City contraction identification method based on time sequence population data change point detection

Abstract

The embodiment of the invention discloses a city shrinkage identification method based on time sequence population data change point detection, and relates to the technical field of geographic information, and the method comprises the steps: obtaining land utilization space raster data and population space raster data; respectively converting the land utilization space raster data and the population space raster data into matrixes; determining urban land use grids based on the land utilization space matrix, extracting population data from the population space matrix, constructing a one-dimensional population array, and performing smoothing processing to obtain smoothed time series data; identifying change trend abrupt change nodes of the population data based on the smoothed time sequence data, and segmenting each one-dimensional population array; and identifying a population data change trend of the segmented array, and determining a city contraction grid. According to the method, time periods are automatically divided by using change point detection, the starting and ending time of urban contraction can be more objectively identified, and data support can be provided for future urban land evolution simulation and regional development policy formulation.

Description

Technical Field

[0001] The present application relates to the field of geographic information technology, and in particular to a method for identifying urban shrinkage based on change point detection of time series population data. Background Art

[0002] Urban shrinkage is a phenomenon in which urban population declines and the economy declines when cities develop to a certain stage. Population decline is still generally regarded as the core sign of urban shrinkage. When cities defined by administrative boundaries are used as research objects, using population decline to describe urban shrinkage may include rural population in the analysis, which may easily lead to biased results. Currently, urban shrinkage is generally described by population decline within natural cities. This method still has the following shortcomings:

[0003] (1) Insufficient spatial accuracy in identifying urban shrinkage. As an important area of ​​urban research, urban land evolution simulation usually uses grids or patches as basic units to display urban processes in space. However, when natural cities are used as the main body to identify urban shrinkage, this may limit the combination of urban shrinkage and urban land evolution spatial display simulation.

[0004] (2) The division of research periods is relatively subjective. At present, most relevant technologies tend to divide the research period into several target periods when identifying urban shrinkage within a certain period of time, and identify the urban status in each target period. However, this time division method is usually relatively subjective, and can be divided into five-year or ten-year time intervals based on data availability. This method is prone to ignoring the key time points when the population change trend actually changes, resulting in inaccurate identification of urban shrinkage periods.

[0005] In recent years, with the availability of high-resolution geographic data, some related technologies have attempted to use annual population data or night light data to identify urban shrinkage trends. However, this method usually only fits the overall population changes through a straight line, ignoring the dynamic changes in population trends during the study period.

[0006] Therefore, there is currently a lack of a method that can accurately identify the location and time interval where the urban shrinkage effect occurs. Summary of the invention

[0007] The embodiment of the present application provides a method for identifying urban shrinkage based on time series population data change point detection to solve the defects in the related technology. The present application uses the annual population grid data to detect the change points of the time series data through the Pelt algorithm, divides the research period according to the change points, and judges the change trend of each time period through regression fitting, thereby identifying the shrinkage grid and the start and end years of its shrinkage, which can more finely understand the spatial imbalance phenomenon in the process of urban development, and can provide data support for the future simulation of urban land evolution and the formulation of regional development policies. The technical solution is as follows:

[0008] In a first aspect, an embodiment of the present application provides a method for identifying urban shrinkage based on change point detection of time series population data, characterized by comprising:

[0009] S1 obtains the land use spatial raster data of the study area in different periods and the population spatial raster data of the corresponding periods;

[0010] S2 converts the land use space raster data into the land use space matrix, and converts the population space raster data into the population space matrix, the matrix elements correspond to the cells, and the matrix space corresponds to the cell space;

[0011] S3 determines the urban land grids based on the land use spatial matrix of each period, extracts the population data of each urban land grid in each period from the population spatial matrix, and constructs a one-dimensional population array of each urban land grid;

[0012] S4 performs smoothing processing on each one-dimensional population array to obtain smoothed time series data;

[0013] S5 identifies the mutation nodes of the population data change trend in each one-dimensional population array based on the smoothed time series data, segments each one-dimensional population array based on the mutation nodes of the change trend, and re-determines the start time node and the end time node of the new one-dimensional population array;

[0014] S6 identifies the population data change trend of the segmented new one-dimensional population array, and determines the location of the urban shrinkage grid and the corresponding start time node and end time node.

[0015] In an optional solution of the first aspect, step S1 includes the following steps:

[0016] S101 reclassifies the land use spatial raster data of multiple different periods to determine the urban land raster and other land use type raster;

[0017] S102 reprojects the land use space raster data and the population space raster data of the same period so that the coordinate projections of the land use space raster data and the population space raster data are consistent;

[0018] S103 resamples the land use spatial raster data of each period to make the grid size of the land use spatial raster data consistent with that of the population spatial raster data.

[0019] In an optional solution of the first aspect, step S2 includes the following steps:

[0020] S201 determines the land use type corresponding to each grid in the land use spatial grid data, assigns a value to each land use type, and obtains the land use spatial matrix;

[0021] S202 determines the population data corresponding to each grid in the population space grid data of each period, and obtains the population space matrix corresponding to the population space grid data of each period, wherein each element in each of the population space matrices represents the population data of the corresponding grid in the corresponding period.

[0022] In an optional solution of the first aspect, step S3 includes the following steps:

[0023] S301 searches for town land grids in the land use space matrix of each period according to the values ​​corresponding to the town land types, and determines the duration of each town land grid;

[0024] S302 extracts the population data of each town land grid in the corresponding period from the population space matrix corresponding to the continuous period, arranges the population data of each period in chronological order, and constructs a one-dimensional population array of each town land grid.

[0025] In an optional solution of the first aspect, step S4 includes the following steps:

[0026] Sliding the one-dimensional population array based on a preset sliding window, and generating smoothed time series data based on the mean of population data calculated by sliding the preset sliding window each time;

[0027] The mean of the population data in the preset sliding window is calculated based on the sliding window smoothing method, using the formula:

[0028]

[0029] Wherein, P(t) is the population data after smoothing, g(·) is the population data before smoothing, the size of the preset sliding window is 2n+1, t is the middle point of the preset sliding window, g(t) is the population data corresponding to the middle point, and i is the displacement from the middle point.

[0030] In an optional solution of the first aspect, step S5 includes the following steps:

[0031] S501 constructs a cost function based on the radial basis function:

[0032] K(y s,t )=exp(-σ||y s -y t || 2 );

[0033]

[0034] Among them, K(y s,t ) is the similarity between the data at time s and the data at time t, and its value range is [0,1], ||y s -y t || represents the Euclidean distance between two data points, σ ​​represents the bandwidth, which controls the decay rate of the similarity between two data points as the distance increases; c(y s,t ) is the cost function of the data at time s and the data at time t, b and a are parameters;

[0035] S502 constructs an objective function according to the cost function, and the optimization goal is to minimize the objective function, including:

[0036] F(t)=Z(s)+c(y s,t )+β;

[0037] Where F(t) is the objective function when data point t is the split point, Z(s) is the objective function value when feasible solution s is the split point, and when the feasible solution set is empty, Z(s) is 0; β is the penalty term, which is used to control the number of identified change points;

[0038] S503 updates the feasible solution set in each cycle of solving the objective function, adds the current time t to the feasible solution set, and filters out the time s in the feasible solution set that does not meet the following conditions:

[0039] z(s)+c(y s,t )≤Z(t);

[0040] S504 determines the mutation node of the population data change trend according to the solution result of the objective function, segments the time series data at each position, segments each one-dimensional population array based on the mutation node of the change trend, and re-determines the start time node and end time node of the new one-dimensional population array.

[0041] In an optional solution of the first aspect, step S6 includes the following steps:

[0042] Based on the new one-dimensional population array after the above segmentation and the start time node and end time node corresponding to the new one-dimensional population array, regression fitting is performed, and the formula is applied:

[0043] P t =kt′+b;

[0044] The urban shrinkage raster is determined based on the following formula:

[0045]

[0046] If the output type of f(k,ΔP) is contraction, the corresponding grid is determined to be an urban contraction grid, and the rest are non-contracting grids, and the urban contraction grid is output;

[0047] Among them, t′ is the time, P t is the population data at time t, k is the slope of the fitting line, b is the intercept, and the trend of population data change in the corresponding grid is judged based on the slope k of the fitting line: when k>0, the population in the grid is on an upward trend, when k<0, the population in the grid is on a downward trend, and ΔP is the rate of change of population data.

[0048] In a second aspect, the embodiment of the present application further provides an urban shrinkage identification device based on time series population data change point detection, comprising:

[0049] The data acquisition module is used to obtain the land use spatial raster data of the study area at different periods and the population spatial raster data of the corresponding periods;

[0050] The data processing module is used to convert the land use space grid data into the land use space matrix, and convert the population space grid data into the population space matrix, where the matrix elements correspond to the cells and the matrix space corresponds to the cell space;

[0051] The data processing module is also used to determine the town land grids based on the land use space matrix of each period, extract the population data of each town land grid in each period from the population space matrix, and construct a one-dimensional population array of each town land grid;

[0052] The data processing module is also used to perform smoothing processing on each one-dimensional population array to obtain smoothed time series data;

[0053] The calculation module is used to identify the change trend mutation nodes of the population data in each one-dimensional population array based on the smoothed time series data, segment each one-dimensional population array based on the change trend mutation nodes, and re-determine the start time node and end time node of the new one-dimensional population array;

[0054] The calculation module is also used to identify the population data change trend of the new one-dimensional population array after segmentation, and determine the position of the urban shrinkage grid and the corresponding start time node and end time node.

[0055] In a third aspect, an embodiment of the present application further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method provided in the first aspect of the embodiment of the present application or any one of the implementation methods of the first aspect is implemented.

[0056] In a fourth aspect, the present application also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method provided by the first aspect of the embodiment of the present application or any one of the implementations of the first aspect.

[0057] The beneficial effects brought about by the technical solutions provided by some embodiments of the present application include at least:

[0058] The embodiment of the present application provides a method for identifying urban shrinkage based on time series population data change point detection. Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0059] (1) This application selected population decline as the core indicator to ensure that the spatial and temporal resolution of the data is fine enough, the spatial resolution of the population spatial grid data can match the basic unit of urban expansion simulation, and the annual data is easy to obtain. The results of the identification can be better combined with the urban expansion simulation study.

[0060] (2) This application uses change point detection to automatically divide time periods, more objectively identifying the start and end time of urban shrinkage, determining the nodes for time period division by searching for mutation points in population time series data, identifying the population change trend in each time period after division, and thus obtaining the start and end time of urban shrinkage grid. This provides support for a more detailed understanding of the spatial imbalance phenomenon in the process of urban development, and provides reliable data support for future urban land evolution simulation and regional development policy formulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solutions in the present application or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0062] Figure 1 It is a flow chart of a method for identifying urban shrinkage based on time series population data change point detection provided by an embodiment of the present application;

[0063] Figure 2This is a schematic diagram of the influence of window size in data smoothing based on time series population data change point detection provided by an embodiment of the present application;

[0064] Figure 3 It is a schematic diagram of the proportion of shrinkage grids in each city in a city shrinkage identification method based on time series population data change point detection provided in an embodiment of the present application;

[0065] Figure 4 It is a schematic diagram of the development status of each city in a city shrinkage identification method based on time series population data change point detection provided in an embodiment of the present application;

[0066] Figure 5 It is a structural schematic diagram of a device for identifying urban shrinkage based on time series population data change point detection provided by an embodiment of the present application;

[0067] Figure 6 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0069] The terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device including a series of steps or modules is not limited to the listed steps or modules, but may optionally include steps or modules that are not listed, or may optionally include other steps or modules that are inherent to these processes, methods, products or devices.

[0070] It should be noted that the terms "first\second" involved in the present application are only used to distinguish similar objects, and do not represent a specific order for the objects. It is understandable that "first\second" can be interchanged with a specific order or sequence where permitted. It should be understood that the objects distinguished by "first\second" can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those described or illustrated herein.

[0071] The present application is described in detail below with reference to specific embodiments.

[0072] Next, combine Figure 1, introduces a method for identifying urban shrinkage based on time series population data change point detection provided by the embodiment of the present application. For details, please refer to Figure 1 , Figure 1 FIG. 1 is a flow chart of a method for identifying urban shrinkage based on time series population data change point detection provided by an embodiment of the present application. Figure 1 As shown, the method comprises the following steps:

[0073] S1 obtains the land use spatial raster data of the study area in different periods and the population spatial raster data of the corresponding periods;

[0074] S2 converts the land use space raster data into the land use space matrix, and converts the population space raster data into the population space matrix, the matrix elements correspond to the cells, and the matrix space corresponds to the cell space;

[0075] S3 determines the urban land grids based on the land use spatial matrix of each period, extracts the population data of each urban land grid in each period from the population spatial matrix, and constructs a one-dimensional population array of each urban land grid;

[0076] S4 performs smoothing processing on each one-dimensional population array to obtain smoothed time series data;

[0077] S5 identifies the mutation nodes of the population data change trend in each one-dimensional population array based on the smoothed time series data, segments each one-dimensional population array based on the mutation nodes of the change trend, and re-determines the start time node and the end time node of the new one-dimensional population array;

[0078] S6 identifies the population data change trend of the segmented new one-dimensional population array, and determines the location of the urban shrinkage grid and the corresponding start time node and end time node.

[0079] Specifically, step S1 includes the following steps:

[0080] S101 reclassifies the land use spatial raster data of multiple different periods to determine the urban land raster and other land use type raster;

[0081] S102 reprojects the land use space raster data and the population space raster data of the same period so that the coordinate projections of the land use space raster data and the population space raster data are consistent;

[0082] S103 resamples the land use spatial raster data of each period to make the grid size of the land use spatial raster data consistent with that of the population spatial raster data.

[0083] Specifically, the Beijing-Tianjin-Hebei region can be used as a research area for urban shrinkage identification. The land use spatial raster data and the population spatial raster data of the corresponding years in 2000, 2005, 2010, 2015, 2020, and 2022 are selected. Each grid in the land use spatial raster data corresponds to the land use type of the year, such as urban land, expandable land, and water area, and each grid in the population spatial raster data corresponds to the population data of the year.

[0084] In some embodiments, step S2 includes the following steps:

[0085] S201 determines the land use type corresponding to each grid in the land use spatial grid data, assigns a value to each land use type, and obtains the land use spatial matrix;

[0086] For example, the values ​​of the elements of the matrix include: urban land is 1, expandable land is 0, and water area is 2, and the value of each element reflects its corresponding land use type. This embodiment of the application is not limited to this.

[0087] S202 determines the population data corresponding to each grid in the population space grid data of each period, and obtains the population space matrix corresponding to the population space grid data of each period, wherein each element in each of the population space matrices represents the population data of the corresponding grid in the corresponding period.

[0088] For example, the population spatial grid data of each period can be converted into a series of population quantity matrices A, B, C, ..., N, etc. The element value in each matrix A, B, C, ..., N represents the population quantity in each grid cell in a specific period.

[0089] In some embodiments, step S3 includes the following steps:

[0090] S301 searches for town land grids in the land use space matrix of each period according to the values ​​corresponding to the town land types, and determines the duration of each town land grid;

[0091] S302 extracts the population data of each town land grid in the corresponding period from the population space matrix corresponding to the continuous period, arranges the population data of each period in chronological order, and constructs a one-dimensional population array of each town land grid.

[0092] Specifically, since the grid size and coordinate projection are consistent in step S1, the land use space matrix and the population space matrix can be stacked into a high-dimensional tensor according to the spatial dimension. Specifically, the value corresponding to the urban land type can be searched in the land use state matrix, for example, the value corresponding to the urban land type is 1, and the spatial coordinates (x, y) of the urban land grid can be obtained. The population data of the grid at the same position in the population space matrix of each period can be obtained accordingly, thereby realizing the extraction of population data of each urban land grid.

[0093] Furthermore, the urban land grid can be updated according to the land use spatial matrix of the next period as time progresses. For the newly added urban land location, the start time of the conversion to urban land can be determined, and the end time of the conversion of part of the urban land in the previous period to other land types can also be determined. In this way, the duration of each urban land grid can be determined, and the population data of the corresponding time point can be extracted from the population spatial matrix of the corresponding period, so as to continuously update the one-dimensional population array of each urban land grid.

[0094] For example, if a grid is an urban land grid from 2000 to 2022, the population data corresponding to the grid from 2000 to 2022 is extracted, and each data is arranged in chronological order, then the one-dimensional population array corresponding to the grid can be obtained. At this time, the one-dimensional population array includes 23 population data, and the data length of the array can be recorded as 23.

[0095] In some embodiments, step S4 includes the following steps:

[0096] Sliding the one-dimensional population array based on a preset sliding window, and generating smoothed time series data based on the mean of population data calculated by sliding the preset sliding window each time;

[0097] The mean of the population data in the preset sliding window is calculated based on the sliding window smoothing method, using the formula:

[0098]

[0099] Wherein, P(t) is the population data after smoothing, g(·) is the population data before smoothing, the size of the preset sliding window is 2n+1, t is the middle point of the preset sliding window, g(t) is the population data corresponding to the middle point, and i is the displacement from the middle point.

[0100] It should be noted that sensitivity analysis can be performed on different window sizes, such as Figure 2As shown, in this embodiment, when the sliding window size is preset to 3, the interference of noise on mutation point detection can be effectively reduced while retaining key changes; when the window size is 5, although key changes can be retained, data will be lost, which will make the time series data with a data length of 8 unreliable; when the window size is 7, more data will be lost, and the characteristics of key changes are not well retained. Therefore, it is preferred to preset the sliding window size to 3.

[0101] In some embodiments, step S5 includes the following steps:

[0102] S501 constructs a cost function based on the radial basis function:

[0103] K(y s,t )=exp(-σ||y s -y t || 2 );

[0104]

[0105] Among them, K(y s,t ) is the similarity between the data at time s and the data at time t, and its value range is [0,1], ||y s -y t || represents the Euclidean distance between two data points, σ ​​represents the bandwidth, which controls the decay rate of the similarity between two data points as the distance increases; c(y s,t ) is the cost function of the data at time s and the data at time t, b and a are parameters used to evaluate the consistency within the time period; the cost function ensures the comparability of cost functions between time periods of different lengths by introducing the (b–a) term and multiplying the sum of the similarities of each point in the time period by a normalization term 1 / (b–a).

[0106] S502 constructs an objective function according to the cost function, and the optimization goal is to minimize the objective function, including:

[0107] F(t)=Z(s)+c(y s,t )+β;

[0108] Where F(t) is the objective function when data point t is the split point, Z(s) is the objective function value when feasible solution s is the split point, and when the feasible solution set is empty, Z(s) is 0; β is the penalty term, which is used to control the number of identified change points;

[0109] S503 updates the feasible solution set in each cycle of solving the objective function, adds the current time t to the feasible solution set, and filters out the time s in the feasible solution set that does not meet the following conditions:

[0110] Z(s)+c(y s,t )≤Z(r);

[0111] S504 determines the mutation node of the population data change trend according to the solution result of the objective function, segments the time series data at each position, segments each one-dimensional population array based on the mutation node of the change trend, and re-determines the start time node and end time node of the new one-dimensional population array.

[0112] In some embodiments, step S6 includes the following steps:

[0113] Based on the new one-dimensional population array after the above segmentation and the start time node and end time node corresponding to the new one-dimensional population array, regression fitting is performed, and the formula is applied:

[0114] P t =kt′+b;

[0115] The urban shrinkage raster is determined based on the following formula:

[0116]

[0117] If the output type of f(k,ΔP) is contraction, the corresponding grid is determined to be an urban contraction grid, and the rest are non-contracting grids, and the urban contraction grid is output;

[0118] Among them, t′ is the time, P t is the population data at time t, k is the slope of the fitting line, b is the intercept, and the trend of population data change in the corresponding grid is judged based on the slope k of the fitting line: when k>0, the population in the grid is on an upward trend, when k<0, the population in the grid is on a downward trend, and ΔP is the rate of change of population data.

[0119] In some embodiments, after S4, the data length of each one-dimensional population array can be determined. If the data length of the one-dimensional population array is greater than a preset threshold, step S5 is executed to segment the one-dimensional population array, and step S6 is executed to perform regression fitting based on the segmented one-dimensional population array. Otherwise, step S6 is executed directly to perform regression fitting based on the unsegmented one-dimensional population array.

[0120] In a specific embodiment, the data used in this application includes population data and land use data, wherein the population data of the Beijing-Tianjin-Hebei region from 2000 to 2022 is used to determine the population change trend and identify urban shrinkage, and the land use data of the Beijing-Tianjin-Hebei region from 2000 to 2020 is used to determine the scope of urban land. The specific situation of these two types of data is as follows:

[0121] The population data used in the embodiments of the present application can be obtained from the LandScan population data set. In order to further ensure the accuracy and representativeness of this data set, the embodiments of the present application can aggregate the population data of 2010 and 2020 to the city scale and verify it with the statistical results of the census to determine whether it is reliable to use the LandScan population data to identify urban shrinkage.

[0122] The land use data used in the embodiments of this application can be obtained from the Resource and Environmental Data Center of the Chinese Academy of Sciences. The data set is based on Landsat remote sensing image data and is built in combination with field surveys. It includes 6 primary categories and 25 secondary categories with a resolution of 30m. Based on the availability of data and the research period, the embodiments of this application selected land use data from 2000 to 2020 at intervals of 5 years, and regarded urban land in the secondary category as urban land.

[0123] Since the projection coordinate system and resolution of population data and land use data are different, the embodiment of the present application also performs reprojection and resampling operations on population data and land use data. First, the population data is reprojected to be the same as the land use data, and then the land use data is resampled to be consistent with the resolution of the reprojected population data. Cities are characterized by high population density. In a grid with both urban land and non-urban land, urban land often contributes more population. Therefore, the resampling principle of the embodiment is that as long as there is urban land in the grid, the grid attribute is urban land.

[0124] All data preprocessing, data smoothing, trend mutation node detection and contraction identification algorithms of the embodiments of the present application can be implemented in the Python 3.11 environment, and Python libraries such as Numpy, Geopandas and Gdal can be used.

[0125] The embodiment of the present application identifies the urban shrinkage at the grid scale in the Beijing-Tianjin-Hebei region from 2000 to 2022, and conducts preliminary statistics on these grids. Figure 3 The figure shows the proportion of grids that have been shrinking since 2000 to the amount of urban land in 2000. The figure shows that in 2000, more than 20% of urban land in all cities in the Beijing-Tianjin-Hebei region shrank, and the shrinkage area in Tangshan and Handan even exceeded 40%. By 2022, the proportion of shrinking area in all cities has decreased, and only four cities still have a shrinkage area of ​​more than 30%, among which Handan has the largest decline, while the shrinkage area in Qinhuangdao has dropped to less than 20%.

[0126] The embodiment of the present application identifies the shrinkage and growth patterns of cities at intervals of 5 years, and identifies the shrinkage and growth patterns of cities in four periods from 2000 to 2020, such as Figure 4 As shown, from 2000 to 2020, a total of four shrinkage and growth patterns were identified in the Beijing-Tianjin-Hebei region, namely, densification (increase in urban population density), shrinkage (decrease in urban population density and decrease in total urban population), sprawl (decrease in urban population density but increase or remain unchanged in total urban population) and reasonable growth (urban population density remains unchanged but total urban population increases).

[0127] from Figure 4 It can be seen that from 2000 to 2005, the cities in the Beijing-Tianjin-Hebei region were mainly of the spreading and shrinking types, with only Shijiazhuang being of the densification type; from 2005 to 2010, most cities were of the spreading type, with only Baoding being of the reasonable growth type, and Cangzhou being of the densification type; from 2010 to 2015, the cities were mainly of the densification and spreading types; from 2015 to 2020, there were mainly spreading and shrinking types, with only Hengshui being of the reasonable growth type. In the four periods, the cities in the Beijing-Tianjin-Hebei region were mainly of the three types of spreading, shrinking and densification, with reasonable growth appearing less frequently. Most cities experienced a densification process from 2010 to 2015, indicating that the phenomenon of urban land expansion being faster than population growth during this period was effectively curbed.

[0128] The following are device embodiments of the present application, which can be used to execute the method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

[0129] See next Figure 5 , is a schematic diagram of the structure of an urban shrinkage identification device based on time series population data change point detection provided by an exemplary embodiment of the present application. The device can be implemented as all or part of a terminal through software, hardware, or a combination of both, and can also be integrated on a server as an independent module. The device 50 includes a data acquisition module 501, a data processing module 502, and a calculation module 503, wherein:

[0130] The data acquisition module 501 is used to acquire the land use spatial grid data of the study area in multiple different periods and the population spatial grid data of the corresponding period;

[0131] The data processing module 502 is used to convert the land use space grid data into the land use space matrix, and convert the population space grid data into the population space matrix, the matrix elements correspond to the cells, and the matrix space corresponds to the cell space;

[0132] The data processing module 502 is also used to determine the town land grids based on the land use space matrix of each period, extract the population data of each town land grid in each period from the population space matrix, and construct a one-dimensional population array of each town land grid;

[0133] The data processing module 502 is also used to perform smoothing processing on each one-dimensional population array to obtain smoothed time series data;

[0134] The calculation module 503 is used to identify the mutation nodes of the change trend of the population data in each one-dimensional population array based on the smoothed time series data, segment each one-dimensional population array based on the mutation nodes of the change trend, and re-determine the start time node and the end time node of the new one-dimensional population array;

[0135] The calculation module 503 is also used to identify the population data change trend of the new one-dimensional population array after segmentation, and determine the position of the urban shrinkage grid and the corresponding start time node and end time node.

[0136] It should be noted that the device 50 provided in the above embodiment only uses the division of the above functional modules as an example when executing the urban shrinkage identification method based on time series population data change point detection. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device provided in the above embodiment and the embodiment of the urban shrinkage identification method based on time series population data change point detection belong to the same concept. The implementation process is detailed in the method embodiment, which will not be repeated here.

[0137] An embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any of the above-mentioned method embodiments when executing the program.

[0138] See also Figure 6 , which is a structural block diagram of an electronic device provided in an embodiment of the present application.

[0139] like Figure 6 As shown, the electronic device 600 includes: a processor 601 and a memory 602 .

[0140] In the embodiment of the present application, the processor 601 is the control center of the computer system, which can be a processor of a physical machine or a processor of a virtual machine. The processor 601 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 601 may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array).

[0141] The processor 601 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also called a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state.

[0142] The memory 602 may include one or more computer-readable storage media, which may be non-transitory. The memory 602 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In some embodiments of the present application, the non-transitory computer-readable storage medium in the memory 602 is used to store at least one instruction, which is used to be executed by the processor 601 to implement the method in the embodiment of the present application.

[0143] In some embodiments, the electronic device 600 further includes: a peripheral device interface 603 and at least one peripheral device 604. The processor 601, the memory 602 and the peripheral device interface 603 can be connected via a bus or a signal line. Each peripheral device 604 can be connected to the peripheral device interface 603 via a bus, a signal line or a circuit board. Specifically, the peripheral device 604 includes: a display screen, a camera and an audio circuit. The peripheral device interface 603 can be used to connect at least one peripheral device related to I / O (Input / Output) to the processor 601 and the memory 602.

[0144] In some embodiments of the present application, the processor 601, the memory 602, and the peripheral device interface 603 are integrated on the same chip or circuit board; in some other embodiments of the present application, any one or two of the processor 601, the memory 602, and the peripheral device interface 603 can be implemented on a separate chip or circuit board. This embodiment of the present application does not specifically limit this.

[0145] The electronic device structure block diagram shown in the embodiment of the present application does not constitute a limitation on the electronic device 600. The electronic device 600 may include more or fewer components than shown in the figure, or combine certain components, or adopt a different component arrangement.

[0146] The embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps of the method of any of the above embodiments are implemented. The computer-readable storage medium may include, but is not limited to, any type of disk, including a floppy disk, an optical disk, a DVD, a CD-ROM, a micro drive, and a magneto-optical disk, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory device, a magnetic card or an optical card, a nanosystem (including a molecular memory IC), or any type of medium or device suitable for storing instructions and / or data.

[0147] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, by hardware. Based on this understanding, the above technical solution can be essentially or in other words, the part that contributes to the relevant technology can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiment.

[0148] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for identifying urban shrinkage based on change point detection of time series population data, characterized in that: include: S1 obtains the land use spatial raster data of the study area in different periods and the population spatial raster data of the corresponding periods; S2 converts the land use space raster data into the land use space matrix, and converts the population space raster data into the population space matrix, the matrix elements correspond to the cells, and the matrix space corresponds to the cell space; S3 determines the urban land grids based on the land use spatial matrix of each period, extracts the population data of each urban land grid in each period from the population spatial matrix, and constructs a one-dimensional population array of each urban land grid; S4 performs smoothing processing on each one-dimensional population array to obtain smoothed time series data; S5 identifies the mutation nodes of the population data change trend in each one-dimensional population array based on the smoothed time series data, segments each one-dimensional population array based on the mutation nodes of the change trend, and re-determines the start time node and the end time node of the new one-dimensional population array; S6 identifies the population data change trend of the segmented new one-dimensional population array, and determines the location of the urban shrinkage grid and the corresponding start time node and end time node.

2. According to claim 1, a method for identifying urban shrinkage based on change point detection of time series population data is characterized in that: Step S1 includes the following steps: S101 reclassifies the land use spatial raster data of multiple different periods to determine the urban land raster and other land use type raster; S102 reprojects the land use space raster data and the population space raster data of the same period so that the coordinate projections of the land use space raster data and the population space raster data are consistent; S103 resamples the land use spatial raster data of each period to make the grid size of the land use spatial raster data consistent with that of the population spatial raster data.

3. The urban shrinkage identification method based on time series population data change point detection according to claim 2 is characterized in that: Step S2 includes the following steps: S201 determines the land use type corresponding to each grid in the land use spatial grid data, assigns a value to each land use type, and obtains the land use spatial matrix; S202 determines the population data corresponding to each grid in the population space grid data of each period, and obtains the population space matrix corresponding to the population space grid data of each period, wherein each element in each of the population space matrices represents the population data of the corresponding grid in the corresponding period.

4. The urban shrinkage identification method based on time series population data change point detection according to claim 3 is characterized in that: Step S3 includes the following steps: S301 searches for town land grids in the land use space matrix of each period according to the values ​​corresponding to the town land types, and determines the duration of each town land grid; S302 extracts the population data of each town land grid in the corresponding period from the population space matrix corresponding to the continuous period, arranges the population data of each period in chronological order, and constructs a one-dimensional population array of each town land grid.

5. The urban shrinkage identification method based on time series population data change point detection according to claim 1 is characterized in that: Step S4 includes the following steps: Sliding the one-dimensional population array based on a preset sliding window, and generating smoothed time series data based on the mean of population data calculated by sliding the preset sliding window each time; The mean of the population data in the preset sliding window is calculated based on the sliding window smoothing method, using the formula: Wherein, P(t) is the population data after smoothing, g(·) is the population data before smoothing, the size of the preset sliding window is 2n+1, t is the middle point of the preset sliding window, g(t) is the population data corresponding to the middle point, and i is the displacement from the middle point.

6. The urban shrinkage identification method based on time series population data change point detection according to claim 5 is characterized in that: Step S5 includes the following steps: S501 constructs a cost function based on the radial basis function: K(and s,t )=exp(-σ||y s -and t || 2 ); Among them, K(y s,t ) is the similarity between the data at time s and the data at time t, and its value range is [0,1], ||y s -y t || represents the Euclidean distance between two data points, σ ​​represents the bandwidth, which controls the decay rate of the similarity between two data points as the distance increases; c(y s,t ) is the cost function of the data at time s and the data at time t, b and a are parameters; S502 constructs an objective function according to the cost function, and the optimization goal is to minimize the objective function, including: F(t)=Z(s)+c(y s,t )+β; Where F(t) is the objective function when data point t is the split point, Z(s) is the objective function value when feasible solution s is the split point, and when the feasible solution set is empty, Z(s) is 0; β is the penalty term, which is used to control the number of identified change points; S503 updates the feasible solution set in each cycle of solving the objective function, adds the current time t to the feasible solution set, and filters out the time s in the feasible solution set that does not meet the following conditions: Z(s)+c(y s,t )≤Z(t); S504 determines the mutation node of the population data change trend according to the solution result of the objective function, segments the time series data at each position, segments each one-dimensional population array based on the mutation node of the change trend, and re-determines the start time node and end time node of the new one-dimensional population array.

7. The urban shrinkage identification method based on time series population data change point detection according to claim 6 is characterized in that: Step S6 includes the following steps: Based on the new one-dimensional population array after the above segmentation and the start time node and end time node corresponding to the new one-dimensional population array, regression fitting is performed, and the formula is applied: P t =kt′+b; The urban shrinkage raster is determined based on the following formula: If the output type of f(k,ΔP) is contraction, the corresponding grid is determined to be an urban contraction grid, and the rest are non-contracting grids, and the urban contraction grid is output; Among them, t′ is the time, P t is the population data at time t, k is the slope of the fitting line, b is the intercept, and the trend of population data change in the corresponding grid is judged based on the slope k of the fitting line: when k>0, the population in the grid is on an upward trend, when k<0, the population in the grid is on a downward trend, and ΔP is the rate of change of population data.

8. A device for identifying urban shrinkage based on change point detection of time series population data, characterized in that: include: The data acquisition module is used to obtain the land use spatial raster data of the study area at different periods and the population spatial raster data of the corresponding periods; The data processing module is used to convert the land use space grid data into the land use space matrix, and convert the population space grid data into the population space matrix, where the matrix elements correspond to the cells and the matrix space corresponds to the cell space; The data processing module is also used to determine the town land grids based on the land use space matrix of each period, extract the population data of each town land grid in each period from the population space matrix, and construct a one-dimensional population array of each town land grid; The data processing module is also used to perform smoothing processing on each one-dimensional population array to obtain smoothed time series data; The calculation module is used to identify the change trend mutation nodes of the population data in each one-dimensional population array based on the smoothed time series data, segment each one-dimensional population array based on the change trend mutation nodes, and re-determine the start time node and end time node of the new one-dimensional population array; The calculation module is also used to identify the population data change trend of the new one-dimensional population array after segmentation, and determine the position of the urban shrinkage grid and the corresponding start time node and end time node.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images