of the medium and small cities: based on a middle and small city border region extraction model, firstly, calculating the landscape disorder degree, POI nuclear density and night light intensity of a research region; then, the weight (landscape disorder degree: 0.10, POI nuclear density: 0.51, night light intensity: 0.39) of each factor is determined by combining the geographic detector and the Woldpop data, so that a comprehensive index is constructed; and finally, identifying the urban border area by using a natural breakpoint method, and carrying out post-processing on the result. FIG. 4 (a) shows the result of classifying the composite index into three types by using the natural breakpoint method; fig. 4 (b) shows the final result of the model extraction in the border region of the medium and small city. As can be seen from the figure, the border area of the medium and small cities is liftedThe model can be taken to accurately and completely identify the border of the city of the research area. Urban border areas of the study area are mainly concentrated in the north and east, and the area is about 39km 2 . Compared with a single-factor landscape disorder threshold method and a POI nuclear density breakpoint analysis method, the performance of the small and medium city edge region extraction model is greatly improved, and particularly, the result of the city edge region outer boundary is obtained. The model and the POI kernel density breakpoint analysis method in the application extract the overall pattern of the inner boundary of the urban border area (urban core area) to be more consistent, but the detail of the model and the POI kernel density breakpoint analysis method is stronger in comparison with the model and the POI kernel density breakpoint analysis method. The difference of the results of the two is mainly concentrated in the southwest of a research area, and the southwest is a Han river new area and is still in a rapid development stage, the landscape pattern changes faster, the POI data update is slower, so that the error exists between the inner boundary of the urban border area extracted by the POI nuclear density breakpoint analysis method and the actual boundary. The model of the application focuses on comprehensive performance of regional landscape disorder, POI nuclear density and night light intensity, and has small dependence on single factor performance, so that the range of urban border areas can be accurately identified.
Model precision evaluation: because the range of the border area of the city of the research area is difficult to identify by the landscape disorder degree threshold method, the application only evaluates the precision of the POI nuclear density breakpoint analysis method and the extraction model of the border area of the city of the middle and small cities. Through detailed visual analysis, the extraction model of the edge region of the medium and small cities is obviously stronger than the POI nuclear density breakpoint analysis method in detail and integrity. In order to further evaluate the extraction precision of different methods, the application adopts two methods of field verification and landscape pattern index evaluation. The field verification is to uniformly select 100 sample points around the urban border area along the road edge, and in fig. 4 (b), the accuracy of the extracted result is analyzed by a field verification method, as shown in table 2. From the table, the overall accuracy of the extraction model of the edge region of the medium and small cities is obviously higher than that of the POI nuclear density breakpoint analysis method, and the total accuracy reaches 98%. The POI nuclear density breakpoint analysis method has more false lifting numbers, and the overall accuracy is only 67%.
Table 2 extraction accuracy by different methods
Figure SMS_9
Landscape pattern index is often used to evaluate the accuracy of extraction of urban border areas, and Plaque Density (PD) and shannon diversity index (SHDI) are chosen to evaluate the accuracy of both methods from grade and landscape levels. PD represents the degree of fragmentation of the landscape and SHDI represents the degree of richness and complexity of the landscape type. Typically, the PD and SHDI of urban border areas are higher, and the urban centers and rural areas are lower. Table 3 shows PD and SHDI values for different regions for two methods based on the Fragstats 4.2 software calculation. It can be seen from the table that in the urban border area, the PD and SHDI values of the urban border area extraction model are obviously higher than those of the POI nuclear density breakpoint analysis method, which shows that the degree of view fragmentation, complexity and diversity in the urban border area extracted by the method is higher than those of the urban border area extracted by the method, and the characteristics of the urban border area can be reflected. In rural areas, PD and SHDI values of the middle and small city edge region extraction model are obviously lower than those of the POI nuclear density breakpoint analysis method, and in the city core region, the PD and SHDI values are approximate because the boundary difference extracted by the two methods is smaller. Comprehensive, the model provided by the application has higher precision, and can realize the accurate extraction of the border area of the city in the research area.
TABLE 3 PD and SHDI values for different regions
Figure SMS_10
Model commonality analysis: in order to further verify the applicability of the model provided by the application in different areas and different types of small and medium cities, urban border areas are respectively identified in the Shangluo city and Hanzhu area of Ankang city. The business district belongs to a typical banded city structure and is obviously limited by resource conditions. As seen in fig. 5 (a), the POI kernel density breakpoint analysis method is incomplete in the edge region of the city of the region of the business, and mainly focuses on the southeast. The main reason is that southeast is an industrial park, the number of POIs is small and scattered, and the POIs are insufficient to support the identification of urban border areas. The Han-Bin area is separated from the middle of Han river, the southeast is mainly an old urban area, and the northwest is a new urban area, belonging to a multi-center urban structure. As can be seen from fig. 5 (b), the POI nuclear density breakpoint analysis method has poor extraction results on urban border areas of the han-bine area, especially in the southeast aged urban area. The method is mainly characterized in that the single factor method is used for extracting the edge regions of the multi-center cities, the data quality requirements are extremely high, the old urban areas are relatively backward in development, population distribution is concentrated, and POI data are not complete enough. In contrast, the middle and small city edge region extraction model can accurately and completely extract city edge regions of two regions. The model identifies urban border areas according to comprehensive performance differences of regional landscape disorder degree, POI nuclear density and night light intensity, has smaller dependence on single factor performance, and can be suitable for small and medium cities in different areas and different types.
While preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the application.
It will be apparent to those skilled in the art that various modifications and variations can be made in the present application without departing from the spirit or scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims and the equivalents thereof, the present application is intended to cover such modifications and variations.

Claims (8)

1. The method for extracting the edge region of the medium and small cities based on the multi-source data is characterized by comprising the following steps of:
acquiring remote sensing images, POI data, lamplight images and population data of cities;
land utilization classification is carried out on the remote sensing images, and the landscape disorder degree of the city is obtained;
evaluating the nuclear density of the city according to the POI data to obtain the POI nuclear density of the city;
determining the night light intensity of the city according to the light image;
determining single explanatory degrees of the view disorder degree, the POI nuclear density and the night light intensity on the population data respectively;
determining the weight of the landscape disorder degree, the POI nuclear density and the night light intensity according to the proportion of the single interpretation effort degree to the sum of the interpretation effort degrees;
determining the comprehensive index of each region in the city according to the weight;
and dividing each region of the city into a plurality of categories according to the comprehensive index, wherein the categories comprise city edge regions.
2. The method for extracting the border region of the medium and small cities based on the multi-source data according to claim 1, wherein the step of classifying the land utilization of the remote sensing image to obtain the landscape disorder of the cities comprises the following steps:
performing land utilization classification on the remote sensing images by using an object-oriented SVM classification method;
and determining the landscape disorder degree according to the land utilization type.
3. The method for extracting border areas of small and medium cities based on multi-source data according to claim 2, wherein before land utilization classification is performed on the remote sensing images by using an object-oriented SVM classification method, the method further comprises:
and preprocessing the remote sensing image.
4. The method for extracting border areas of small and medium cities based on multi-source data according to claim 1, wherein the step of evaluating the nuclear density of the cities according to the POI data to obtain the POI nuclear density of the cities comprises the following steps:
and evaluating the POI data by using a nuclear density evaluation tool to obtain the POI nuclear density.
5. The method for extracting a border region of a medium and small city based on multi-source data according to claim 4, further comprising, prior to evaluating the POI data using a nuclear density evaluation tool:
and preprocessing the POI data.
6. The method for extracting border areas of small and medium cities based on multi-source data as set forth in claim 1, wherein the determining the night light intensity of the cities according to the light images includes:
and determining DN value of each image in the lamplight images to obtain the night lamplight intensity.
7. The method for extracting border areas of small and medium cities based on multi-source data as set forth in claim 1, wherein said determining a single degree of interpretation of said demographic data by said degree of landscape turbulence, POI kernel density and night light intensity, respectively, comprises:
and determining the single explanatory degree of the landscape disorder degree, the POI nuclear density and the night light intensity on the population data respectively by utilizing the difference and factor detection of the geographic detector.
8. The method for extracting border areas of small and medium cities based on multi-source data as set forth in claim 1, further comprising, after dividing each area of the cities into a plurality of categories according to the composite index:
and carrying out grid conversion vector and smoothing on the grid images obtained after category division to obtain the range of the urban border area.
CN202211565899.2A 2022-12-07 2022-12-07 Multi-source data-based extraction method for edge regions of medium and small cities Pending CN116310853A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211565899.2A CN116310853A (en) 2022-12-07 2022-12-07 Multi-source data-based extraction method for edge regions of medium and small cities

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211565899.2A CN116310853A (en) 2022-12-07 2022-12-07 Multi-source data-based extraction method for edge regions of medium and small cities

Publications (1)

Publication Number Publication Date
CN116310853A true CN116310853A (en) 2023-06-23

Family

ID=86813778

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211565899.2A Pending CN116310853A (en) 2022-12-07 2022-12-07 Multi-source data-based extraction method for edge regions of medium and small cities

Country Status (1)

Country Link
CN (1) CN116310853A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116645012A (en) * 2023-07-27 2023-08-25 河北工业大学 High-precision dynamic identification method for spatial range of urban border area
CN117710746A (en) * 2023-12-22 2024-03-15 中国科学院地理科学与资源研究所 Multi-scale landscape function main body type identification method integrating open geographic data
CN118115872A (en) * 2024-03-04 2024-05-31 中国科学院地理科学与资源研究所 Identification method for north-shift boundary of corn planting zone

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116645012A (en) * 2023-07-27 2023-08-25 河北工业大学 High-precision dynamic identification method for spatial range of urban border area
CN116645012B (en) * 2023-07-27 2023-10-10 河北工业大学 High-precision dynamic identification method for spatial range of urban border area
CN117710746A (en) * 2023-12-22 2024-03-15 中国科学院地理科学与资源研究所 Multi-scale landscape function main body type identification method integrating open geographic data
CN118115872A (en) * 2024-03-04 2024-05-31 中国科学院地理科学与资源研究所 Identification method for north-shift boundary of corn planting zone
CN118115872B (en) * 2024-03-04 2024-08-06 中国科学院地理科学与资源研究所 Identification method for north-shift boundary of corn planting zone

Similar Documents

Publication Publication Date Title
CN111666918B (en) Coastline change identification method based on multiple factors
Zhu et al. Flood disaster risk assessment based on random forest algorithm
CN116310853A (en) Multi-source data-based extraction method for edge regions of medium and small cities
CN102750703B (en) Remote sensing image airport automatic detecting method based on linear cutting
CN110929939B (en) Landslide hazard susceptibility spatial prediction method based on clustering-information coupling model
Huang et al. Modelling landslide susceptibility prediction: a review and construction of semi-supervised imbalanced theory
CN111062446A (en) Land type classification method based on multi-source homeland resource data
Gong et al. Extracting tidal creek features in a heterogeneous background using Sentinel-2 imagery: a case study in the Yellow River Delta, China
Xue et al. Water body automated extraction in polarization SAR images with dense-coordinate-feature-concatenate network
Wang et al. A framework for fine classification of urban wetlands based on random forest and knowledge rules: taking the wetland cities of Haikou and Yinchuan as examples
Ma et al. A multisource remotely sensed data oriented method for “ghost city” phenomenon identification
Liu et al. Spatial-temporal hidden Markov model for land cover classification using multitemporal satellite images
Aahlaad et al. An object-based image analysis of worldview-3 image for urban flood vulnerability assessment and dissemination through ESRI story maps
Peng et al. Detailed wetland-type classification using Landsat-8 time-series images: a pixel-and object-based algorithm with knowledge (POK)
Zhang et al. Extracting physical urban areas of 81 major Chinese cities from high-resolution land uses
CN105701856B (en) A kind of vegetation extracting method and system
Lu et al. Mapping the terraces on the Loess Plateau based on a deep learning-based model at 1.89 m resolution
Han et al. A method of the coverage ratio of street trees based on deep learning
Chen et al. Built-Up Area Extraction Combing Densely Connected Dual-Attention Network and Multiscale Context
Deng et al. Automated and refined wetland mapping of Dongting Lake using migrated training samples based on temporally dense Sentinel 1/2 imagery
CN109543729A (en) Time series data land cover classification method based on characteristic parameter cluster
CN111401683B (en) Method and device for measuring tradition of ancient villages
Liu et al. Impervious surface Mapping and its spatial–temporal evolution analysis in the Yellow River Delta over the last three decades using Google Earth Engine
CN116205522B (en) Landslide vulnerability evaluation method and system based on multidimensional CNN coupling
Chen et al. A novel water change tracking algorithm for dynamic mapping of inland water using time-series remote sensing imagery

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination