Street space livable measurement method under multi-source data fusion

By using multi-source data fusion and comprehensive evaluation technology, the problems of subjectivity and insufficient data quantification in the evaluation of urban spatial livability have been solved, enabling scientific assessment and guiding transformation of the livability of existing spaces, thereby improving spatial quality and resident satisfaction.

CN121920864APending Publication Date: 2026-04-24YUNNAN ARTS UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310846677.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-07-11
Publication Date
2026-04-24

AI Technical Summary

Technical Problem

Existing technologies for evaluating urban spatial livability suffer from problems such as strong subjectivity, lack of consistent standards, insufficient data quantification, and insufficient micro-perspective, resulting in insufficient scientific rigor and guidance in the evaluation results.

Method used

A multi-source data fusion approach is adopted, which collects street view data, spatial syntax data and wind environment data through active sensing technology. Combined with machine learning and CFD numerical simulation, comprehensive evaluation and feature mining are carried out. The entropy weight method and analytic hierarchy process are used to determine the index weights, so as to achieve an objective quantitative assessment of the livability of existing spaces.

Benefits of technology

It provides a more scientific and accurate assessment of urban spatial livability, which can guide the micro-renewal of existing spaces, improve spatial quality, and increase resident satisfaction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121920864A_ABST
    Figure CN121920864A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of urban planning, in particular to a street space livable measurement method under multi-source data fusion, which adopts an active sensing technology to carry out quantitative measurement research on street view data, space syntactic data and wind environment data in a space based on a human livable sensing view angle. On the basis of a human-based space livability evaluation index, an evaluation index weight and an index evaluation standard are further perfected, and a new scientific basis is provided for stock space update and development. For urban design practice requirements of urban micro-update under the human scale, the street space quality research needs continuous, large-scale and high-precision street space characteristic measurement instead of a few street section surfaces. Based on this, the research data of the method has good geographic position information attribute characteristics, rapid and accurate evaluation of the space livable quality is realized, and the research purpose of truly guiding human-scale street space quality micro-update practice is achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of urban planning technology, specifically to a method for measuring the livability of street spaces based on multi-source data fusion. Background Technology

[0002] As times have progressed, urban livability has seen new focuses and research dimensions regarding people's experiences within spaces. With advancements in data acquisition and analysis technologies, research on human perception of space has gradually delved into the level of subjective feelings. This study primarily focuses on the quantitative analysis of subjective feelings about livability in physical spaces.

[0003] Limited data source fusion technology refers to a technology that couples and matches collected data with spatial simulation data based on the geographic information characteristics of the data and the survey method of actively sensing spatial climate characteristics by sensors.

[0004] Quantitative analysis uses concrete data to represent vague and ambiguous factors, thereby achieving the purpose of data scoring and comparison. This study will focus on the evaluation indicators constructed through quantitative analysis, namely primary and secondary indicators, to make the final evaluation results more objective and scientific.

[0005] Livability analysis based on small-scale urban residential plots can characterize urban planning and development issues at a micro-scale, facilitating targeted adjustments and optimizations. In the past, limitations in data collection technology and the lack of real-time data on existing spaces meant that discussions of urban livability at the micro-scale were limited, and often confined to static analyses of single time periods, lacking dynamic and real-time monitoring and evaluation. Specifically:

[0006] The current research on spatial habitability is largely based on qualitative descriptions and is highly subjective. Even though some studies use the AHP (Analytic Hierarchy Process) algorithm to quantitatively evaluate spatial habitability, the results are subject to subjective fluctuations due to the inherent human factor in the algorithm's subjective assignment.

[0007] The lack of a comprehensive system has led to the adoption of quantitative research in assessing spatial livability. To obtain more objective evaluation results, different weights were assigned to multiple indicators during the research, thereby reducing the volatility caused by subjective human influence. Currently, the method for determining indicator weights mostly adopts the objective weighting method. However, due to the inherent characteristics of livability indicators, there are extremely large indicators, extremely small indicators, intermediate indicators, and interval indicators. This results in a lack of consistent standard expression for the final indicator evaluation, making the interpretation of the conclusions somewhat difficult.

[0008] The "insufficient quantification" stems from the fact that many previous indicators of spatial livability were based on macro and meso perspectives. While these indicators could effectively reflect the patterns of various regional spatial livability indicators, and some livability studies also touched upon the micro perspective of space, they mostly focused on aspects such as the distribution of public facilities, road network density, and per capita personal economic status. Few studies addressed the issue of urban micro-renewal, resulting in conclusions that lacked specificity in providing practical guidance for spatial micro-renewal. Summary of the Invention

[0009] The purpose of this invention is to provide a method for measuring the livability of street spaces under multi-source data fusion. For existing spaces, data simulation analysis technology is used to realize the human-centered perception identification of the livability of existing spaces. Then, combined with data on street livability, a coupled analysis is performed to explore the livability characteristics of existing spaces, providing a scientific operation method for the micro-update of existing spaces.

[0010] To achieve the above-mentioned technical objectives and effects, the present invention is implemented through the following technical solution:

[0011] A method for measuring the livability of street space based on multi-source data fusion includes: using active sensing technology to conduct quantitative measurement research on street view data, spatial syntax data and wind environment data in the space based on the perspective of human-centered livability perception, realizing a human-centered livability evaluation index system within the region, which can be used to adjust spatial micro-renewal practices.

[0012] Furthermore, the street view data measurement specifically includes the following steps:

[0013] 1.1 Street view data collection: Taking street view image data of the study area as the research object, the study aims to achieve intuitive and visual display of the constituent elements and proportions of street view image data.

[0014] 1.2 To create a sample dataset, the input images were randomly cropped to 500 pixels × 500 pixels. Using Photoshop, the images were further subdivided into 10 categories, including roads, buildings, cars, pedestrians, signs, utility poles, and sidewalks. A Photoshop sample set of 500 images was generated by manually annotating them.

[0015] 1.3 Different colors are used to divide each element. A color swatch is created using fixed RGB colors, and the paint bucket tool is used to segment the artificial architectural elements. Secondly, to ensure that each color corresponds to a value in a one-dimensional vector, the colors in the image need to be encoded, reducing the three-dimensional representation to two-dimensional. This can be achieved by treating BGR colors as three-digit numbers in base 256 and converting them to decimal representations.

[0016] 1.4 Convert the colormap to a two-dimensional matrix and remove duplicate numbers to obtain the color table. Finally, put the processed label images into the program and run it to obtain semantic segmentation data and the proportion of each element in the street view image. By combining street view data with machine learning, data visualization is achieved, solving the previous problem of difficulty in batch collection of street view data. It can quickly realize an intuitive display of the constituent elements and proportions of street view image data.

[0017] Furthermore, the spatial syntactic data measurement specifically includes the following steps:

[0018] 2.1 Data Preprocessing: Collect street-related geospatial information and textual descriptions, and integrate them into structured data. Geospatial information may include the street's geographical location, area, road width, etc., while textual descriptions may include descriptions of the street's history, culture, commerce, etc.

[0019] 2.2 Spatial Relationship Modeling: The geospatial information of the streets is modeled as an axis representation.

[0020] 2.3 Spatial Relationship Analysis: Based on the spatial relationship model, the spatial relationships between various geographic spaces in the street are analyzed, and combined with textual description information for analysis, thereby determining information such as the division of areas, major attractions, and commercial areas in the street.

[0021] 2.4 Spatial Quantification: Based on the results of spatial relationship analysis, quantitative analysis is conducted on the spatial structure, landscape, and clusters of the streets.

[0022] 2.5 Syntactic Analysis: Based on the spatial structure and context of the street, explain the historical and cultural connotations and commercial value of the street, and provide recommendations for shopping, entertainment and other services.

[0023] Furthermore, the wind environment data measurement specifically includes the following steps:

[0024] 3.1 Study the climate characteristics of the area where the existing space is located and the characteristics of wind flow around the building. Compare relevant regulations and wind environment evaluation standards proposed by domestic and foreign scholars to determine wind environment evaluation standards in four aspects: wind speed, wind speed ratio, wind pressure and wind chill temperature.

[0025] 3.2 Understand the CFD numerical simulation process, determine the simulation requirements such as turbulence model, discretization and solution methods, and boundary condition settings, and verify the CFD numerical simulation by combining typical wind tunnel tests conducted by foreign scholars to ensure the accuracy of the turbulence model and solution method selected in this invention;

[0026] 3.3 Conduct a survey of the study area and summarize common building cluster layouts; simulate the wind environment around individual buildings under different conditions of cross-sectional shape, height, and wind direction angle, as the basis for subsequent research on wind environment simulation around different building layouts.

[0027] 3.4 Based on the survey results, the wind environment around the residential area in winter and summer was simulated for different layout forms. The simulation results were compared and analyzed to find the most suitable layout forms for winter and summer.

[0028] 3.5 Simulate and analyze the existing spatial environment and conduct a wind environment assessment, while also considering the impact of public buildings near the settlement on the wind environment.

[0029] Furthermore, the street view data evaluation, i.e., semantic segmentation quantization, includes green view rate and sky rate; the semantic segmentation quantization score is shown in Table 1:

[0030] Table 1 Semantic Segmentation Quantitative Scoring Table

[0031]

[0032] Furthermore, the spatial syntax data evaluation includes, but is not limited to, connectivity, global integration, local integration, selectivity, synergy / comprehensibility, horizon connectivity, and horizon integration. The spatial syntax scores are shown in Table 2.

[0033] Table 2 Spatial Syntax Scoring Table

[0034]

[0035]

[0036]

[0037] Furthermore, the wind environment data evaluation includes, but is not limited to, the wind speed ratio in the comfortable wind zone, the wind speed ratio in the calm wind zone, the wind speed ratio in the strong wind zone, air age, and PMV. The wind environment data scores are shown in Table 3.

[0038] Table 3 Wind Environment Data Scoring Table

[0039]

[0040]

[0041]

[0042]

[0043] The beneficial effects of this invention are:

[0044] This invention proposes a method for measuring the livability of street spaces based on multi-source data fusion. Building upon a human-centered perspective on spatial livability evaluation indicators, it further refines the weighting and judgment criteria of these indicators, providing new scientific basis for the renewal and development of existing spaces. Facing the needs of urban design practices at the human-centered scale in urban micro-renewal, street space quality research requires continuous, large-scale, and high-precision measurements of street space characteristics, rather than just a few street cross-sections. Therefore, the data in this study all possess good geographical location information attributes, enabling rapid and accurate evaluation of spatial livability quality and achieving the research objective of truly guiding human-centered micro-renewal practices in street space quality.

[0045] 1. Multi-Source Data Fusion Evaluation: This technical solution employs active sensing technology to comprehensively evaluate multiple aspects of street space. Street view data provides visual perception of the street's physical environment, building structure, and landscape features; spatial syntax data provides information on street layout, connectivity, and overall structure; and wind environment data reveals the impact of wind speed, wind pressure, and air quality on the living environment. By actively collecting multi-source data from the study area using multi-carrier adaptive sensors, accurate real-time data is obtained. Combined with data cleaning, processing, and statistical analysis based on the data's characteristics, a comprehensive consideration of people's perception of livability in different aspects can be achieved.

[0046] 2. Human-centered perception identification of the livability of existing spaces: Through data simulation and analysis techniques, the livability of existing spaces can be evaluated. Specifically, street view data can be used to extract elements such as buildings, roads, sidewalks, and green spaces through image processing and analysis techniques, and their distribution and proportional relationships can be analyzed; spatial syntax data can be used to analyze the spatial structure of streets, block divisions, and the location of attractions; wind environment data can reveal the livability of streets in terms of wind speed, wind pressure, and air quality. By integrating this information, the livability characteristics of existing spaces can be identified, such as green view ratio, sky view ratio, connectivity value, and integration degree.

[0047] 3. Coupled Analysis and Feature Mining: Coupled analysis of data on human-centered livability and street-level livability can further uncover the livability characteristics of existing spaces. For example, by performing correlation analysis on street view data and spatial syntax data, the correlation patterns between different elements can be discovered, such as the continuity between buildings and roads, and the spatial layout of scenic spots and commercial areas. The SegNet (semantic segmentation) machine learning algorithm is used for semantic analysis of street view data; the Depthmap software is used for quantitative research on the topological relationships of street spatial elements at the graphical level; and the PHOENICS wind environment simulation software is used for digital simulation analysis of various space-related microclimate factors. The mining of these features can provide scientific operational methods and decision-making basis for the micro-updates of existing spaces.

[0048] 4. Scientific Operational Methods: The extracted spatial human-centered data undergoes data forwarding, standardization, and normalization to facilitate the collection, preprocessing, and analysis of street view data. An entropy weighting method is used for objective weighting, while the analytic hierarchy process (AHP) is employed for subjective weighting analysis. The weight values ​​obtained from both methods are then averaged to determine the weight values ​​for each indicator, minimizing subjective interference. For example, during street view data collection, the specific elements of the street view image data in the study area are analyzed and visualized, intuitively displaying the constituent elements and proportions of the street view image data. Simultaneously, in the measurement and analysis of spatial syntactic data, techniques such as axis representation and spatial relationship analysis are used to accurately quantify the spatial structure and landscape characteristics of the streets. These operational methods ensure the accuracy and reliability of the measurement results.

[0049] 5. Scientific Basis for Practical Guidance: Through the fusion and evaluation of multi-source data and the mining of livability characteristics, this technical solution can provide a scientific basis for practical guidance in the micro-renewal of existing spaces. Specifically, by comprehensively analyzing information from different data sources, hidden livability characteristics and potential in existing spaces can be discovered, providing scientific support and decision-making references for the renewal and development of existing spaces. Based on a human-centered perspective, an evaluation matrix using "wind environment data," "street view data," and "spatial syntax data" as dimensions is used to classify the livability of spaces into quality categories: high-quality, high-comfort spaces (90-100), high-quality, low-comfort spaces (70-90), low-quality, high-comfort spaces (60-70), and low-quality, low-comfort spaces (60-0). These evaluation results and feature mining methods can guide the micro-renewal practices of street space quality, improving people's perception and satisfaction with the urban environment.

[0050] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0051] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a schematic diagram of the overall framework of the present invention;

[0053] Figure 2 This is a schematic diagram illustrating the principle of spatial syntactic data measurement in this invention;

[0054] Figure 3 This is a schematic diagram illustrating the principle of wind environment data measurement in this invention.

[0055] Figure 4 This is a schematic diagram illustrating the street view data measurement principle of the present invention;

[0056] Figure 5 This is a schematic diagram illustrating the quantitative analysis of street view data according to an embodiment of the present invention;

[0057] Figure 6 This is a schematic diagram of semantic quantification index analysis as described in an embodiment of the present invention.

[0058] Figure 7 This is a schematic diagram of the street view data collection distribution according to an embodiment of the present invention;

[0059] Figure 8 This is a schematic diagram of the connection value analysis described in an embodiment of the present invention;

[0060] Figure 9 This is a schematic diagram of the global integration degree analysis described in an embodiment of the present invention;

[0061] Figure 10 This is a schematic diagram of the local integration degree analysis described in an embodiment of the present invention;

[0062] Figure 11 This is a schematic diagram of the selectivity analysis described in an embodiment of the present invention;

[0063] Figure 12 This is a scatter plot of the synergy analysis described in the embodiments of the present invention;

[0064] Figure 13 This is a schematic diagram (a) of the field integration analysis described in an embodiment of the present invention;

[0065] Figure 14 This is a schematic diagram (b) of the field integration analysis described in an embodiment of the present invention;

[0066] Figure 15These are cloud maps of wind speed, air age, and PMV in different regions as described in the embodiments of the present invention.

[0067] Figure 16 This is a schematic diagram of image coupling analysis according to an embodiment of the present invention.

[0068] Figure 17 Figure a shows the design node results as described in this embodiment of the invention;

[0069] Figure 18 Figure b shows the design node results as described in this embodiment of the invention;

[0070] Figure 19 Figure c shows the design node results as described in this embodiment of the invention. Detailed Implementation

[0071] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0072] Example 1

[0073] like Figure 1-5 As shown

[0074] This embodiment describes a method for measuring the livability of street spaces based on multi-source data fusion. It employs active sensing technology to conduct quantitative measurement research on spatial data such as street view data, spatial syntax data, and wind speed from a human-centered livability perception perspective. This results in a human-centered livability evaluation index system for a region, guiding spatial micro-renewal practices. The technical solutions used for the three types of data are as follows:

[0075] 1. Street View Data

[0076] Using street view image data of the study area as the research object, this study aims to achieve an intuitive and visual display of the constituent elements and proportions of street view image data. Input images are randomly cropped to 500 pixels × 500 pixels. Photoshop is used to subdivide the data into 10 categories, including roads, buildings, cars, pedestrians, signs, utility poles, and sidewalks. A Photoshop sample set of 500 images is generated through manual annotation. Different colors are used to classify each element, and a fixed RGB color palette is created. The paint bucket tool is used for manual segmentation of building elements. Secondly, to ensure each color corresponds to a value in a one-dimensional vector, the colors in the image are encoded, reducing the three-dimensional representation to two-dimensional. BGR colors can be viewed as three-digit numbers in base 256 and converted to decimal representation. Thirdly, the colormap is converted into a two-dimensional matrix and duplicate numbers are removed, resulting in the color table. Finally, the processed labeled images are run in the program to obtain semantic segmentation data and the proportion of each element in the street view image. By combining street view data with machine learning, data visualization is achieved, solving the problem of difficult batch collection of street view data and enabling a rapid and intuitive display of the constituent elements and proportions of street view image data.

[0077] 2. Spatial Syntax

[0078] 2.1 Data Preprocessing: Collect street-related geospatial information and textual descriptions, and integrate them into structured data. Geospatial information may include the street's geographical location, area, road width, etc., while textual descriptions may include descriptions of the street's history, culture, commerce, etc.

[0079] 2.2 Spatial Relationship Modeling: The geospatial information of the streets is modeled as an axis representation.

[0080] 2.3 Spatial Relationship Analysis: Based on the spatial relationship model, the spatial relationships between various geographic spaces in the street are analyzed, and combined with textual description information for analysis, thereby determining information such as the division of areas, major attractions, and commercial areas in the street.

[0081] 2.4 Spatial Quantification: Based on the results of spatial relationship analysis, quantitative analysis is conducted on the spatial structure, landscape, and clusters of the streets.

[0082] 2.5 Syntactic Analysis: Based on the spatial structure and context of the street, explain the historical and cultural connotations and commercial value of the street, and provide recommendations for shopping, entertainment and other services.

[0083] This street space quantification technology based on spatial syntax can help people better understand the geospatial structure and commercial value of streets, improving business operation efficiency and urban planning and management efficiency. At the same time, this technology can also be applied to other geospatial analysis fields, such as urban planning and traffic management.

[0084] 3. Wind Environment Data

[0085] 3.1 Study the climate characteristics of the area where the existing space is located and the characteristics of wind flow around the building. Compare relevant regulations and wind environment evaluation standards proposed by domestic and foreign scholars to determine wind environment evaluation standards in four aspects: wind speed, wind speed ratio, wind pressure and wind chill temperature.

[0086] 3.2 Understand the CFD numerical simulation process, determine the simulation requirements such as turbulence model, discretization and solution methods, and boundary condition settings, and verify the CFD numerical simulation by combining typical wind tunnel tests conducted by foreign scholars to ensure the accuracy of the turbulence model and solution method selected in this invention;

[0087] 3.3 Conduct a survey of the study area and summarize common building cluster layouts; simulate the wind environment around individual buildings under different conditions of cross-sectional shape, height, and wind direction angle, as the basis for subsequent research on wind environment simulation around different building layouts.

[0088] 3.4 Based on the survey results, the wind environment around the residential area in winter and summer was simulated for different layout forms. The simulation results were compared and analyzed to find the most suitable layout forms for winter and summer.

[0089] 3.5 Simulate and analyze the existing spatial environment and conduct a wind environment assessment, while also considering the impact of public buildings near the settlement on the wind environment.

[0090] All three types of data have geographic information attributes. By using data overlay, we can conduct data coupling and visualization research on the existing spatial environment of the study area, thereby discovering the problems existing in the existing space, and then using corresponding design intervention methods to update and optimize the spatial environment.

[0091] This method integrates street view data, spatial syntax data, and wind environment data to analyze the human perception of livability in existing spaces from multiple perspectives. It also utilizes automated tools to construct a data collection process, providing a rich and reliable data source for subsequent research on existing spaces and improving the systematic nature of data collection in existing space research to a certain extent.

[0092] Based on multi-source data, a livability evaluation system for existing spaces is constructed, providing a basic scientific basis and reference for proposing renewal and development plans for existing spaces. On the basis of the livability evaluation system for existing spaces, the artistic attributes of local spaces are fully explored by introducing artistic intervention methods, thereby designing relevant protection and development plans, driving the enthusiasm of local residents to participate and bringing new vitality to existing spaces, thus helping relevant departments to better protect and develop existing spaces.

[0093] Multi-carrier adaptive sensors are used to actively collect multi-source data from the study area to obtain accurate real-time data. Based on the characteristics of the data itself, relevant data cleaning, sorting and statistical work are carried out.

[0094] The SegNet machine learning algorithm was used to perform semantic analysis on street view data; the Depthmap software was used to conduct a quantitative study of the topological relationships of street spatial elements at the graphical level; and the PHOENICS wind environment simulation software was used to perform digital simulation analysis of various spatially related microclimate factors.

[0095] The extracted spatial human-centered data is processed by data positiveization, standardization and normalization. The entropy weight method is used for objective weighting, and the analytic hierarchy process is used for subjective weighting analysis. The weight values ​​obtained by combining the two are averaged to determine the weight values ​​of each indicator, so as to minimize subjective interference factors.

[0096] Example 2

[0097] Taking the Jinning Ancient City Street Renewal Project as an example

[0098] 1.1 Street view data, such as Figure 5-7 As shown

[0099] 1.1.1 Sample Dataset Creation

[0100] 1.1.1.1 Street View Data Collection

[0101] Since there was no street view data for Jincheng Ancient Town, street view videos were manually filmed and then extracted frame by frame using Premiere Pro to obtain street view images.

[0102] 1.1.1.2 Create a sample dataset using the collected street view data;

[0103] The images were further subdivided into 10 categories, including roads, buildings, cars, pedestrians, signs, utility poles, and sidewalks. A Photoshop sample set of 500 images was generated using manual annotation. The street view images and sample labels were randomly cropped to 500 pixels x 500 pixels and then fed into the program.

[0104] 1.1.2 Semantic Analysis Based on Street View Images

[0105] 1.1.2.1 Input the street view images and corresponding sample labels into the program;

[0106] Convert the colormap to a two-dimensional matrix and remove duplicate numbers to obtain the color table. Then, categorize the processed label images and street view images into corresponding folders, and run them in a Python program. First, configure the Python environment; after successful configuration, run the program.

[0107] 1.1.2.2 Semantic Analysis Results

[0108] The program can effectively identify the types of elements in an image, such as sky, sidewalks, lanes, buildings, and greenery, and obtain the cutout and proportion of each element. After processing street view images, the program can obtain the proportion of each semantic component of the street view in the entire street. Based on the images used in the experiment, it can be concluded that unreasonable street areas should be redesigned based on the analyzed semantics. The proportion of greenery yields the green view rate of nodes, the proportion of sky represents its sky rate, the proportion of people represents vitality value, the proportion of windows represents transparency, and the proportion of public facilities yields the occupancy rate of public facilities. By analyzing these proportion values, design intervention can be carried out. By leveraging the spatial grounding of street views, this invention not only discovered the number of strong and weak perception zones in Jincheng Ancient Town, but also obtained the specific landing points of these zones. These landing points are the focus of the next design step of this invention. In areas with weak imagery, this invention will design to address the problem; in areas with strong imagery perception, the advantages should be leveraged to strengthen the characteristics.

[0109] 1.2 Spatial Syntax Data

[0110] Space syntax-based redesign refers to the process of analyzing spatial syntax data of a space, utilizing the principles and rules of space syntax, and proposing improvement measures and redesigning the space to address its existing problems and needs. Specifically, space syntax-based redesign may include the following steps:

[0111] 1.2.1 Collect spatial syntax data: Collect spatial syntax data of spatial locations through field surveys, remote sensing technology, GIS technology, etc., including information such as the location, shape, size, and relationship of spatial elements.

[0112] 1.2.2 Analyze spatial syntax data: By analyzing spatial syntax data, we can understand the problems and needs of spatial locations, such as the connectivity, accessibility, and organization of spatial elements, as well as their defects and shortcomings.

[0113] 1.2.3 Develop improvement strategies: Based on the principles and rules of spatial syntax, develop improvement strategies to address the problems and needs of spatial locations, such as optimizing the layout of spatial elements, improving traffic organization, and increasing public facilities.

[0114] 1.2.4 Conduct modification design: Based on the established improvement strategy, conduct modification design, including specific design schemes, post-modification syntactic reanalysis, etc.

[0115] 1.2.5 Syntactic Evaluation: After the renovation work is completed, a syntactic comparison study is conducted on the renovation effect before and after to understand whether the renovation effect has achieved the expected goal, and whether there are any new problems and needs, so as to further optimize the design and management of the space.

[0116] In conclusion, the design of space syntax-based transformation is a systematic process that requires combining practical situations and professional knowledge with scientific methods and technologies to optimize and improve spatial environments.

[0117] Optimize the visual representation based on the syntactic results, and then conduct on-site research and comparative analysis. Combining the two can intuitively reflect the internal problems of the spatial pattern, as shown in Figure 8-13:

[0118] Connectivity analysis can intuitively reflect the permeability of space. The number of all other nodes directly connected to a given node is its connectivity value. It is a special quantitative value that can be obtained without calculation in the system. From the perspective of spatial cognition, the larger the connectivity value of a node, the more neighboring spaces a person can see within that space. In other words, it indicates that the node has better spatial permeability.

[0119] Global integration reflects the degree of clustering or dispersion of a space relative to other spaces in a system. It represents the degree of connection between one axis and other axes, reflecting the accessibility of a space. It measures a space's potential to attract arrival traffic and embodies the centrality of that space. Areas with higher global integration have greater potential to attract traffic.

[0120] Local integration (typically considering a radius of 3) expresses the reachability of each axis to other axes within a maximum of 3 topological steps.

[0121] Choice refers to the frequency with which a particular element in a spatial system is the shortest topological distance between two nodes. It examines the advantages of a spatial unit as the shortest path for travel and reflects the likelihood of a space being traversed. A space with a higher choice is more likely to be traversed by people, so choice is also called traversability.

[0122] Synergy is a variable describing the correlation between the integration degree at radius 3 and the integration degree at radius n. It reflects whether the local space of the area contributes to the understanding of the overall spatial morphology and structure, and whether the overall and local spaces of the village are consistent. In the scatter plot generated by Depthmap software, the horizontal and vertical axes are set as global integration and local integration, respectively, forming a straight line. The sum of the distances from all points in the scatter plot to this line is minimized, and this line most closely simulates the trend of the scatter plot. R² is called the goodness of fit, used to indicate the degree of fit between the line and the scatter plot. Generally, an R² of 0.5 or higher is considered a good fit, and an R² of 0.7 or higher is considered a very good fit. Spatial structures with high synergy often exhibit single-core structural characteristics, while those with low synergy exhibit multi-core structural characteristics. High synergy indicates that the local center can be well integrated into the entire spatial system; for example... Figure 12 .

[0123] The viewport connectivity value means the total number of other elements seen when looking outwards from a given element. This sum is fed back and recorded above the element's header. After this process is done for every element in the system, each element will have a Connectivity value. Then, based on the Connectivity values, elements are divided into ten ranges from high to low. A specific element is displayed with the corresponding color depending on which range it falls into, thus generating... Figure 13 .

[0124] Visual Integration [HH], or visual integration degree, indicates that a higher value means that the element requires fewer turns to see other elements in the system. A lower value means that more turns are needed to see other elements from this element. Conversely, a higher Visual Integration [HH] value means that it is easier to see from any position in the system. A higher Visual Integration [HH] value makes an element more attractive and easier to see. A lower value makes it less attractive.

[0125] 1.3 Wind Environment Data

[0126] 1.3.1 Sample Dataset Creation

[0127] 1.3.1.1 Data Acquisition

[0128] During actual measurements, fixed-point measurements were selected, with multiple detection points set up within the village. Since this invention primarily tests wind speed data within the village, a handheld split-type anemometer was used. The measurement height was 1.5 meters above the ground. During actual measurements, the anemometer was fixed at a height of 1.5 meters to record the site's wind environment. Due to the significant influence of the external environment, hourly recording was adopted. The selection of measurement points included the main roads, activity spaces, water bodies, and green spaces of Jinning Ancient City—spaces frequently used by village residents and where outdoor activities are common. The aim was to compare the consistency between the simulated and actual wind environments.

[0129] 1.3.1.2 Simulation analysis is performed using the collected wind direction data;

[0130] To better understand the characteristics of wind environment under different environmental factors within the ancient city, the wind environment at different measuring points at different time periods within the ancient city was analyzed. The average value of the collected measured data was taken, and a line graph was used to express the differences between each measuring point and the differences between the measured wind speed and the simulated wind speed.

[0131] 1.3.2 Analysis based on numerical simulation

[0132] 1.3.2.1 Input the exported simulated image into the recognition program and run it;

[0133] To ensure the accuracy of the identification process, the image was cropped to... Figure 11 The data is input into the recognition software in the form of a color table. The color range from blue to red corresponds to 0-15 in the table generated by the recognition program. Figure 12 Then, a line graph was used to compare the percentage of each color range in the wind speed cloud map generated by the recognition program.

[0134] 1.3.2.2 Analysis, Comparison, and Scoring

[0135] like Figure 15As shown, the software exports wind speed cloud maps, air age, and PMV cloud maps for different intervals for scoring. Wind speed cloud maps in the 0-4 m / s interval are used to score the wind speed ratio in comfortable wind zones and strong wind zones, with every four small intervals forming a large interval. Wind speed cloud maps in the 0-4.8 m / s interval are used to score the wind speed ratio in calm wind zones, with every two small intervals forming a large interval. Both methods use the highest and lowest intervals for scoring, and the final score is the average of the two. The overall wind environment of Jinning Ancient Town is judged based on the scoring. The results demonstrate that the wind environment quality in Jinning Ancient Town is poor, making wind environment-based design a meaningful approach. Leveraging the efficiency of wind environment simulation, this invention not only identified areas with poor wind environment quality in Jinning Ancient Town but also identified specific areas for improvement. These areas represent the focus of the next design phase of this invention. For areas with poor wind environment quality, this invention addresses the specific problems; for areas with better wind environment quality, the invention should leverage its advantages and strengthen its characteristics.

[0136] 1.4 Three types of data were used to conduct quantitative analysis of the study area, thereby quantitatively evaluating the environmental conditions of the area. Image coupling was used to highlight and analyze the problems.

[0137] like Figure 16-19 This invention aims to utilize space syntax theory to conduct a detailed analysis of the historical and cultural district of Jincheng, Kunming, and provide targeted micro-renewal designs to enhance its spatial sense and cultural appeal.

[0138] Based on the analysis of spatial syntactic data, and combining seven relevant parameters including connectivity, integration, selectivity, and visibility analysis, the analysis revealed six nodes within the renovation area that require optimization and improvement. These nodes all suffer from poor spatial accessibility, lack of cultural elements, and poor environmental quality. To address these issues, the design employs two micro-update strategies: adding culturally distinctive facilities to enhance interaction between the nodes and their surroundings and improve accessibility; and optimizing internal elements of the nodes, such as renovating the appearance of existing shops and adding greenery, to enhance their cultural appeal. The design selected two nodes for detailed syntactic analysis and micro-update design methods, enabling the Kunming Jincheng Historical and Cultural District to maintain its historical and cultural atmosphere while providing a better tourist experience and driving economic development in the surrounding area.

[0139] This study utilizes semantic segmentation to analyze street view image data, examining the objective composition of elements within the underlying interface of commercial streets. Traditional blocks in ancient towns represent a unique spatial form developed under specific social and cultural environments, and their constituent elements embody their historical and cultural connotations. By analyzing the characteristics of traditional block architecture, this research primarily focuses on three semantic elements—sky view ratio, green view ratio, and permeability—to provide guidance and suggestions for the design of the underlying interface of commercial streets, offering a new approach to the inheritance and promotion of outstanding architectural culture among Yunnan's ethnic minorities.

[0140] In the future, it is still necessary to further improve the accuracy of element classification and the precision of maps, combine ethnic minority traditional culture to develop the architectural decoration and craftsmanship of traditional blocks, and use semantic element analysis as the basis and premise of design to make the design of Jincheng more scientific. While promoting the distinctive culture of ethnic minorities, it can also attract people and achieve the common development of culture and economy, avoid the commercialization of Jincheng, and thus effectively promote the coordinated and sustainable development of tourism and local ethnic minority culture.

[0141] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for measuring the livability of street spaces based on multi-source data fusion, characterized in that, include: Active sensing technology is used to conduct quantitative measurement research on street view data, spatial syntax data and wind environment data in the space based on the perspective of human-centered livability perception, so as to realize the human-centered livability evaluation index system in the region and adjust the spatial micro-renewal practice activities.

2. The method for measuring the livability of street space based on multi-source data fusion as described in claim 1, characterized in that: The street view data measurement specifically includes the following steps: S1.1: Street view data collection takes street view image data of the study area as the research object, and realizes intuitive and visual display of the constituent elements and proportions of street view image data; S1.2: Create a sample dataset by randomly cropping the input images to 500 pixels × 500 pixels, using Photoshop to subdivide each category, and generating a Photoshop sample set of 500 images using manual annotation; S1.3: Divide each element into different colors, create a color swatch using fixed RGB colors, and use the paint bucket tool to segment the artificial architectural elements; secondly, in order to make each color correspond to the corresponding value in a one-dimensional vector, the colors in the image need to be encoded, reducing the three-dimensional to two-dimensional, and converting the BGR color as a three-digit number in base 256 into a decimal representation. S1.4: Convert the colormap into a two-dimensional matrix and remove duplicate numbers to obtain the color table. Finally, put the processed label image into the program and run it to obtain the semantic segmentation data and the proportion of each element in the street view image, realizing an intuitive display of the constituent elements and proportions of street view image data.

3. The method for measuring the livability of street space based on multi-source data fusion as described in claim 1, characterized in that: The spatial syntactic data measurement specifically includes the following steps: S2.1: Data preprocessing: Collect street-related geospatial information and textual description information, and integrate them into structured data; S2.2: Spatial Relationship Modeling: Modeling the geospatial information of streets into an axis representation; S2.3: Spatial Relationship Analysis: Based on the spatial relationship model, analyze the spatial relationships between various geographic spaces in the street, and combine this with textual description information to determine information such as the division of areas, major attractions, and commercial areas in the street; S2.4: Spatial Quantification: Based on the results of spatial relationship analysis, conduct quantitative analysis on the spatial structure, landscape, and clusters of the streets; S2.5: Syntactic Analysis: Based on the spatial structure and context of the street, explain the historical and cultural connotations and commercial value of the street, and provide recommendations for shopping, entertainment and other services.

4. The method for measuring the livability of street space based on multi-source data fusion as described in claim 1, characterized in that: The measurement of wind environment data specifically includes the following steps: S3.1: Study the climate characteristics of the area where the existing space is located and the characteristics of wind flow around the building. Compare relevant regulations and wind environment evaluation standards proposed by domestic and foreign scholars to determine wind environment evaluation standards in four aspects: wind speed, wind speed ratio, wind pressure and wind chill temperature. S3.2: Understand the CFD numerical simulation process, determine the simulation requirements such as turbulence model, discretization and solution methods, and boundary condition settings, and verify the CFD numerical simulation by combining typical wind tunnel tests conducted by foreign scholars, so as to ensure the accuracy of the turbulence model and solution method selected in this invention. S3.3: Conduct a survey of the study area and summarize common building cluster layouts; simulate the wind environment around individual buildings under different conditions of cross-sectional shape, height, and wind direction angle, as the basis for subsequent research on wind environment simulation around different building layouts; S3.4: Based on the survey results, simulate the wind environment around the residential area in winter and summer for different layout forms, compare and analyze the simulation results, and find the most suitable layout form for winter and summer. S3.5: Simulate and analyze the existing spatial environment and conduct a wind environment assessment, while also considering the impact of public buildings near the settlement on the wind environment.

5. The method for measuring the livability of street space based on multi-source data fusion as described in claim 2, characterized in that: The street view data evaluation system includes, but is not limited to, green view rate and sky view rate.

6. The method for measuring the livability of street space based on multi-source data fusion as described in claim 3, characterized in that: The spatial syntax data evaluation system includes, but is not limited to, connectivity, global integration, local integration, selectivity, synergy / comprehensibility, horizon connectivity, and horizon integration.

7. The method for measuring the livability of street space based on multi-source data fusion as described in claim 4, characterized in that: The wind environment data evaluation includes, but is not limited to, the wind speed ratio in the comfortable wind zone, the wind speed ratio in the calm wind zone, the wind speed ratio in the strong wind zone, air age, and PMV.