Historical population space reconstruction method and system based on altitude driving, terminal and storage medium

Abstract

The invention discloses a historical population space reconstruction method and system based on altitude driving, a terminal and a storage medium, and the method comprises the steps: obtaining population statistical data in a historical period, and carrying out the preprocessing, and obtaining target historical population data; acquiring historical spatial data, performing matching processing on the historical spatial data and the target historical population data to obtain a three-dimensional structure database, and performing mask extraction and resampling processing to obtain target altitude data; constructing a population and altitude quantitative mapping relation model according to the target historical population data and the target altitude data; and according to the population and altitude quantitative mapping relation model, performing preliminary space distribution processing and loop iterative calculation on the target historical population data to obtain a historical population space reconstruction result. According to the method, the population distribution is quantitatively associated with the altitude, so that the population number in the historical period can be subjected to refined space distribution in the geographic unit, and the accuracy of generation of the historical population space reconstruction result is effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a historical population spatial reconstruction method and system based on altitude driving, a terminal and a computer readable storage medium. BACKGROUND

[0002] The historical population spatial distribution pattern is a key indicator for quantifying the intensity of human activity on the environment, and is also an important scientific basis for revealing the evolution mechanism of human-earth relationship. Historical population spatial pattern reconstruction is a process of integrating historical population statistical data, geographical zoning information and environmental factors and other multi-source data, and restoring macro population data to high-resolution grid using spatialization model. The core is to realize the quantitative expression of population spatial distribution in a long time sequence, so as to accurately reveal the time evolution law of population in spatial position and density.

[0003] However, in the prior art, historical population statistical data generally have time breakpoints (i.e. years or time nodes with clear population statistical records in historical documents, which have the characteristics of discontinuity and no periodicity) and data missing phenomenon, and the geographical zoning changes frequently in different dynasties, thereby causing difficulty in historical population spatial reconstruction in a long time sequence, and the accuracy of the reconstruction result is not high.

[0004] Therefore, the prior art still needs to be improved and developed. SUMMARY

[0005] The main purpose of the present application is to provide a historical population spatial reconstruction method and system based on altitude driving, a terminal and a computer readable storage medium, which aims to solve the problem that historical population statistical data in the prior art generally have time breakpoints and data missing phenomenon, and the geographical zoning changes frequently in different dynasties, thereby causing difficulty in historical population spatial reconstruction in a long time sequence, and the accuracy of the reconstruction result is not high.

[0006] To achieve the above-mentioned purpose, the present application provides a historical population spatial reconstruction method based on altitude driving, which comprises the following steps: Obtain historical population statistical data, and pre-process the historical population statistical data to obtain target historical population data; Obtain historical spatial data, match the historical spatial data with the target historical population data, obtain a three-dimensional structure database, and perform mask extraction and resampling processing on the three-dimensional structure database to obtain target altitude data; Construct a population-altitude quantitative mapping relationship model according to the target historical population data and the target altitude data; According to the population and elevation quantitative mapping relationship model, the target historical population data is subjected to preliminary spatial allocation processing and cyclic iteration calculation, so as to obtain a historical population spatial reconstruction result.

[0007] Optionally, the elevation-driven historical population spatial reconstruction method, wherein the preprocessing comprises structured processing, data cleaning processing and time series interpolation processing. The historical period population statistical data is obtained and preprocessed, so as to obtain target historical population data, specifically comprising: The historical period population statistical data in a preset time period is obtained, and the structured processing is performed on the historical period population statistical data by using an OCR character recognition method, so as to obtain a population data recognition result. The data cleaning processing is performed on the population data recognition result, so as to obtain a target population data recognition result, wherein the data cleaning processing comprises format unification processing, field name standardization processing, missing value marking processing and abnormal value elimination processing. A hierarchical interpolation model is constructed, and the time series interpolation processing is performed on the target population data recognition result according to the hierarchical interpolation model, so as to obtain target historical population data.

[0008] Optionally, the elevation-driven historical population spatial reconstruction method, wherein the structured processing is performed on the historical period population statistical data by using an OCR character recognition method, so as to obtain a population data recognition result, specifically comprising: The historical period population statistical data is subjected to rendering processing, so as to obtain a high-resolution image. The high-resolution image is subjected to grayscale processing, so as to obtain a grayscale image. The grayscale image is subjected to full-text image recognition processing, so as to obtain a population data recognition result.

[0009] Optionally, the elevation-driven historical population spatial reconstruction method, wherein the historical spatial data is obtained, the historical spatial data is matched with the target historical population data, so as to obtain a three-dimensional structure database, and the three-dimensional structure database is subjected to mask extraction and resampling processing, so as to obtain target elevation data, specifically comprising: The historical spatial data is obtained and subjected to vectorization processing, so as to obtain vector polygon data. A preset reference boundary is determined, and the vector polygon data is subjected to optimization adjustment according to the preset reference boundary, so as to obtain a target vector graph, wherein the optimization adjustment comprises translation adjustment, rotation adjustment and scale adjustment. The historical spatial data is matched with the target vector graph, so as to obtain matching data. The area weighting method or the population proportion method is used to perform population data allocation on the target historical population data according to the matching data, so as to obtain a three-dimensional structure database; A space mask is obtained by performing mask extraction on the three-dimensional structure database. A digital elevation model is constructed, and elevation information is obtained by performing elevation information extraction on the three-dimensional structure database according to the space mask and the digital elevation model. The data resampling processing is performed on the elevation information by using a bilinear interpolation resampling method, so as to obtain target elevation data.

[0010] Optionally, the historical population space reconstruction method based on the elevation driving, wherein the expression of the population and elevation quantitative mapping relationship model is: ; wherein, and is a grid space index, is a population number in the grid, is an elevation in the grid, and are period-specific parameters.

[0011] Optionally, the historical population space reconstruction method based on the elevation driving, wherein the preliminary space allocation processing and the loop iteration calculation are performed on the target historical population data according to the population and elevation quantitative mapping relationship model, so as to obtain the historical population space reconstruction result, and specifically include: The initial population distribution result is obtained by performing the normalization processing and the space allocation processing on the target historical population data according to the population and elevation quantitative mapping relationship model. The double constraint mechanism is determined, and the loop iteration calculation is performed on the initial population distribution result according to the double constraint mechanism, so as to obtain the historical population space reconstruction result.

[0012] Optionally, the historical population space reconstruction method based on the elevation driving, wherein the double constraint mechanism is determined, and the loop iteration calculation is performed on the initial population distribution result according to the double constraint mechanism, so as to obtain the historical population space reconstruction result, and specifically include: The allocation result after each iteration is obtained by performing the loop iteration calculation on the initial population distribution result, wherein the allocation result includes a total quantity error and a density constraint error. The double constraint mechanism is determined, wherein the double constraint mechanism includes a space total quantity constraint and a population density extreme constraint. The distribution result is compared with the double constraint mechanism, if the total amount error is less than the space total amount constraint and the density constraint error is less than the population density extreme value constraint in continuous iteration times, iteration is stopped, and a historical population space reconstruction result is obtained.

[0013] In addition, to achieve the above object, the application further provides a historical population space reconstruction system based on altitude driving, wherein the historical population space reconstruction system based on altitude driving comprises: A historical population data extraction module is configured to acquire historical period population statistical data, and pre-process the historical period population statistical data to obtain target historical population data. An altitude data extraction module is configured to acquire historical space data, match the historical space data with the target historical population data to obtain a three-dimensional structure database, and perform mask extraction and resampling processing on the three-dimensional structure database to obtain target altitude data. A mapping relationship model construction module is configured to construct a population and altitude quantitative mapping relationship model according to the target historical population data and the target altitude data. A space reconstruction result output module is configured to perform preliminary spatial distribution processing and cyclic iteration calculation on the target historical population data according to the population and altitude quantitative mapping relationship model to obtain a historical population space reconstruction result.

[0014] In the application, historical period population statistical data is acquired, and the historical period population statistical data is pre-processed to obtain target historical population data; historical space data is acquired, the historical space data is matched with the target historical population data to obtain a three-dimensional structure database, and mask extraction and resampling processing are performed on the three-dimensional structure database to obtain target altitude data; a population and altitude quantitative mapping relationship model is constructed according to the target historical population data and the target altitude data; and preliminary spatial distribution processing and cyclic iteration calculation are performed on the target historical population data according to the population and altitude quantitative mapping relationship model to obtain a historical population space reconstruction result. By quantitatively correlating population distribution and altitude, a population and altitude quantitative mapping relationship model is constructed, and preliminary spatial distribution processing and cyclic iteration calculation are performed on historical population data according to the population and altitude quantitative mapping relationship model, so that the population quantity in a historical period can be finely spatially distributed in a geographical unit, and the accuracy of generating a historical population space reconstruction result is effectively improved. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 is a flowchart of a preferred embodiment of the historical population space reconstruction method based on altitude driving of the application; Figure 2is a schematic diagram of the overall structure implementation process of a preferred embodiment of the altitude-driven historical population spatial reconstruction method of the present application; Figure 3 is a structure diagram of a preferred embodiment of the altitude-driven historical population spatial reconstruction system of the present application; Figure 4 is a structure diagram of a preferred embodiment of the terminal of the present application. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical scheme and advantages of the present application more clear and explicit, the present application will be further described in detail below with reference to the drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0017] The historical population spatial distribution pattern is a key indicator for quantifying the intensity of human activity on the environment, and is also an important scientific basis for revealing the evolution mechanism of human-earth relationship. Historical population spatial pattern reconstruction is a process of integrating historical population statistical data, geographical division information and environmental factors and other multi-source data, and using spatialization model to restore macro population data to high-resolution grid. The core is to realize the quantitative expression of population spatial distribution in a long time sequence, so as to accurately reveal the time evolution law of population in spatial position and density. The current historical population spatial reconstruction method has the following challenges: 1. Data spatio-temporal discontinuity problem: historical population statistical data generally have time breakpoints and missing phenomena, and the geographical division changes frequently, making it difficult to finely reconstruct in a long time sequence; 2. Insufficient portability of modern methods: the influence factors commonly used in modern population gridding modeling are usually not available in historical period, which makes the historical portability of the reconstruction method controversial.

[0018] To solve the above problems, the present application proposes a historical population spatial reconstruction method based on altitude driving, which takes the relationship between modern population and altitude as the modeling basis, combines historical geographical division data and time series population statistical data, and performs gridding distribution under the dual constraint mechanism, thereby realizing high-precision historical population spatial reconstruction across the millennium scale.

[0019] The altitude-driven historical population spatial reconstruction method according to the preferred embodiment of the present application, as shown in Figure 1 and Figure 2 The altitude-driven historical population spatial reconstruction method comprises the following steps: Step S10, obtaining historical population statistical data, and preprocessing the historical population statistical data to obtain target historical population data. The preprocessing includes structured processing, data cleaning processing and time series interpolation processing.

[0020] The historical period population statistics in the present application are derived from the Population History of a Certain Country, which are obtained by separately extracting and collating the relevant contents (including tables and text parts) and their page numbers in the Population History of a Certain Country involving population statistics, and after subsequent pretreatment, can be convenient for subsequent OCR (Optical Character Recognition, character recognition technology) recognition and structured extraction.

[0021] Specifically, historical period population statistics in a preset time period are obtained, and the historical period population statistics are rendered to obtain a high-resolution image; the high-resolution image is subjected to grayscale processing to obtain a grayscale image; the grayscale image is subjected to full-text image recognition processing to obtain a population data recognition result.

[0022] The acquisition process of the historical period population statistics is as follows: the population statistics of each historical period in the Population History of a Certain Country during 1000-2000 (i.e. in the preset time period in the present application) are collected, and are uniformly collated according to the corresponding AD year, so that different time systems correspond to AD years, forming a population sequence framework with time comparability.

[0023] Further, the OCR text recognition technology is used to batch analyze the collated historical literature PDF files such as the Population History of a Certain Country, automatically detect and extract the table regions. In view of the common problems such as broken lines, merged cells and character recognition errors in historical tables, processing algorithms including character correction, column alignment and separation of numbers and text are designed to realize high-precision structured extraction of historical population data.

[0024] As shown in Figure 2 The specific process of the OCR structured processing method in the present application is as follows: 1. The historical PDF literature page is rendered, converted into a high-resolution image and pretreated (i.e. grayscale processing); 2. The OCR engine is used to perform full-text recognition on the grayscale image to obtain a population data recognition result. In order to solve the common errors and format disorder problems in the recognition result, the present application designs a multi-stage post-processing and repair process, including: word break repair based on regular expressions, targeted data cleaning to merge and split numbers and separate text numbers, and through analysis of the text structure and key anchor point extraction of the core three-line table data area, effectively excluding the interference of non-data content such as headers and footers. Finally, the processed data is converted into a structured electronic table, realizing efficient and accurate conversion from unstructured literature images to regular data.

[0025] It can be understood that the present application first utilizes the PyMuPDF library (a Python library specially used for PDF document processing) to render the specified pages of the PDF literature into high-resolution images, and uses OpenCV (an open-source computer vision library) for grayscale and other preprocessing to optimize image quality. Subsequently, the Tesseract OCR engine (an open-source optical character recognition engine) is called, combined with a Chinese language model and a page segmentation mode (such as PSM 6), to perform full-text recognition on the grayscale image.

[0026] To solve the common recognition errors and format confusion problems in the OCR process, a multi-stage post-processing and repair process is designed: 1. Intelligent repair of word breaks caused by OCR misrecognition through regular expressions; 2. Development of a targeted data cleaning module that can automatically merge small numbers that have been incorrectly split and separate closely connected text and numbers within the same cell; 3. By analyzing the number of columns in the text line and using the "total" line as a key anchor point, the core three-line table data area is effectively identified and extracted, while excluding the interference of headers, footers, and annotation text. Finally, the cleaned and structured data is converted to Pandas DataFrame format (DataFrame is a core data structure in the Pandas library, which is a two-dimensional, tabular data structure that can be used to store and process various types of data), and output as an Excel file, thereby achieving efficient and accurate conversion from unstructured historical literature images to structured spreadsheets.

[0027] The data cleaning process is performed on the population data recognition result to obtain a target population data recognition result, wherein the data cleaning process includes format unification processing, field name standardization processing, missing value marking processing, and outlier removal processing; a hierarchical interpolation model is constructed, and the target population data recognition result is subjected to the time series interpolation processing according to the hierarchical interpolation model to obtain a target historical population data.

[0028] As shown in Figure 2 , the population data recognition result is subjected to data cleaning, and the process is as follows: the population data recognition result is subjected to format unification, field name standardization, missing value marking, and outlier removal.

[0029] As shown in Figure 2 , after data cleaning, the target population data recognition result obtained after cleaning is subjected to time series interpolation, wherein the process of time series interpolation is as follows: a hierarchical interpolation model is designed, combined with the annual growth rate of adjacent time points (here, "adjacent time points" refer to two adjacent time points in the historical period for which population statistics data exists) and the spatial estimation method of adjacent geographical areas, to complete the population data in the discontinuous period and reconstruct the time series sequence of the total population in the century interval.

[0030] The process of complementing the population data in the missing period is as follows: obtaining the annual population growth rate between two adjacent time points (population known), taking it as the population growth rate between the base year (population known) and the target year (population unknown), combining the time interval between the base year and the target year, and calculating the population of the target year.

[0031] The linear extrapolation formula based on the annual growth rate is: ; Wherein, The population of the target year, The population of the base year, The annual growth rate based on literature research or back calculation, The base year.

[0032] It can be understood that, for the problem of space-time discontinuity of historical population data, the application constructs a hierarchical interpolation model combining time sequence and space to complement the population data in the missing period, and reconstructs the population total sequence in the interval of 100 years, and the specific process is as follows: 1. Time series data interpolation: if there are ≥2 reliable time point data (years) in the same historical period and no big data change, the linear extrapolation based on the annual growth rate is carried out to obtain the adjacent 100-year population data, wherein the formula of linear extrapolation is: .

[0033] 2. Spatial data interpolation: for the case that individual geographical unit historical population statistical data is missing, according to the spatial adjacency and social economic-geographical environment similarity characteristics, the spatial interpolation method is adopted, taking the population growth rate of the adjacent administrative region meeting the condition as the proxy variable, combining the reliable population base of the target unit at the latest time point, to estimate the data of the missing year.

[0034] Step S20, obtaining historical spatial data, matching the historical spatial data with the target historical population data, obtaining a three-dimensional structure database, and performing mask extraction and resampling processing on the three-dimensional structure database to obtain target elevation data.

[0035] The historical spatial data in the application is obtained from the PDF document of "a country historical atlas", and the historical spatial data is matched with the historical population data, so as to obtain the elevation data corresponding to the historical population data.

[0036] Specifically, historical spatial data is acquired, and vectorization is performed on the historical spatial data to obtain vector polygon data; a preset reference boundary is determined, and the vector polygon data is adjusted and optimized according to the preset reference boundary to obtain a target vector graph, wherein the adjustment and optimization include translation adjustment, rotation adjustment, and scale adjustment.

[0037] The processing procedure of the spatial data is as follows: relying on a PDF document of a historical atlas of a certain country, a boundary line is drawn along a hierarchical geographical boundary explicitly marked on a map by means of artificial vectorization on a GIS platform to generate vector polygon data. In order to ensure the spatial comparability of geographical divisions in different historical periods, a "reference period" boundary layer is established with a preset geographical division as a reference, and the establishment procedure of the "reference period" boundary layer is as follows: taking the preset geographical division as a space-time reference, a polygon is generated by artificial vectorization operation on an ArcGIS platform (ArcGIS is a comprehensive geographic space platform) based on corresponding boundary information in the historical atlas of the certain country, the polygon is generated by drawing and closing a polygon along the boundaries of multiple-level geographical units to form the "reference period" boundary layer. The boundary layer serves as a spatial reference for subsequent registration and alignment of geographical boundaries in different periods.

[0038] Further, the spatial registration and overlay analysis functions of ArcGIS are used to realize registration and correction of geographical boundaries in different periods with the main body consistency and spatial continuity taken into account, and the specific steps are as follows: 1. The vectorized geographical boundary shp files (shp file is short for ESRIShapefile, which is a kind of vector graphics format capable of saving the position of geometric graphics and related attributes) in other historical periods are imported according to the same coordinate system; 2. Control points of the reference layer and the target period boundary are set to perform translation, rotation, and scale adjustment to make the main boundary line correspond to the reference boundary.

[0039] The registration and correction procedure is based on the spatial analysis function of ArcGIS, and the "reference period" boundary layer is used as a reference to realize the unification of geographical boundaries in different periods by means of spatial registration (Spatial Adjustment) and overlay analysis (Overlay Analysis).

[0040] The historical spatial data and the target vector graph are matched to obtain matching data; an area weighting method or a population proportion method is used to distribute the target historical population data according to the matching data to obtain a three-dimensional structure database.

[0041] As Figure 2As shown, the process of matching the historical spatial data with the target vector map is as follows: the cleaned and interpolated historical population table (i.e. target historical population data, containing the geographical area name and the corresponding year population) is associated with the digitized geographical area vector map (i.e. historical spatial data) through attribute table association, wherein the matching rules include: 1. matching through geographical area name (considering homophonic and historical evolution alias table); 2. for boundary adjustment or combination, the population data is distributed to the corresponding new boundary unit by using area weighting or population proportion method, and finally a three-dimensional structure database covering "population statistics-administrative division-geographical space" is formed.

[0042] The three-dimensional structure database is subjected to mask extraction to obtain spatial mask; a digital elevation model is constructed, and elevation information is extracted from the three-dimensional structure database according to the spatial mask through the digital elevation model to obtain altitude information.

[0043] As shown in Figure 2 As shown, the process of extracting the altitude information is as follows: based on the SRTM global digital elevation model (DEM), the vector boundary (shp file) of the preset watershed range in the historical period is used as a spatial mask to extract the elevation information in the study area range by using the mask extraction (Extract by Mask) method in the ArcGIS platform, thereby providing basic data for subsequent altitude-driven modeling.

[0044] The altitude information is subjected to data resampling processing by using a bilinear interpolation resampling method to obtain target altitude data.

[0045] As shown in Figure 2 As shown, the process of data resampling is as follows: the extracted DEM altitude information is uniformly resampled to a grid resolution of 10 km x 10 km by using the bilinear interpolation resampling method, thereby providing a unified spatial data basis for subsequent nonlinear relationship modeling and spatial distribution.

[0046] In step S30, a population and altitude quantitative mapping relationship model is constructed according to the target historical population data and the target altitude data.

[0047] As shown in Figure 2 As shown, the process of modeling the modern population and altitude relationship is as follows: based on the multi-period modern population data and altitude data, an exponential decay type nonlinear fitting method is used to construct a population and altitude quantitative mapping relationship model, so as to verify the dynamic and regular changes of the exponential decay model parameters with time, and to provide a reference for adaptive adjustment of parameters in historical population reconstruction.

[0048] Specifically, the expression of the population and altitude quantitative mapping relationship model is as follows: ; in, and For raster space indexing, For grid (i.e., by) and Population count in a grid (composed of cells) For grid (i.e., by) and Elevation (m) in the grid (composed of the grid) and As a time-specific parameter, this sequence pair exhibits an increasing / decreasing pattern over time.

[0049] The quantitative mapping model between population and altitude set in this invention demonstrates a negative exponential relationship between population and altitude, that is, there is a quantitative mapping relationship between population and altitude in a statistical sense.

[0050] Step S40: Perform preliminary spatial allocation processing and iterative calculation on the target historical population data according to the quantitative mapping relationship model between population and altitude to obtain the spatial reconstruction result of the historical population.

[0051] After constructing a quantitative mapping model between population and altitude, the historical population can be iteratively spatially allocated using this model to obtain the spatial reconstruction results of the historical population.

[0052] Specifically, the initial population distribution results are iteratively calculated to obtain the allocation results after each iteration, wherein the allocation results include total error and density constraint error.

[0053] The specific process of iterative spatial allocation of historical population is as follows: First, initial allocation is performed: using the fitted nonlinear relationship between population and altitude (i.e., the quantitative mapping relationship model between population and altitude), the total population data after interpolation is initially spatially allocated to generate an initial gridded population distribution.

[0054] Understandably, once the quantitative mapping model between population and altitude is established, the spatial distribution probability of the population can be estimated using the altitude values ​​of each grid cell. The specific process is as follows: 1. Input the DEM data into the model and determine the altitude value of each grid cell. 2. Calculate the theoretical population weight value for each grid cell based on the negative exponential model. 3. Re-normalize the weights of all grid cells within each cell, so that... 4. Allocate the total population of each unit to each grid according to the proportions of different historical periods: 5. Output the initial population distribution results to form a gridded historical population initial value layer based on altitude. Through the above steps, the historical population distribution is spatially matched with the terrain features, reflecting the regularity of population decrease with altitude.

[0055] A dual constraint mechanism is determined, which includes a spatial total constraint and a population density extreme value constraint. The allocation result is compared with the dual constraint mechanism. If, in a series of iterations, the total error is less than the spatial total constraint and the density constraint error is less than the population density extreme value constraint, the iteration is stopped, and the historical population spatial reconstruction result is obtained.

[0056] Furthermore, such as Figure 2 As shown, this invention employs dual constraints for optimization, wherein a dual constraint mechanism is established, specifically including: 1. Set a spatial total constraint, using no more than ±1% of the total population count for each period as the total raster population after spatial allocation (i.e., SUM( The constraints of the ()) condition ensure that the total population of the raster within the cell is consistent with the total population count. 2. Set extreme population density constraints to limit the value of a single raster cell from exceeding the highest theoretical density of its constituent units. This is determined by the preset highest population density for each period (total population within the geographic unit / total area of ​​the unit) and the raster area (100km²). 2 The product of ) is used as the population of a single raster during the same period. The upper limit of ).

[0057] Furthermore, such as Figure 2 As shown, this invention incorporates parameter iteration and adaptive step size adjustment. The specific process includes: continuously optimizing the allocation result in a loop through parameter iteration and adaptive step size adjustment until the parameters are optimized. The iteration process satisfies two constraints: ; ; in, This represents the total population for the corresponding period. Population of an administrative unit This refers to the area of ​​the corresponding administrative unit.

[0058] It is understandable that after obtaining the initial allocation results, the parameters are adjusted through iterative loops. Adaptive adjustments are made to simultaneously satisfy both "total population constraints" and "population density constraints." The specific process is as follows: 1. Initialization parameters: The parameters of the modern population and altitude model are used as the initial values. ; 2. Error calculation: In each iteration, the deviation of the allocation result from the target constraint is calculated, including: Total quantity error: wherein, is the total population of the corresponding period; Density constraint error: wherein, is the maximum population density of the period.

[0059] 3. Parameter update: If or , adjust the parameter to obtain: ; ; wherein, and are adaptive step size coefficients that automatically scale according to the error size (the larger the error, the smaller the step size to prevent over-adjustment).

[0060] 4. Convergence determination: When and for three consecutive iterations, the model is considered to have reached a stable convergence state; wherein the model convergence condition is: when the total quantity error is less than 1% and all grid population values do not exceed the theoretical maximum density for a certain number of iterations, it is considered that the parameter combination has reached stability, and the parameters at this time are defined as the "optimal combination parameters" for the period, and the corresponding population spatial distribution result is the final output.

[0061] 5. Output result: Save the optimal parameter combination and the corresponding gridded population result as the final reconstruction data for the period.

[0062] The above process makes the model adaptively approach the reasonable structure of the population spatial distribution of the historical period through iterative optimization, and simultaneously satisfies the total quantity consistency and density constraint requirements in the spatial statistical sense.

[0063] Further, a loop and convergence process is also set, the specific process is: repeat the above steps until the model solves the optimal combination parameters for different historical time points. When the total quantity error is less than 1% and all grid population values do not exceed the theoretical maximum density for a certain number of iterations, it is considered that the parameter combination has reached stability. The parameters at this time are defined as the "optimal combination parameters" for the period, and the corresponding population spatial distribution result is the final output.

[0064] Result data output: Based on the above results, generate and output historical population distribution data sets from 1000-2000 AD, with a time resolution of every hundred years and a spatial resolution of 10 km.

[0065] The beneficial effects of the present application are: The present application provides a historical population spatial reconstruction method based on altitude driving. Through multi-source data integration and hierarchical interpolation method, the time breakpoints and missing data of statistical data are effectively completed, and the spatial correlation between population statistics and administrative division is realized, overcoming the long-time sequence fine reconstruction problem caused by the time and space discontinuity of historical data and the frequent change of administrative division. The present application uses a modern population-altitude nonlinear relationship model and introduces a double constraint optimization mechanism, overcoming the challenges of difficult access to historical period impact factors and insufficient portability of modern methods. The present application can stably obtain high-resolution population spatial distribution data on a long-time sequence and a 10km grid scale, significantly improving the accuracy and reliability of the reconstruction results, and providing reliable data support for ecological protection and human-earth relationship evolution research.

[0066] Further, as shown in Figure 3 Based on the historical population spatial reconstruction method based on altitude driving, the present application also correspondingly provides a historical population spatial reconstruction system based on altitude driving, wherein the historical population spatial reconstruction system based on altitude driving comprises: A historical population data extraction module 51 is configured to obtain population statistics in a historical period, pre-process the population statistics in the historical period, and obtain target historical population data. An altitude data extraction module 52 is configured to obtain historical spatial data, match the historical spatial data with the target historical population data, obtain a three-dimensional structure database, and perform mask extraction and resampling processing on the three-dimensional structure database to obtain target altitude data. A mapping relationship model construction module 53 is configured to construct a population-altitude quantitative mapping relationship model according to the target historical population data and the target altitude data. A spatial reconstruction result output module 54 is configured to perform preliminary spatial allocation processing and cyclic iteration calculation on the target historical population data according to the population-altitude quantitative mapping relationship model to obtain a historical population spatial reconstruction result.

[0067] Further, as shown in Figure 4 Based on the historical population spatial reconstruction method and system based on altitude driving, the present application also correspondingly provides a terminal, which comprises a processor 10, a memory 20 and a display 30. Figure 4 Only part of the components of the terminal are shown, but it should be understood that all the shown components are not required to be implemented, and more or fewer components can be alternatively implemented.

[0068] The memory 20 can be an internal storage unit of the terminal in some embodiments, such as a hard disk or a memory of the terminal. The memory 20 can also be an external storage device of the terminal in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. Further, the memory 20 can include both an internal storage unit and an external storage device of the terminal. The memory 20 is used to store application software and various data installed on the terminal, such as program codes of the terminal, etc. The memory 20 can also be used to temporarily store data that has been output or will be output. In an embodiment, the memory 20 stores a history population spatial reconstruction program based on altitude driving 40, which can be executed by the processor 10 to implement the history population spatial reconstruction method based on altitude driving in the present application.

[0069] The processor 10 can be a Central Processing Unit (CPU), a microprocessor or other data processing chip in some embodiments, which is used to run program codes or process data stored in the memory 20, such as to execute the history population spatial reconstruction method based on altitude driving, etc.

[0070] The display 30 can be an LED display, a liquid crystal display, a touch liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, etc. in some embodiments. The display 30 is used to display information of the terminal and to display a visualized user interface.

[0071] In an embodiment, the processor 10 implements the steps of the history population spatial reconstruction method based on altitude driving when executing the history population spatial reconstruction program based on altitude driving 40 in the memory 20.

[0072] In summary, the application provides a kind of elevation-driven historical population space reconstruction method, system and terminal based on it, the method comprises: obtaining historical population data, and the historical population data is preprocessed, and target historical population data is obtained;Historical spatial data is obtained, and the historical spatial data is matched with the target historical population data, and three-dimensional structure database is obtained, and the three-dimensional structure database is carried out mask extraction and resampling processing, and target elevation data is obtained;According to the target historical population data and the target elevation data, a population and elevation quantitative mapping relationship model is constructed;According to the population and elevation quantitative mapping relationship model, the target historical population data is preliminarily spatially distributed and iteratively calculated, and historical population space reconstruction result is obtained.The application quantitatively correlates population distribution and altitude, constructs a population and elevation quantitative mapping relationship model, and preliminarily spatially distributes and iteratively calculates historical population data through the population and elevation quantitative mapping relationship model, which can realize fine spatial distribution of population quantity in geographical unit in historical period, and effectively improves the accuracy of historical population space reconstruction result generation.

[0073] It should be noted that in this document, the terms "comprising", "including", or any other variant thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal including a list of elements does not only include those elements, but also other elements not expressly listed or inherent to such process, method, article, or terminal. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article, or terminal including the element.

[0074] Of course, those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by a computer program instructing relevant hardware (such as a processor, a controller, etc.) to complete, and the program can be stored in a computer-readable computer-readable storage medium, and the program can include the processes of the above-mentioned method embodiments when executed. The computer-readable storage medium can be a memory, a disk, an optical disk, etc.

[0075] It should be understood that the application is not limited to the above examples, and those skilled in the art can improve or modify it according to the above description, and all these improvements and modifications shall fall within the scope of the appended claims of the present application.

Claims

1. A method for spatial reconstruction of historical population based on altitude, characterized in that, The elevation-driven spatial reconstruction method for historical population includes: Obtain historical population statistics data and preprocess the historical population statistics data to obtain target historical population data; Historical spatial data is acquired, and the historical spatial data is matched with the target historical population data to obtain a three-dimensional structure database. The three-dimensional structure database is then subjected to mask extraction and resampling to obtain the target elevation data. A quantitative mapping model between population and altitude is constructed based on the target's historical population data and target altitude data. Based on the quantitative mapping relationship model between population and altitude, the target historical population data is subjected to preliminary spatial allocation processing and iterative calculation to obtain the spatial reconstruction results of the historical population.

2. The method for spatial reconstruction of historical population based on altitude-driven approach according to claim 1, characterized in that, The preprocessing includes structure processing, data cleaning processing, and time-series interpolation processing; The process of acquiring historical population statistics and preprocessing them to obtain target historical population data specifically includes: Obtain historical population statistics within a preset time period, and use OCR text recognition method to perform the structured processing on the historical population statistics to obtain population data recognition results; The population data identification results are subjected to the data cleaning process to obtain the target population data identification results. The data cleaning process includes format unification, field name standardization, missing value marking, and outlier removal. A hierarchical interpolation model is constructed, and the target population data identification results are subjected to the time-series interpolation processing based on the hierarchical interpolation model to obtain the target historical population data.

3. The method for spatial reconstruction of historical population based on altitude-driven methods according to claim 2, characterized in that, The method of using OCR text recognition to perform structured processing on the historical population statistics data to obtain population data recognition results specifically includes: The historical demographic data is rendered to obtain a high-resolution image. The high-resolution image is converted to grayscale to obtain a grayscale image; The grayscale image is subjected to full-text image recognition processing to obtain population data recognition results.

4. The method for spatial reconstruction of historical population based on altitude-driven approach according to claim 1, characterized in that, The process of acquiring historical spatial data, matching the historical spatial data with the target historical population data to obtain a three-dimensional structure database, and performing mask extraction and resampling on the three-dimensional structure database to obtain target elevation data specifically includes: Historical spatial data is acquired and vectorized to obtain vector polygon data; A preset baseline boundary is determined, and the vector polygon data is optimized and adjusted according to the preset baseline boundary to obtain the target vector map. The optimization and adjustment include translation adjustment, rotation adjustment and scale adjustment. The historical spatial data is matched with the target vector map to obtain matching data; The target historical population data is allocated according to the matching data using either the area-weighted method or the population ratio method to obtain a three-dimensional structural database. The spatial mask is obtained by performing mask extraction on the three-dimensional structure database; A digital elevation model is constructed, and elevation information is extracted from the three-dimensional structure database using the digital elevation model and the spatial mask to obtain altitude information. The altitude information is resampled using a bilinear interpolation resampling method to obtain the target altitude data.

5. The method for spatial reconstruction of historical population based on altitude-driven methods according to claim 1, characterized in that, The expression for the quantitative mapping relationship model between population and altitude is: ； in, and For raster space indexing, The population count in the grid. The elevation in the grid. and These are time-specific parameters.

6. The method for spatial reconstruction of historical population based on altitude-driven approach according to claim 1, characterized in that, The process of performing preliminary spatial allocation processing and iterative calculations on the target historical population data based on the quantitative mapping relationship model between population and altitude to obtain the spatial reconstruction result of the historical population specifically includes: The target historical population data is normalized and spatially allocated based on the quantitative mapping relationship model between population and altitude to obtain the initial population distribution results. A dual constraint mechanism is determined, and the initial population distribution results are iteratively calculated based on the dual constraint mechanism to obtain the historical population spatial reconstruction results.

7. The method for spatial reconstruction of historical population based on altitude-driven methods according to claim 6, characterized in that, The determination of the dual constraint mechanism, and the iterative calculation of the initial population distribution results based on the dual constraint mechanism to obtain the historical population spatial reconstruction results, specifically includes: The initial population distribution results are iteratively calculated to obtain the allocation results after each iteration, wherein the allocation results include total error and density constraint error; A dual constraint mechanism is established, wherein the dual constraint mechanism includes a total spatial constraint and a population density extreme value constraint; The allocation result is compared with the dual constraint mechanism. If, in consecutive iterations, the total error is less than the total spatial constraint and the density constraint error is less than the extreme population density constraint, the iteration is stopped, and the historical population spatial reconstruction result is obtained.

8. An altitude-driven spatial reconstruction system for historical population, characterized in that, The altitude-driven historical population spatial reconstruction system includes: The historical population data extraction module is used to obtain population statistics data for historical periods and to preprocess the population statistics data for historical periods to obtain target historical population data. The elevation data extraction module is used to acquire historical spatial data, match the historical spatial data with the target historical population data to obtain a three-dimensional structure database, and perform mask extraction and resampling processing on the three-dimensional structure database to obtain the target elevation data. The mapping relationship model construction module is used to construct a quantitative mapping relationship model between population and altitude based on the target's historical population data and the target's altitude data. The spatial reconstruction result output module is used to perform preliminary spatial allocation processing and iterative calculation on the target historical population data according to the quantitative mapping relationship model between population and altitude, so as to obtain the spatial reconstruction result of the historical population.

9. A terminal, characterized in that, The terminal includes: a memory, a processor, and an altitude-driven historical population spatial reconstruction program stored in the memory and executable on the processor, wherein the altitude-driven historical population spatial reconstruction program, when executed by the processor, implements the steps of the altitude-driven historical population spatial reconstruction method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores an elevation-driven historical population spatial reconstruction program, which, when executed by a processor, implements the steps of the elevation-driven historical population spatial reconstruction method as described in any one of claims 1-7.

Images