Method for analyzing urban spatio-temporal evolution based on multi-source historical data

Abstract

The application discloses a kind of urban space-time evolution analysis method based on multi-source historical data.Its steps are as follows:1) obtain and digitize multi-period historical city map, point of interest data and realistic painting;2) adopt the hybrid model of topology and angle distance to carry out syntax analysis on city road network and water system, calculate integrated degree and angle selection degree and other core indexes;3) based on optimization algorithm, normalizing treatment is carried out to space syntax index and POI data;4) extract dynamic city information from historical painting;5) statistical test is carried out to dynamic data extracted from painting and space syntax analysis result;6) using location quotient formula and kernel density estimation, analyze the space-time aggregation pattern and function evolution of public service facilities.The application builds a closed-loop research framework of multi-source data fusion, algorithm optimization and quantitative verification, and provides a highly reliable scientific method for revealing the internal driving mechanism of urban form evolution.

Description

Technical Field

[0001] This invention relates to the interdisciplinary fields of urban planning, geographic information science, computer vision, and history, and particularly to a method for analyzing the spatiotemporal evolution of cities based on multi-source historical and geographical data. This invention integrates and deeply mines historical maps, historical paintings, and multi-period point-of-interest (POI) data through scientific quantitative methods to reveal the complex, long-term interaction between urban spatial structure and socio-economic activities. Background Technology

[0002] Exploring the evolutionary patterns of urban morphology and social characteristics is a core issue for understanding the intrinsic logic of urban development and guiding future urban planning. Traditional research methods often rely on document interpretation, qualitative descriptions of images, or analysis based on a single data source. These methods have significant limitations in comprehensively, objectively, and quantitatively revealing the complex processes of urban development. Historical maps and paintings, as valuable visual archives of urban history, contain extremely rich information on spatial structure and human activities, but their unstructured and non-standardized characteristics pose significant challenges to quantitative analysis. At the same time, modern urban research increasingly demands higher precision, spatiotemporal scale, and multidimensional analysis of data. Traditional methods are inefficient when processing large-scale, multi-period, and multi-source heterogeneous data, and struggle to guarantee the scientific objectivity of conclusions.

[0003] Specifically, existing technologies suffer from the following bottlenecks:

[0004] 1. Limitations of Spatial Analysis Methods: Space Syntax, as a mature urban network analysis tool, has shown great potential in assessing urban spatial vitality. However, its classic model still has room for improvement when dealing with historical maps that reflect the cognitive habits of ancient people, especially in fully considering pedestrians' perception of changes in angle (rather than pure topological connectivity).

[0005] 2. Statistical Bias Issues in POI Data: When conducting spatial correlation analysis between Points of Interest (POI) data and street networks, directly associating discrete point data with line segments can easily lead to a highly skewed data distribution. This means a few streets may have a large number of POIs, while most streets may have few. This skewed distribution severely violates the fundamental assumptions of normality in data presented by classic statistical methods such as Pearson correlation analysis, potentially distorting the analysis results and failing to accurately reflect the true relationship between facility distribution and street centrality.

[0006] 3. Lack of cross-source data verification: Although advanced image recognition technologies such as YOLO have been applied in fields such as art detection, research on their systematic application in quantitatively extracting urban social activity information from historical paintings and conducting rigorous statistical cross-validation with spatial structure indicators analyzed from historical maps is currently lacking. This makes the analytical conclusions based on a single historical map lack corroboration from another independent information source, casting doubt on their reliability and authenticity.

[0007] 4. Ambiguity in Evolution Mechanisms: Existing research often fails to construct an integrated analytical framework to fully reveal the sustained impact of urban frameworks such as water systems and road systems on the development of urban functional areas, and the specific mechanisms by which urban socio-economic activities (such as commercial agglomeration) in turn shape historical urban forms and socio-spatial patterns. Summary of the Invention

[0008] To systematically overcome the aforementioned shortcomings of existing technologies in studying the spatiotemporal evolution of urban morphology and social characteristics, this invention proposes a method for analyzing urban spatiotemporal evolution based on multi-source historical and geographical data. The core objective of this invention is to construct a comprehensive research framework integrating data fusion, algorithm optimization, quantitative analysis, and cross-validation. This framework aims to scientifically and rigorously integrate and quantitatively analyze urban maps, points of interest (POI) data, and realistic paintings from different historical periods to deeply explore and quantify the structural roles of water and road systems in urban development, and how various socio-economic activities in cities trigger and respond to changes in urban morphology and social characteristics.

[0009] This invention provides the following technical solution:

[0010] The method for analyzing urban spatiotemporal evolution based on multi-source historical and geographical data includes the following steps:

[0011] Step S1, Data Collection and Preprocessing: Acquire and digitize city maps, corresponding Points of Interest (POI) data of public service facilities, and realistic paintings reflecting the urban landscape of the time for at least two historical periods within the study area. Each historical period spans no less than 50 years.

[0012] Step S2, Spatial Syntax Analysis: The road and water network in the digitized city map is analyzed using spatial syntax to calculate the integration degree and angle selectivity index.

[0013] Step S3, Quantification and Standardization of Public Service Facilities: Standardize the angle selection index and POI data to eliminate data skew and make them meet the requirements of statistical analysis;

[0014] Step S4, Painting Data Extraction and Verification: Using image recognition technology, dynamic urban information is extracted from the realistic painting and spatialized.

[0015] Step S5, Correlation Analysis: The dynamic urban information extracted in step S4 is correlated with the spatial syntactic integration degree and angle selectivity degree indices calculated in steps S2 and S3 to cross-validate the reliability of historical information.

[0016] Step S6: Use location quotient (LQ) analysis and kernel density estimation (KDE) to perform spatiotemporal evolution analysis on POI data from multiple periods;

[0017] Step S7, Comprehensive Analysis: Integrate the spatial syntactic integration degree and angle selectivity index characteristics, location quotient (LQ) analysis, and kernel density estimation (KDE) analysis results to reveal the spatiotemporal evolution mechanism of urban morphology and social characteristics.

[0018] Furthermore, the standardization process in step S3 includes:

[0019] a) Log-normalize the angular selectivity index ACH to obtain the angular selectivity standardized index NACH, which is calculated using the following formula:

[0020] ;

[0021] in, To select the angle, radius Total angular depth at the location;

[0022] b) The breadth-first search algorithm is used on the POI data, and a weighted summation calculation is performed in combination with distance decay weights to quantify the number of facilities within a specified topological radius. Then, a logarithmic transformation is performed to obtain the POI count standardization index NACO.

[0023] Furthermore, the image recognition technology in step S4 is the YOLO deep learning object detection algorithm.

[0024] Furthermore, the correlation analysis in step S5 is a Pearson correlation analysis, and its calculation formula is as follows:

[0025] ;

[0026] in, The correlation coefficient, and For sample values ​​of two variables, and This is the sample mean.

[0027] Furthermore, in step S6, the calculation formula for the location quotient (LQ) analysis is as follows:

[0028] ;

[0029] in, for area The number of POIs in the industry for The total number of POIs across all industries in the region. For the entire research area The number of POIs in the industry This represents the total number of Points of Interest (POIs) across all industries in the entire study area.

[0030] An electronic device includes: a memory, a processor, and a graphics processing unit (GPU); the memory is used to store computer-executable instructions, and the processor and GPU are used to execute the computer-executable instructions.

[0031] A computer-readable storage medium having computer-executable instructions stored thereon.

[0032] By employing the above-described technology, the beneficial effects of the present invention compared to the prior art are as follows:

[0033] 1) This invention achieves the scientific fusion and in-depth quantification of multi-source heterogeneous historical data: This invention innovatively integrates three vastly different data sources: historical maps, multi-period POI data, and realistic paintings with high artistic value. Through precise digitization, advanced YOLO image recognition technology, and optimized spatial syntax models, it achieves objective quantitative extraction and analysis of unstructured historical data (especially paintings), which is traditionally difficult to process. This greatly enriches the data dimensions of urban studies and compensates for the lack of depth and breadth in traditional research due to single data sources or reliance on qualitative analysis.

[0034] 2) An optimized spatial syntax algorithm tailored to the characteristics of historical data is proposed: Considering the characteristics of historical cognitive maps and the inherent complexity of POI data spatial distribution, this invention designs and verifies an optimized spatial syntax algorithm model. By introducing angle selection normalization (NACH) and POI count normalization (NACO) algorithms, the serious data skew problem arising when associating POI data with streets is scientifically resolved, making the processed data closer to a normal distribution, thus meeting the prerequisite for subsequent reliable statistical analysis. This significantly improves the accuracy, applicability, and scientific rigor of spatial syntax analysis in historical city research.

[0035] 3) A cross-validation mechanism based on independent historical information sources was constructed: This invention is the first to systematically apply the YOLO algorithm to historical paintings, quantitatively extracting key information reflecting the dynamic vitality of the city (commercial activities and population distribution), and then performing Pearson correlation analysis with the spatial syntactic analysis results based on historical maps. This step constitutes a cross-validation closed loop, that is, using an independent historical information source (painting) to verify conclusions derived from another information source (map), thereby significantly enhancing the reliability, authenticity, and scientific persuasiveness of the research results;

[0036] 4) This invention provides an in-depth study of the long-term evolution mechanism of cities: Utilizing mature spatial econometric models such as location quotient (LQ) analysis and kernel density estimation (KDE), this invention can quantitatively assess the structural contributions of water systems and road systems to urban development in different historical periods, and identify the spatial distribution patterns, agglomeration patterns, and coupling relationships between different types of public service facilities and street accessibility. By comparing and analyzing data from multiple periods, the historical trajectory of urban functional centers can be clearly outlined, thus revealing more comprehensively and deeply the intrinsic driving mechanisms of the spatiotemporal evolution of urban morphology and social characteristics.

[0037] 5) Providing scientific decision-making support for modern urban planning and heritage protection: This study, with a broader historical and spatial perspective and a complete set of quantitative research methods, systematically explores the distribution logic and development characteristics of various public service facilities in different periods. Its findings not only deepen the understanding of the historical evolution of specific cities but also provide a solid scientific basis for contemporary urban renewal, historical district protection, and cultural heritage revitalization planning practices, helping to guide future urban development to achieve sustainability while respecting historical context. Attached Figure Description

[0038] Figure 1 This is a flowchart of a method for analyzing urban spatiotemporal evolution based on multi-source historical and geographical data provided by the present invention;

[0039] Figure 2 This is a visualization of the integration index obtained after performing spatial syntactic analysis on a historical map in this embodiment of the invention.

[0040] Figure 3 In this embodiment of the invention, a visualization of a heatmap of temple type kernel density and integration degree superimposed is obtained after spatial syntactic analysis of a historical map.

[0041] Figure 4 In this embodiment of the invention, a visualization of a heatmap of school type kernel density and integration degree superimposed is obtained after spatial syntactic analysis of a historical map;

[0042] Figure 5In this embodiment of the invention, a visualization of a heatmap overlaid with government type kernel density and integration degree is obtained after performing spatial syntactic analysis on a historical map.

[0043] Figure 6 This is a schematic diagram illustrating how the YOLOv5 algorithm is used to automatically identify and label commercial activities and crowds from historical paintings in an embodiment of the present invention. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0045] Reference Figure 1 The specific steps of the urban spatiotemporal evolution analysis method based on multi-source historical and geographical data provided by this invention are as follows:

[0046] Step 1: Collection and Preprocessing of Multi-Source Heterogeneous Data

[0047] Taking Suzhou Ancient City as an example, this study acquired and digitized city maps, corresponding Points of Interest (POI) data for public service facilities, and realistic paintings reflecting the city's appearance at that time for three key historical periods (1745 during the Qianlong period of the Qing Dynasty, 1938, and 2019). As shown in Table 1:

[0048] .

[0049] Step 2: Spatial Syntax Analysis

[0050] To quantify the spatial structural attributes of Suzhou's ancient city, this study employs spatial syntax theory and uses Depthmap software to analyze the road and waterway networks across three periods. The core computational indicators are Integration and Choice. Integration measures the ease with which each street in Suzhou's ancient city connects to all other streets, reflecting its potential as a "destination." Choice measures the frequency with which a street appears on the shortest path between any two points, reflecting its "through" traffic potential. To adapt to historical maps of Suzhou and more realistically simulate pedestrian path selection, a hybrid computational model of topological distance and angular distance was used, ensuring that analyses across all periods were conducted within a unified evaluation framework to guarantee the scientific validity of longitudinal comparisons.

[0051] Step 3: Quantification and Standardization of Optimized Public Service Facilities

[0052] This step is a key technological innovation in this study, aiming to scientifically solve the problem of statistical bias that may occur when analyzing Suzhou data.

[0053] a) Angular Choice Standardization (NACH): To make the original Angular Choice (ACH) index calculated from the road network of Suzhou Ancient City more in line with the requirements of statistical analysis for data distribution, this study uses the following formula (1) for standardization:

[0054] Angular Choice Standardization (NACH): To make the original Angular Choice (ACH) index more consistent with the requirements of statistical analysis for data distribution, this invention uses the following formula (1) for standardization:

[0055] ;

[0056] In this formula, Indicates the radius The original angle selectivity within, Represents radius Total angular depth within. Original The values ​​often exhibit a long-tailed distribution, meaning they range extremely wide and are highly right-skewed. Directly using such data violates many assumptions of statistical models. The logarithmic transformation employed... ) is a standard variance-stabilizing transformation that can effectively compress the data range, significantly alleviate data skewness, and make its distribution closer to a normal distribution. Simultaneously logarithmizing the numerator and denominator is a form of relativistic standardization, thereby making the angle selection values ​​from different regions or periods comparable.

[0057] b) POI Count Normalization (NACO): To address the data skew caused by directly associating Suzhou's POIs with street lines, this study designed a breadth-first search (BFS) algorithm using Python programming (with the help of the Geopandas and Pandas libraries). This algorithm starts from each street segment in Suzhou's ancient city and searches for reachable POIs within a specified topological radius (e.g., 5 steps). To more accurately reflect the distance decay effect, a weighted summation method was adopted, where the weight of POIs farther away is lower (e.g., weight decays by a factor of 0.9 per step). The final weighted count value is then incremented by 1 and logarithmically transformed. This method not only avoids data skew but also makes its model assumptions (distance decay) more consistent with the first law of geography.

[0058] Step 4: Quantitative extraction and cross-validation of painting data

[0059] This study employs image recognition technology, using a dataset sample to train a target detection model for "shops" and "people." This dataset originates from slices of historical realistic paintings such as "Prosperous Suzhou" and "Emperor Qianlong's Southern Inspection Tour," and then uses LabelImg annotation software to manually select and label the two key elements, "shops" and "people," within the images. These precisely labeled samples form the basis for training and validating the YOLO-v5 model, aiming to enable the model to automatically identify and locate these two types of targets in ancient paintings with high accuracy. Dynamic urban information, such as commercial activities and population distribution, is extracted from the realistic paintings reflecting the urban landscape of Suzhou during the Qing Dynasty (Figure 2) and precisely spatialized onto the corresponding historical maps.

[0060] Step 5: Correlation Analysis

[0061] The core of this step is to perform cross-validation of multi-source data to verify whether the patterns revealed by data from different sources are consistent. This invention uses the classic Pearson Correlation Analysis. This method is used to measure the strength and direction of the linear relationship between two continuous variables. Its calculation formula is shown in (2):

[0062] ;

[0063] In this formula, and The sample values ​​represent two variables (NACH value and its corresponding NACO value for a street in Suzhou). and It is the sample mean of these two variables. It is the number of samples (total number of streets). The values ​​range from -1 to +1, where +1 indicates a perfect positive correlation, -1 indicates a perfect negative correlation, and 0 indicates no linear correlation. This invention examines the correlations between variables such as NACH and NACO, and NACH and crowd density in paintings, and sets a statistical significance level (usually 1). This invention aims to scientifically and quantitatively determine the strength of the correlation between urban spatial structure and socio-economic vitality. A high and significant correlation coefficient will strongly support the effectiveness of the method proposed in this invention.

[0064] Step 6: Location Quotient (LQ) Analysis

[0065] To reveal the evolution of urban functional zoning in different periods, this invention employs the classic indicator of location quotient (LQ) from economic geography. LQ is an indicator used to measure the concentration of a particular industry (here referring to a type of POI) in a specific region relative to a larger reference region (the entire city). Its calculation formula is shown in (3):

[0066] ;

[0067] In the application of this invention, refer to within the area The number of types of POIs yes The total number of all types of POIs within the region; It is in the entire city (reference area N) The total number of types of POIs This is the total number of all Points of Interest (POIs) in the entire city. When When, it indicates Type of facilities in The region's agglomeration level is higher than the urban average, which is a characteristic or dominant function of the region. When When the LQ value is high, it indicates that the facility is relatively sparse in the area. By calculating and comparing LQ values ​​at different times, the rise, decline, and spatial migration of various functional zones in the city can be accurately and quantitatively depicted.

[0068] Step 7: Kernel Density Estimation (KDE)

[0069] This invention employs kernel density estimation (KDE) to more intuitively understand the spatial distribution pattern of various public service facilities within the ancient city of Suzhou. KDE generates smooth, continuous density surfaces, clearly revealing the "hotspot" areas of various facilities at different times. Overlaying the KDE results with the integration degree map of spatial syntax allows for in-depth exploration of facility layout preferences: whether there is a tendency to choose the most accessible "central" areas, or other specific locational orientations. By overlaying the spatial distribution heatmap (KDE) of public service facilities with the structural accessibility (integration degree R20) of the urban road network and water system, the functional layout logic of the ancient city of Suzhou in 1745 is intuitively revealed. The overall spatial agglomeration pattern of all types of public service facilities is displayed. The figure clearly identifies two main "hotspot" areas within the city, namely the core agglomeration areas of public service functions. Different types of public service facilities (such as temples, education, and government institutions) are categorized and displayed, revealing the unique locational preferences and spatial distribution patterns of each type of facility. This map not only verifies the geographical accuracy of the computational model in this study, but also clearly shows that the areas of high concentration of urban public service facilities at that time highly overlapped with the core areas of integration identified by spatial syntax analysis, confirming the strong guiding role of urban morphological structure (high accessibility) on socio-economic functions (facility layout). Figure 6 ).

[0070] Step 8: Comprehensive Analysis and Conclusion

[0071] This invention analyzes historical and modern urban map data of Suzhou's ancient city and constructs a representative historical urban spatiotemporal dataset using the YOLO algorithm. Figure 2-5 Based on the aforementioned optimized spatial syntax algorithm, and utilizing Python libraries such as Geopandas and Numpy, this study provides an in-depth analysis of the impact of the waterway system on the historical development and urban form of Suzhou.

[0072] By applying multiple spatial econometric models to map data of Suzhou's ancient city from different periods, the following findings were obtained: 1) By controlling for waterways as a variable, the study quantitatively confirmed that waterways were an important component of Suzhou's historical urban system, playing a crucial role in urban development and activities, but were gradually neglected in modern times. 2) Different types of public service facilities exhibited varying accessibility and integration patterns with the water and road systems at different periods. 3) Changes in the distribution of public service facilities corresponded to the phased shifts in Suzhou's historical city center, which corroborated observations from realistic paintings and historical maps, demonstrating the authenticity and effectiveness of the research methodology.

[0073] This invention, through a broader historical and spatial perspective and supplemented by quantitative research methods, explores the distribution and development characteristics of various public service facilities in the ancient city of Suzhou in different periods, revealing the impact of water and road systems on urban development, as well as the historical urban form and social changes caused by urban activities.

[0074] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for analyzing the spatiotemporal evolution of cities based on multi-source historical and geographical data, characterized in that: Includes the following steps: Step S1, Data Collection and Preprocessing: Acquire and digitize city maps, corresponding Points of Interest (POI) data of public service facilities, and realistic paintings reflecting the urban landscape of the time for at least two historical periods within the study area. Each historical period spans no less than 50 years. Step S2, Spatial Syntax Analysis: The road and water network in the digitized city map is analyzed using spatial syntax to calculate the integration degree and angle selectivity index. Step S3, Quantification and Standardization of Public Service Facilities: Standardize the angle selection index and POI data to eliminate data skew and make them meet the requirements of statistical analysis; Step S4, Painting Data Extraction and Verification: Using image recognition technology, dynamic urban information is extracted from the realistic painting and spatialized. Step S5, Correlation Analysis: The dynamic urban information extracted in step S4 is correlated with the spatial syntactic integration degree and angle selectivity degree indices calculated in steps S2 and S3 to cross-validate the reliability of historical information. Step S6: Use location quotient (LQ) analysis and kernel density estimation (KDE) to perform spatiotemporal evolution analysis on POI data from multiple periods; Step S7, Comprehensive Analysis: Integrate the spatial syntactic integration degree and angle selectivity degree index characteristics, location quotient (LQ) analysis, and kernel density estimation (KDE) analysis results to reveal the spatiotemporal evolution mechanism of urban morphology and social characteristics; The standardization process in step S3 includes: a) Log-normalize the angular selectivity index ACH to obtain the angular selectivity standardized index NACH, which is calculated using the following formula: ； in, To select the angle, radius Total angle depth at the location; b) The breadth-first search algorithm is used on the POI data, and a weighted summation calculation is performed in combination with distance decay weights to quantify the number of facilities within a specified topological radius. Then, a logarithmic transformation is performed to obtain the POI count standardization index NACO.

2. The urban spatiotemporal evolution analysis method based on multi-source historical and geographical data according to claim 1, characterized in that, The image recognition technology in step S4 is the YOLO deep learning object detection algorithm.

3. The urban spatiotemporal evolution analysis method based on multi-source historical and geographical data according to claim 1, characterized in that, The correlation analysis in step S5 is Pearson correlation analysis, and its calculation formula is as follows: ； in, The correlation coefficient is... and For sample values ​​of two variables, and This is the sample mean.

4. The urban spatiotemporal evolution analysis method based on multi-source historical and geographical data according to claim 1, characterized in that, In step S6, the calculation formula for the location quotient (LQ) analysis is as follows: ； in, for area The number of POIs in the industry for The total number of POIs across all industries in the region. For the entire research area The number of POIs in the industry This represents the total number of Points of Interest (POIs) across all industries in the entire study area.

5. An electronic device, characterized in that, include: The memory, processor, and graphics processing unit (GPU) are provided; the memory is used to store computer-executable instructions, and the processor and GPU are used to execute the computer-executable instructions to implement the method as described in any one of claims 1-4.

6. A computer-readable storage medium, characterized in that, It stores computer-executable instructions that, when executed by a processor, enable the implementation of the method as described in any one of claims 1-4.

Images