A community public space renewal potential assessment method based on spatial entropy

Abstract

The application discloses a community public space updating potential evaluation method based on spatial entropy, which is based on the multi-element spatial entropy measurement to construct a public space updating potential evaluation system from the quantitative evaluation perspective of the public space updating potential, so as to evaluate the updating potential of the community public space and analyze the updating orientation. The application can realize the fine and human-scale public space updating potential evaluation, better serve the planning decision and management of the community public space, guide the orderly development of the community public space updating work, utilize the multi-source data of building data, plot data, administrative boundary data and street view data, construct an evaluation system from three dimensions of the planar space order, the facade image and the landscape facility matching, realize the more fine and accurate updating potential evaluation of the community public space, and improve the problems of the previous community public space evaluation method, such as strong subjectivity, large personnel investment and low work efficiency.

Description

Technical Field

[0001] This invention relates to the field of urban planning technology, specifically a method for assessing the potential for renewal of community public spaces based on spatial entropy. Background Technology

[0002] Currently, there are two main methods for assessing the potential for community public space renewal. One method involves professional planners conducting on-site surveys, interviews, and surveys to assess the current state of community public spaces. This method requires significant manpower and resources in the initial research phase and relies heavily on qualitative judgments based on the strong knowledge of the researchers, making it difficult to widely apply. The other method involves constructing a quantitative indicator system based on collected economic and social indicators to analyze the potential of the overall built environment's public space coverage, accessibility, and usability. This method lacks a systematic evaluation from a renewal-oriented perspective and neglects the visual characteristics of small-scale features and facilities. Therefore, current methods for assessing the potential for community public space renewal cannot efficiently grasp the multi-dimensional characteristics of community public spaces under the current community renewal orientation, nor can they accurately identify the renewal potential level and renewal orientation of community public spaces, thus failing to provide an effective basis for the implementation of community renewal projects. Summary of the Invention

[0003] To address the shortcomings mentioned in the background section, the present invention aims to provide a method for assessing the potential for renewal of community public spaces based on spatial entropy.

[0004] The objective of this invention can be achieved through the following technical solution: a method for assessing the renewal potential of community public spaces based on spatial entropy, the method comprising the following steps:

[0005] Acquire and store geospatial vector data, community information, and street view data of the administrative area to be evaluated;

[0006] The acquired geospatial vector data and community information are input into the geographic information platform, and the public space patches within the boundary of each community are marked and numbered. Based on spatial location, the geospatial vector data, community information and public space patches are associated to form a basic database of public space patches.

[0007] The planar walkable paths and street scene elements of public space patches are identified, and the planar path spatial entropy A of each public space patch is measured. i Facade style spatial entropy B i Landscape facility spatial entropy C i There are three types of spatial entropy values, which are uniformly stored as attribute data of the public spatial patches in the public spatial patch basic database;

[0008] Extract and classify the spatial entropy value of each type of public space patch, and determine the renewal potential level from three aspects: "planar path order - facade appearance - landscape facilities". Combine with community information to verify the coefficient of renewal potential level to obtain the final public space patch value result. And make renewal potential judgment according to the final public space patch value result.

[0009] The regeneration potential of identified public space patches is visualized in the geographic information platform by category and level. The visualization of regeneration potential includes the entropy distribution of each public space patch and the regeneration potential level attribute information. The visualization of regeneration potential is linked with various external databases to export a community public space regeneration potential evaluation report.

[0010] Preferably, the process of acquiring and storing geospatial vector data and community information within the administrative area to be evaluated includes the following steps:

[0011] Obtain the street administrative boundaries where community public spaces require updating;

[0012] The geospatial vector data of the street to be evaluated is obtained through the OpenstreetMap open data platform. The geospatial vector data includes building data and land parcel data. The building data includes information on the number of building floors, building height, year of construction, and building location. The land parcel data includes land use type and geographic coordinate information.

[0013] Obtain community information within the street area to be evaluated; the community information is area vector data, and includes community name, community geographical location information, and the number of permanent residents in the community;

[0014] Street view data of the community was collected using a panoramic 3D scanner; the street view data of the community includes point cloud data and street view measurement images; the street view measurement images include geographical location, street view measurement image pixels, street view measurement image RGB information, and street view information.

[0015] Preferably, the building data includes information on the number of building floors, building height, year of construction, and building location; the land parcel data includes land use type and geographic coordinates.

[0016] Preferably, the process of forming a basic database of public space patches includes the following steps:

[0017] The collected geospatial vector data is imported into the geographic information system and spatially calibrated to form a basic geospatial database.

[0018] Using spatial cutting tools, the plot is spatially divided by the plot unit boundary line and the building base outline line, and all spaces containing the building base are marked as building patches, while the remaining spaces outside the buildings are marked as public space patches.

[0019] Using spatial connectivity tools, the collected community information and street view data of the community are connected to various public space patches to form a basic database of public space patches to be evaluated;

[0020] Preferably, in the basic database of public space patches, each public space patch is sequentially numbered 1, 2, ..., k, and includes information such as the number of building floors, building height, year of construction, building location, land use type, geographic coordinates, community name, community geographic location, number of permanent residents in the community, geographic location information of street view measurement images, pixel information of street view measurement images, and RGB information of street view measurement images.

[0021] Preferably, the planar path spatial entropy A i Facade style spatial entropy B i Landscape facility spatial entropy C i The measurement process includes the following steps:

[0022] Planar path space entropy A i The planar complexity is calculated based on public space patch data, using the average walking time for all walkable paths, as shown in the formula:

[0023]

[0024] Walking time T for any walkable path n The data collection method is as follows: A timed agent with a walking speed of 1 m / s is placed in the site. Any open point on the boundary of a unit is set as the starting and ending point. The center line of the public space is used as the selectable path to generate a walkable path n. The total movement time for this walkable path is then obtained as T. n ;

[0025] The method for generating the center line of the public space is as follows: starting from the westernmost boundary of the plot, draw a north-south vertical normal line every 1 meter from west to east. Using the plot boundary and building outline as the cutting basis, identify the center point of the line segment after cutting, excluding the building space, and connect the center points in sequence to generate the center line of the public space.

[0026] Facade style spatial entropy B i This is achieved by identifying street scene elements from street view data of public space patches, calculating the color and style complexity of building facades to obtain building color entropy and building style entropy respectively, and then weighting the entropy values ​​of building color entropy and building style entropy to obtain the facade style spatial entropy. This is accomplished based on the following steps:

[0027] The street scene element identification refers to the identification and extraction of building facades and landscape elements from street scene data in the public space patch database by using the InfoGAN generative adversarial network method.

[0028] The method for identifying and extracting building facades and landscape elements is as follows: the AWB white balance algorithm is used to perform batch white balance processing on street view images, and the InfoGAN information generative adversarial network is used to perform semantic segmentation and element recognition on street view images, identify building areas and non-building areas, extract building areas as building facades, and extract non-building areas as landscape elements.

[0029] The architectural color entropy refers to the calculation of the color richness of a building facade. First, the RGB information of the building facade colors is converted into HSV information. Then, based on the K-Nearest Neighbors (KNN) algorithm, the building facade colors are grouped into categories using the HSV information to obtain the number of facade color categories in each public space patch. Finally, the color category richness of the facade in each public space patch is calculated. The formula for calculating architectural color entropy is:

[0030]

[0031] Among them, H i Let B be the architectural color entropy of the i-th public space patch. ij It is the ratio of the number of facade color categories in the i-th public space patch to the total number of facade color categories in all public space patches;

[0032] The method for classifying facade color information is as follows: the extracted RGB image of the building facade is transferred to the HSV space mode to obtain HSV information. Then, the color information is grouped into color categories based on the HSV information, and the KNN algorithm is used to cluster the color information of the building facade according to the corresponding groups. The color category grouping is based on the following criteria: hue (H) is grouped into 36 groups of 10°; saturation (S) is grouped into 5 groups of 20°; and lightness (V) is grouped into 5 groups of 20°, for a total of 900 color categories.

[0033] The architectural style entropy refers to the process of transforming street view data from the basic database of public space patches into structured street view feature data using the Region-CNN algorithm, and then using the Support Vector Machine (SVM) algorithm for style classification to identify the architectural style categories and numbers in each public space patch, thereby calculating the richness of architectural style categories in each public space patch. The formula for calculating architectural style entropy is as follows:

[0034]

[0035] Among them, V iLet P be the architectural style entropy in the i-th public space patch. ie V represents the percentage of architectural style category e in the i-th public space patch, where n is the total number of architectural style categories in all public space patches. i The larger the building, the richer its architectural style;

[0036] The structured street view feature data is clustered using the Selective Search algorithm. Street view image regions with similar colors and textures are then clustered as candidate regions for landscape recognition. The landscape classification rule is as follows: the R-CNN algorithm is used to extract features from the candidate regions for landscape recognition of each street view image, and the SVM algorithm is used to automatically cluster the feature vectors. Different clustering results are summarized into corresponding feature landscape categories to identify the feature landscape categories and number contained in each street view image.

[0037] The facade style spatial entropy is a weighted calculation of the color and style complexity of building facades based on street view data of public space patches. The calculation formula is as follows:

[0038] B i =H i *a q +V i *a w

[0039] Among them, B i Let a be the spatial entropy of the architectural style of the i-th public space patch. q a w For the functional factor weights, and a q a w All values ​​are 0.5;

[0040] Landscape facility spatial entropy C i Landscape elements, including streetlights, trash cans, trees, railings, and signs, are identified and their numbers are recorded based on the Cityscapes training set. The information entropy of the landscape elements and their numbers is calculated using the following formula:

[0041]

[0042] Among them, O ij It is the ratio of the number of landscape facilities in the i-th public space patch to the total number of landscape facilities in all public space patches;

[0043] The method for identifying landscape facilities and their numbers involves clustering based on the SelectiveSearch algorithm using the Cityscapes training set to generate candidate regions containing landscape facilities in street view images of each public space patch. A CNN algorithm is then used to extract features from each candidate region, and an SVM algorithm is used to classify the feature vectors. This identifies five types of landscape facilities and their numbers in each street view image: streetlights, trash cans, trees, railings, and signs. The Cityscapes training set is a semantic understanding image dataset of urban street scenes, containing 5000 high-quality pixel-level annotated images of driving scenes in an urban environment.

[0044] Preferably, the process of extracting and classifying the spatial entropy values ​​of each public space patch into three types, and determining the renewal potential level from three aspects: "planar path order - facade appearance - landscape facilities", and verifying the coefficient of the renewal potential level in conjunction with the objective attribute information of the community, and judging the renewal potential according to the final assignment result, includes the following steps:

[0045] The spatial entropy A of the planar path i Facade style spatial entropy B i Landscape facility spatial entropy C i The values ​​are normalized separately, as shown in the following formula:

[0046]

[0047] In the formula, D' xi D represents the normalized entropy value of the i-th common space patch under the x-th type, i.e., the standard value of the spatial entropy of the common space patch. xi Let D be the spatial entropy value of the common space patch under type x. xmin D represents the maximum entropy value of the common space patch under the xth type. xmin Let x be the minimum value of the spatial entropy of the unit under the x-th type, where x takes the values ​​A, B, or C.

[0048] After normalizing different types of spatial entropy values, standard values ​​of spatial entropy are obtained. The standard value of spatial entropy D' of the planar path is evaluated from three aspects: "planar path order - facade appearance - landscape facilities". Ai Facade style spatial entropy D' Bi Standard value of spatial entropy of landscape facilities D' Ci The process involves assigning values ​​according to the rules and determining the update potential, as follows:

[0049] Based on the planar path order, divide the planar path spatial entropy A i grade:

[0050] The standard value of the spatial entropy of the plane path D' AiPotential assessment is performed according to a Gaussian discrete distribution model, where the Gaussian discrete distribution model refers to the normal distribution model obtained from the Gaussian model. If the planar path spatial entropy A of the i-th common spatial patch... i Located in the last 25%, i.e., D' Ai If the value is ≤0.25, it is determined to be a high-potential update region of planar path order, and the label value is assigned as 1; otherwise, it is determined to be a low-potential update region, and the label value is assigned as 0.

[0051] Based on the facade style and image, the facade style spatial entropy B is divided into... i grade:

[0052] The standard value of spatial entropy of facade style D' Bi According to the natural breakpoint method, the standard value D' of the facade style spatial entropy of the i-th public space patch is... Bi >0.75 or D' Bi If the value is ≤0.25, it is determined to be a low-potential area for facade appearance renewal, and its assigned value is marked as 0; if the standard value of the facade appearance spatial entropy of the i-th public space patch is D' Bi It falls in the middle of the entropy value range, i.e., 0.25 < D' Bi If the value is ≤0.75, it is determined to be a high-potential area for facade renovation, and its assigned value is marked as 1;

[0053] Based on the classification of landscape facility spatial entropy C, landscape facility spatial entropy is determined. i grade:

[0054] The standard value of the spatial entropy D' of landscape facilities in public space patches is determined based on the Gaussian discrete distribution model. Ci If the standard value of the spatial entropy of the i-th public space patch landscape facilities is D' Ci Located in the last 25%, i.e., D' Ci If the value is ≤0.25, it is determined to be a high-potential redevelopment area with supporting landscape facilities, and its value is marked as 1; the rest are marked as 0.

[0055] Based on the objective characteristics of the community, and considering the building age and the coverage rate of public spaces, the total renewal potential of each community's public space patches was verified and optimized according to the following rules, whereby...

[0056] For low-potential redevelopment areas of facade appearance, the potential of building age characteristics is optimized. If the proportion of buildings in the block unit that are 50 years or more older than the time of determination reaches 50%, it is determined to be a high-potential redevelopment area of ​​facade appearance, and its assigned value is marked as 1.

[0057] For low-potential regeneration areas of planar path order, the potential of community public space characteristics is optimized. If the community public space coverage rate is lower than the standard, it is determined to be a high-potential regeneration area of ​​planar path order, and its assignment result is marked as 1. The community public space coverage rate refers to the ratio of public space green area to the number of permanent residents in the community. The standard refers to the green space rate of residential blocks not less than 30% or the per capita public green space area not less than 70% according to the building climate zoning.

[0058] The three types of spatial entropy values ​​are superimposed to obtain the final public space patch assignment result. The rules for classifying the update potential level are as follows: public space patches with a final assignment value of 3 are determined as high-potential update spaces, public space patches with a final assignment value of 2 are determined as medium-potential update spaces, public space patches with a final assignment value of 1 are determined as low-potential update spaces, and public space patches with a final assignment value of 0 are deleted.

[0059] The orientation type of public space renewal is determined based on the spatial entropy standard values ​​of three types: "planar path order - facade appearance - landscape facilities".

[0060] Preferably, the process of linking the visualized update potential content with various external databases to export the community public space update potential evaluation report includes the following steps:

[0061] In the geographic information platform, the results of the regeneration potential identification of the public spatial patches are visualized by type and level through the output device. The regeneration potential identification results include the spatial entropy value distribution of the three types of each public spatial patch, the regeneration potential level attribute, and the regeneration potential type information.

[0062] The obtained results of the renewal potential are linked to the databases of various planning and management departments, and a community public space renewal potential evaluation report is printed out to guide the site selection for public space renewal in each street and community.

[0063] An apparatus comprising:

[0064] One or more processors;

[0065] Memory, used to store one or more programs;

[0066] When one or more of the programs are executed by one or more of the processors, the one or more of the processors implement a method for assessing the potential for renewal of community public spaces based on spatial entropy, as described above.

[0067] A storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform a method for assessing the potential for renewal of community public spaces based on spatial entropy, as described above.

[0068] The beneficial effects of this invention are:

[0069] 1. Effectiveness of the work. This invention achieves a technological breakthrough, shifting from relying on professionals to being easily and directly mastered by non-professionals, breaking down professional technical barriers, and providing a simple and direct technical management method for the future renewal path of the community.

[0070] 2. Objectivity in Decision-Making. This invention utilizes multi-source data, including building data, land parcel data, administrative boundary data, and street view data, to construct an evaluation system from three dimensions: planar spatial order, facade appearance, and landscape facilities. This enables a more refined and accurate assessment of the renewal potential of community public spaces, improving upon previous community public space evaluation methods that relied on qualitative judgments, were highly subjective, required significant personnel investment, and were inefficient.

[0071] 3. Assessment Accuracy. Based on a three-tiered assessment of renewal potential, the method for assessing the renewal potential of community public spaces in this invention employs building age and landscape planning and design standards for secondary verification, further improving the accuracy of the assessment. Attached Figure Description

[0072] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0073] Figure 1 This is a flowchart of a method according to an embodiment of the present invention;

[0074] Figure 2 The planar path space entropy A in an embodiment of the present invention i Measurement method diagram;

[0075] Figure 3 This is a schematic diagram of the normalized coupling output process for three types of spatial entropy values ​​in an embodiment of the present invention;

[0076] Figure 4 This is a schematic diagram illustrating the output of the potential assessment for public space patch renewal units in Yuejianglou Street, according to an embodiment of the present invention.

[0077] Figure 5 This is a visualization output image of the potential for public space patch renewal in Yuejianglou Street, as described in an embodiment of the present invention. Detailed Implementation

[0078] 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.

[0079] like Figure 1 As shown, a method for assessing the renewal potential of community public spaces based on spatial entropy includes the following steps:

[0080] Geospatial vector data and community information within the administrative area to be evaluated are acquired and stored through an open data platform and panoramic 3D scanners.

[0081] All acquired data is uniformly input into the geographic information platform, marking the public space patches within the boundary of each community, and associating the collected geospatial vector data and community information with each public space patch based on spatial location to form a basic database of public space patches;

[0082] The public space elements are identified from the obtained public space patches to be evaluated, and the planar complexity is calculated based on the walking time of walkable paths. The planar path spatial entropy measure A is then performed. i The color and style complexity of building facades are calculated using street view data of public space patches, and the spatial entropy B of the facade style is measured by weighted calculation. i The number of landscape elements identified from public space patch street view data is extracted, and the spatial entropy C of landscape facilities is calculated. i Measurement calculation.

[0083] The spatial entropy values ​​of each public space patch were extracted and divided into three types. The degree of renewal potential was determined from three aspects: "planar spatial order, facade appearance, and landscape facilities". The coefficient of renewal potential level was checked in combination with the objective attribute information of the community. The renewal potential was judged according to the final value result.

[0084] The identified public space patches are visualized by category and level. The visualization of the renewal potential includes the entropy distribution of each public space patch and the renewal potential level attribute information. This data is then linked to external databases of planning and renewal-related departments and community administrative departments to export a community renewal potential evaluation map.

[0085] Example

[0086] The following will use the assessment and demonstration of the potential for community public space renewal based on spatial entropy in Yuejianglou Street, Nanjing City as an example to illustrate the technical solution of the present invention in detail.

[0087] Taking Yuejianglou Street in Nanjing City as the research object, spatial data, community data, and street view data were collected using drone oblique photography equipment, open-source data interfaces, and panoramic 3D scanners. Specifically, this included:

[0088] According to the project brief, the administrative boundaries of Yuejianglou Subdistrict where the community public space needs to be updated were determined.

[0089] Spatial vector data, including building data and land parcel data, was obtained within the Yuejianglou Subdistrict using the OpenStreetMap open data platform. The data is in shapefile format. The building data includes information on the number of building floors, building height, year of construction, and building location; the land parcel data includes land use type and geographic coordinates.

[0090] The planar vector shapefile data of the communities within the street to be evaluated was obtained from the official website of the Nanjing Municipal Government through an open-source data interface. This data includes the community name, the community's geographical location information, and the number of permanent residents in the community.

[0091] Street view data of the community was collected using a panoramic 3D scanner equipped with four 20-megapixel fisheye lenses. The data includes point cloud data collected using the accompanying trajectory software and high-resolution street view images with GPS positioning. The point cloud data is stored as binary files, with corresponding metadata stored as fields and records. The street view images are saved in uncompressed BMP format, with a resolution of 4*20 megapixels, and include geographic location, street view image pixels, street view image RGB information, and street view information.

[0092] All acquired data will be uniformly input into the geographic information platform to mark all public space patches within the boundary of Yuejianglou Street. Based on spatial location, the collected geospatial vector data and community information will be associated with each public space patch to form a basic database of public space patches, specifically including:

[0093] The collected spatial data is imported into a geographic information system and spatially calibrated using the CGS-WGS-1984 coordinate system to form a basic geospatial database.

[0094] Using the spatial segmentation tool in ArcGIS 10.8, the plot was spatially divided using the boundary lines of the Yuejianglou Street plot units and the outlines of the building bases. All spaces containing building bases were marked as building patches, while the remaining spaces outside the buildings were marked as public space patches. A total of 20 public space patches were identified.

[0095] Using the spatial connectivity tool in ArcGIS 10.8, the collected information data and street view data of Yuejianglou Community were connected to the identified public space patches in Yuejianglou Street, forming a basic database of public space patches to be evaluated. The public space patches in Yuejianglou Street were numbered sequentially as 1, 2, ..., 20. The basic database of public space patches to be evaluated includes information such as building floor number, building height, construction year, building location, land use type, geographic coordinates, community name, community geographic location, community resident population, street view image geographic location, street view image pixel count, and street view image RGB information. Figure 2 As shown.

[0096] Multi-factor spatial entropy measurement was performed on 20 public space patches identified in Yuejianglou Street. Planar complexity was calculated based on walking time of walkable paths, and planar path spatial entropy was measured. Color and style complexity of building facades was calculated from the street view data of the acquired public space patches, and weighted calculation of facade style spatial entropy was performed. The number of landscape elements identified from the street view data of the public space patches was extracted, and spatial entropy measurement of landscape facilities was calculated. Specifically, this included:

[0097] Yuejianglou Street Planar Path Spatial Entropy A i Measure, such as Figure 3 As shown. Starting from the westernmost boundary of Yuejianglou Street, perpendicular normals are drawn every 1 meter from west to east in a north-south direction. Using the plot boundary and building outline as the cutting basis, the center points of the line segments after cutting (excluding the building spaces) are identified, and the center points are connected sequentially to generate the center lines of the public spaces. The traversal time of all walkable public space center lines is calculated to determine the planar complexity. The calculation formula is:

[0098]

[0099] The method for collecting the walking time of any walkable path is as follows: A timed agent with a walking speed of 1 m / s is placed in the site, and any open point at the boundary of a unit is set as the starting and ending point. Using the center line of the public space as the selectable path, a walkable path n is generated, and the total movement time of this walkable path is obtained as T. n .

[0100] The calculation results are as follows:

[0101]

[0102] Yuejianglou Street Facade Style Space Entropy B iMeasurement. First, the AWB white balance algorithm is used to perform batch white balance processing on street view images. Then, based on the convolutional neural network, the street view images are used to identify elements, distinguishing between building areas and non-building areas. Building areas are extracted as building facades, and non-building areas are extracted as landscape elements. The street view data in the public space patch database is used to identify and extract building facades and landscape elements.

[0103] The extracted RGB images of the building facades were transferred to the HSV color space, and then grouped into color categories using HSV. The KNN (K-nearest neighbors) algorithm was then used to cluster the color information of the building facades according to the corresponding groups. The color category grouping was based on three criteria: H (hue) grouped in 10° increments (36 groups); S (saturation) grouped in 20° increments (5 groups); and V (value) grouped in 20° increments (5 groups), resulting in a total of 900 color categories. Based on this, the building color entropy H was calculated. i The measure of is calculated using the following formula:

[0104]

[0105] Among them, B ij It is the ratio of the number of building color categories in the i-th public space patch to the number of building color categories in all public space patches.

[0106] After extracting building facades and clustering them using the Selective Search algorithm on street view data, areas with similar colors and textures are clustered as candidate regions for architectural style identification. R-CNN algorithm is used to extract features from the candidate regions of each street view image, and SVM algorithm is used to automatically cluster the feature vectors. Different clustering results are summarized into corresponding architectural style categories. Based on this, the architectural style categories and their number in each street view image are identified, and the richness of architectural style categories in each public space patch is calculated. The calculation formula is as follows:

[0107]

[0108] Among them, V i Let P be the architectural style entropy in the i-th public space patch. ie V represents the percentage of architectural style category e in the i-th public space patch, where n is the total number of architectural style categories in all public space patches. i The larger the value, the richer the architectural style.

[0109] Finally, the spatial entropy B of the facade style of Yuejianglou Community was analyzed. i The calculation formula is as follows:

[0110] B i=H i *a q +V i *a w

[0111] Among them, B i Let H be the spatial entropy of the facade appearance of the i-th public space patch. i Let V be the architectural color entropy of the i-th public space patch. i Let a be the architectural style entropy of the i-th public space patch. q a w For the functional factor weights, a q a w All values ​​are 0.5.

[0112] The calculation results are as follows:

[0113]

[0114] Landscape facility spatial entropy C i Measurement. Spatial entropy of landscape facilities C i This refers to identifying five categories of landscape features (streetlights, trash cans, trees, railings, and signs) and their quantities in street view images based on the Cityscapes training set, and calculating the information entropy of the landscape features and their quantities. The calculation formula is as follows:

[0115]

[0116] Among them, O ij It is the ratio of the number of landscape facilities in the i-th public space patch to the total number of landscape facilities in all public space patches.

[0117] The identification of landscape facilities and their quantities refers to clustering and identification based on the Selective Search algorithm using the Cityscapes training set. This generates candidate regions containing landscape facilities in street view images of various public space patches. A CNN algorithm is used to extract features from these candidate regions, and an SVM algorithm is used to classify the feature vectors. This identifies the five categories of landscape facilities (streetlights, trash cans, trees, railings, and signs) and their quantities in each street view image. The Cityscapes training set refers to a semantic understanding image dataset of urban street scenes, containing 5000 high-quality pixel-level annotated images of driving scenes in urban environments. The Selective Search algorithm is a feature-based object detection algorithm used in R-CNN to generate candidate regions.

[0118] The calculation results are as follows:

[0119]

[0120] An urban renewal potential assessment and verification was conducted on Yuejianglou Subdistrict. This assessment extracted and categorized spatial entropy values ​​of three types for each public space patch generated in the preceding steps, determining the degree of renewal potential from three aspects: "planar spatial order," "three-dimensional urban image," and "landscape facility support." Furthermore, the coefficient verification of the renewal potential level was performed in conjunction with the community's objective attribute information, primarily including:

[0121] The spatial entropy A of the Yuejianglou Street planar path is calculated using the following formula. i Facade style spatial entropy B i Landscape facility spatial entropy C i Normalize them separately:

[0122]

[0123] D' xi D represents the normalized entropy value of the i-th common space patch under the x-th type, i.e., the standard value of the spatial entropy of the common space patch. xi D represents the spatial entropy value of the common space patch under this type. xmin D represents the maximum entropy value of the common space patch under the xth type. xmin Let x be the minimum value of the unit space entropy under the x-th type, where x can be A, B, or C.

[0124] Three types of spatial entropy levels were assigned values, and the spatial entropy values ​​of different dimensions were normalized to obtain the standard value of spatial entropy. The standard value D' of the spatial entropy of the planar path was evaluated from three aspects: "planar spatial order - three-dimensional urban image - landscape facility support". Ai Facade style spatial entropy D' Bi Standard value of spatial entropy of landscape facilities D' Ci Assign values ​​to it according to the following rules to determine its update potential.

[0125] Based on the planar spatial order, the entropy A of the planar path space is divided. i grade

[0126] The standard value of the spatial entropy of the plane path D' Ai Potential assessment is performed using a discrete distribution model, where the planar path spatial entropy A of the i-th common spatial patch is... i Located in the last 25%, i.e., D' Ai If the value is ≤0.25, it is determined as a high-potential renewal region under the guidance of planar spatial order improvement, and a value of 1 is assigned to it; otherwise, it is determined as a low-potential renewal region, and a value of 0 is assigned to it.

[0127] The results are shown in the table below:

[0128]

[0129] Based on the three-dimensional urban image, the spatial entropy of facade style is divided into B. i grade

[0130] The standard value of spatial entropy of facade style D' Bi According to the natural breakpoint method, the standard value D' of the facade style spatial entropy of the i-th public space patch is... Bi >0.75 or D' Bi If the value is ≤0.25, then it is identified as a plot of land in urgent need of updating the city's three-dimensional image, and its value is assigned to 0; if the standard value of the three-dimensional spatial entropy of the i-th public space patch is D' Bi It falls in the middle of the entropy value range, i.e., 0.25 < D' Bi If the value is ≤0.75, it is identified as a high-potential urban renewal area under the guidance of three-dimensional urban image enhancement, and its value is assigned as 1.

[0131] The results are shown in the table below:

[0132]

[0133] Based on the classification of landscape facility spatial entropy C, landscape facility spatial entropy is determined. i grade

[0134] The standard value of the spatial entropy of landscape facilities in public space patches, D', is determined based on the discrete distribution model. Ci If the standard value of the spatial entropy of the i-th public space patch landscape facilities is D' Ci Located in the last 25%, i.e., D' Ci If the value is ≤0.25, it is considered a high-potential redevelopment area under the guidance of landscape facility improvement, and is assigned a value of 1; otherwise, it is assigned a value of 0.

[0135] The results are shown in the table below:

[0136]

[0137] The public space patch assignment results obtained by matching "planar path order, facade style and image, and landscape facility support" are coupled, and the normalized coupling output flowchart of the three types of spatial entropy values ​​is shown in the figure. Figure 3As shown in the diagram. The renewal potential of Yuejianglou Street is assessed based on the final assignment results. Public space patches with a final assignment value of 3 are identified as high-potential renewal spaces; those with a final assignment value of 2 are identified as medium-potential renewal spaces (Level 2); those with a final assignment value of 1 are identified as low-potential renewal spaces; and those with a final assignment value of 0 are identified as Level 0 renewal potential spaces. The public space renewal guidance type is determined based on the spatial entropy standard values ​​of three types: "planar path order," "facade appearance," and "landscape facilities." A schematic diagram of the judgment data for each public space patch renewal unit is shown in the diagram. Figure 4 As shown in the table below, the judgment results are as follows:

[0138]

[0139] High-potential space No. 10 is currently a parking lot. The space is open and flat with low plan complexity. The surrounding area mainly consists of residential buildings with a monotonous architectural style. The internal function is parking, and the landscaping and facilities are weak, making it a high-potential space for public space renewal. Secondary medium-potential renewal spaces No. 8 and No. 16 are public space patches within the community. Their functions are unclear, currently mainly parking, with moderate plan complexity. The surrounding architectural style is simple, and they have some landscaping and facilities. Secondary medium-potential renewal spaces No. 3, No. 17, and No. 20 are community-external public space patches, but their plans are monotonous, lacking recreational routes and landscaping. Low-potential renewal spaces and Level 0 renewal potential spaces all have relatively complete landscaping and facilities. Among them, No. 2 and No. 4, as urban parks, have high architectural complexity and no need for renewal.

[0140] Based on the objective characteristics of Yuejiang Tower, and considering the building's age and the coverage rate of community public spaces, the total renewal potential of each community public space patch was verified and optimized according to the following rules, among which,

[0141] For low-potential redevelopment areas of facade appearance, the potential of building age characteristics is optimized. If the proportion of buildings in the block unit that are 50 years or more older than the time of determination reaches 50%, it is determined to be a high-potential redevelopment area of ​​facade appearance, and its assigned value is marked as 1.

[0142] The potential of community public space characteristics is optimized for low-potential regeneration areas of planar path order. If the community public space coverage rate is lower than the standard, it is determined to be a high-potential regeneration area of ​​planar path order and its assignment result is marked as 1. The community public space coverage rate refers to the ratio of public space green area to the number of permanent residents in the community. The standard refers to the green space rate of the residential block in the building climate zone where Yuejiang Tower is located not less than 30% or the per capita public green space area not less than 70%.

[0143] The identified Yuejianglou public space patches' renewal potential is visualized by categorizing and classifying them. This visualization displays the entropy distribution and renewal potential level attributes of each public space patch. This data is then linked to the databases of various planning and management departments to export a community renewal potential evaluation map. The resulting visualization output is shown in the image below. Figure 5 As shown. Mainly includes:

[0144] The obtained public space patch renewal potential identification results are visualized by type and level through output device. The renewal potential identification results include the entropy distribution and renewal potential level attribute information of each public space patch. The results are exported to a large projection LED screen for interactive display.

[0145] The obtained results on the potential for renewal are linked to the databases of various planning and management departments, and a report on the potential for renewal of community public spaces is printed out to guide the site selection and direction of public space renewal in Yuejianglou Community.

[0146] Based on the same inventive concept, this invention also provides a computer device, comprising: one or more processors, and a memory for storing one or more computer programs; the programs include program instructions, and the processor executes the program instructions stored in the memory. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, used to implement one or more instructions, specifically for loading and executing one or more instructions stored in a computer storage medium to implement the above-described method.

[0147] It should be further explained that, based on the same inventive concept, the present invention also provides a computer storage medium storing a computer program, which, when executed by a processor, performs the above-described method. This storage medium can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0148] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0149] The foregoing has shown and described the basic principles, main features, and advantages of this disclosure. Those skilled in the art should understand that this disclosure is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of this disclosure. Various changes and modifications can be made to this disclosure without departing from its spirit and scope, and all such changes and modifications fall within the scope of this disclosure as claimed.

Claims

1. A method for assessing the renewal potential of community public spaces based on spatial entropy, characterized in that, The method includes the following steps: Acquire and store geospatial vector data, community information, and street view data of the administrative area to be evaluated; The acquired geospatial vector data and community information are input into the geographic information platform, and the public space patches within the boundary of each community are marked and numbered. Based on spatial location, the geospatial vector data, community information and public space patches are associated to form a basic database of public space patches. The planar walkable paths and street scene elements of public space patches are identified, and the planar path spatial entropy of each public space patch is then measured. Facade style and spatial entropy Landscape facility spatial entropy There are three types of spatial entropy values, which are uniformly stored as attribute data of the public spatial patches in the public spatial patch basic database; The planar path space entropy Facade style and spatial entropy Landscape facility spatial entropy The measurement process includes the following steps: Planar path space entropy The planar complexity is calculated based on public space patch data, using the average walking time for all walkable paths, as shown in the formula: = Walking time for any walkable path The data collection method is as follows: A timed agent with a walking speed of 1 m / s is placed in the site. Any open point on the boundary of a unit is set as the starting and ending point. The center line of the public space is used as the selectable path to generate a walkable path n. The total movement time of this walkable path is then obtained. ; The method for generating the center line of the public space is as follows: starting from the westernmost boundary of the plot, draw a north-south vertical normal line every 1 meter from west to east. Using the plot boundary and building outline as the cutting basis, identify the center point of the line segment after cutting, excluding the building space, and connect the center points in sequence to generate the center line of the public space. Facade style spatial entropy This is achieved by identifying street scene elements from street view data of public space patches, calculating the color and style complexity of building facades to obtain building color entropy and building style entropy respectively, and then weighting the entropy values ​​of building color entropy and building style entropy to obtain the facade style spatial entropy. This is accomplished based on the following steps: The street scene element identification refers to the identification and extraction of building facades and landscape elements from street scene data in the public space patch database by using the InfoGAN generative adversarial network method. The method for identifying and extracting building facades and landscape elements is as follows: the AWB white balance algorithm is used to perform batch white balance processing on street view images, and the InfoGAN information generative adversarial network is used to perform semantic segmentation and element recognition on street view images, identify building areas and non-building areas, extract building areas as building facades, and extract non-building areas as landscape elements. The architectural color entropy refers to the calculation of the color richness of a building facade. First, the RGB information of the building facade colors is converted into HSV information. Then, based on the K-Nearest Neighbors (KNN) algorithm, the building facade colors are grouped into categories using the HSV information to obtain the number of facade color categories in each public space patch. Finally, the color category richness of the facade in each public space patch is calculated. The formula for calculating architectural color entropy is: in, Let i be the architectural color entropy of the i-th public space patch. It is the ratio of the number of facade color categories in the i-th public space patch to the total number of facade color categories in all public space patches; The method for classifying facade color information is as follows: the extracted RGB image of the building facade is transferred to the HSV space mode to obtain HSV information. Then, the color information is grouped into color categories based on the HSV information, and the KNN algorithm is used to cluster the color information of the building facade according to the corresponding groups. The color category grouping is based on the following criteria: hue (H) is grouped into 36 groups of 10°; saturation (S) is grouped into 5 groups of 20°; and lightness (V) is grouped into 5 groups of 20°, for a total of 900 color categories. The architectural style entropy refers to the process of transforming street view data from the basic database of public space patches into structured street view feature data using the Region-CNN algorithm, and then using the Support Vector Machine (SVM) algorithm for style classification to identify the architectural style categories and numbers in each public space patch, thereby calculating the richness of architectural style categories in each public space patch. The formula for calculating architectural style entropy is as follows: in, Let i be the architectural style entropy in the i-th public space patch. Let be the percentage of building style category e in the i-th public space patch, and n be the total number of building style categories in all public space patches. The larger the building, the richer its architectural style; The structured street view feature data is clustered using the Selective Search algorithm. Street view image regions with similar colors and textures are then clustered as candidate regions for landscape recognition. The landscape classification rule is as follows: the R-CNN algorithm is used to extract features from the candidate regions for landscape recognition of each street view image, and the SVM algorithm is used to automatically cluster the feature vectors. Different clustering results are summarized into corresponding feature landscape categories to identify the feature landscape categories and number contained in each street view image. The facade style spatial entropy is a weighted calculation of the color and style complexity of building facades based on street view data of public space patches. The calculation formula is as follows: in, Let i be the spatial entropy of the architectural style of the i-th public space patch. , For functional factor weights, and , All values ​​are 0.5; Landscape facility spatial entropy Landscape elements, including streetlights, trash cans, trees, railings, and signs, are identified and their numbers are recorded based on the Cityscapes training set. The information entropy of the landscape elements and their numbers is calculated using the following formula: in, It is the ratio of the number of landscape facilities in the i-th public space patch to the total number of landscape facilities in all public space patches; The method for identifying landscape facilities and their numbers is based on clustering identification using the Selective Search algorithm on the Cityscapes training set. This generates candidate regions containing landscape facilities in each public space patch street view image. The CNN algorithm is used to extract features from each candidate region of the street view image, and the SVM algorithm is used to classify the feature vectors. This identifies the five types of landscape facilities and their numbers in each street view image: streetlights, trash cans, trees, railings, and signs. The Cityscapes training set is a semantic understanding image dataset about urban street scenes, containing 5,000 high-quality pixel-level annotated images of driving scenes in urban environments. Extract and classify the spatial entropy value of each type of public space patch, and determine the renewal potential level from three aspects: "planar path order - facade appearance - landscape facilities". Combine community information to verify the coefficient of renewal potential level to obtain the final public space patch value result. And make renewal potential judgment according to the final public space patch value result. The regeneration potential of identified public space patches is visualized in the geographic information platform by category and level. The visualization of regeneration potential includes the entropy distribution of each public space patch and the regeneration potential level attribute information. The visualization of regeneration potential is linked with various external databases to export a community public space regeneration potential evaluation report.

2. The method for assessing the renewal potential of community public spaces based on spatial entropy according to claim 1, characterized in that, The process of acquiring and storing geospatial vector data and community information within the administrative area to be evaluated includes the following steps: Obtain the street administrative boundaries where community public spaces require updating; The geospatial vector data of the street to be evaluated is obtained through the OpenstreetMap open data platform. The geospatial vector data includes building data and land parcel data. The building data includes information on the number of building floors, building height, year of construction, and building location. The land parcel data includes land use type and geographic coordinate information. Obtain community information within the street area to be evaluated; the community information is area vector data, and includes community name, community geographical location information, and the number of permanent residents in the community; Street view data of the community was collected using a panoramic 3D scanner; the street view data of the community includes point cloud data and street view measurement images; the street view measurement images include geographical location, street view measurement image pixels, street view measurement image RGB information, and street view information.

3. The method for assessing the renewal potential of community public spaces based on spatial entropy according to claim 1, characterized in that, The process of forming the basic database of public space patches includes the following steps: The collected geospatial vector data is imported into the geographic information system and spatially calibrated to form a basic geospatial database. Using spatial cutting tools, the plot is spatially divided by the plot unit boundary line and the building base outline line, and all spaces containing the building base are marked as building patches, while the remaining spaces outside the buildings are marked as public space patches. Using spatial connectivity tools, the collected community information and street view data are connected to various public space patches to form a basic database of public space patches to be evaluated.

4. The method for assessing the renewal potential of community public spaces based on spatial entropy according to claim 3, characterized in that, In the basic database of public space patches, each public space patch is sequentially numbered 1, 2, ..., k, and contains information such as the number of building floors, building height, year of construction, building location, land use type, geographic coordinates, community name, community geographic location, number of permanent residents in the community, geographic location information of street view measurement images, pixel information of street view measurement images, and RGB information of street view measurement images.

5. The method for assessing the renewal potential of community public spaces based on spatial entropy according to claim 1, characterized in that, The spatial entropy values ​​of each public space patch are extracted and divided into three types, and the potential for renewal is determined from three aspects: "planar path order - facade appearance - landscape facilities". The process of updating the potential level coefficients by combining objective attribute information from the community, and then updating the potential judgment based on the final assignment results, includes the following steps: The entropy of the planar path space Facade style and spatial entropy Landscape facility spatial entropy The values ​​are normalized separately, as shown in the following formula: In the formula, For the first The first type The normalized entropy value of a common space patch, i.e., the standard value of the spatial entropy of the common space patch. For the first The spatial entropy value of public space patches under this type, For the first The maximum value of the entropy of the common space patch under this type, For the first The minimum value of the unit space entropy under this type, Choose A, B, and C; After normalizing different types of spatial entropy values, standard values ​​of spatial entropy are obtained. The standard values ​​of spatial entropy of planar paths are then evaluated from three aspects: "planar path order," "facade appearance," and "landscape facilities." Facade style and spatial entropy Standard value of spatial entropy of landscape facilities The process involves assigning values ​​according to the rules and determining the update potential, as follows: Defining the spatial entropy of planar paths according to their order grade: Standard value of spatial entropy of planar path Potential assessment is performed using a Gaussian discrete distribution model, which refers to a normal distribution model derived from the Gaussian model. If the planar path spatial entropy of the i-th common spatial patch... Located in the last 25%, that is If the value is ≤0.25, it is determined to be a high-potential update region of planar path order, and the label value is assigned as 1; otherwise, it is determined to be a low-potential update region, and the label value is assigned as 0. Delineate the spatial entropy of facade style based on facade appearance. grade: spatial entropy of facade appearance According to the natural breakpoint method, the standard value of the spatial entropy of the facade of the i-th public space patch is... >0.75 or If the value is ≤0.25, it is determined to be a low-potential area for facade appearance renewal, and its assigned value is marked as 0; if the standard value of the facade appearance spatial entropy of the i-th public space patch is ≤0.25, it is determined to be a low-potential area for facade appearance renewal, and its assigned value is marked as 0; It falls in the middle of the entropy value range, i.e., 0.25 < If the value is ≤0.75, it is determined to be a high-potential area for facade renovation, and its assigned value is marked as 1; Landscape facility spatial entropy is classified according to the supporting landscape facilities. grade: Standard values ​​of spatial entropy of landscape facilities in public space patches based on Gaussian discrete distribution model. Divide the space into sections, and let the spatial entropy of the i-th public space patch be... Located in the last 25%, that is If the value is ≤0.25, it is determined to be a high-potential redevelopment area with supporting landscape facilities, and its value is marked as 1; the rest are marked as 0. Based on the objective characteristics of the community, and considering the building age and the coverage rate of public spaces, the total renewal potential of each community's public space patches was verified and optimized according to the following rules, whereby... For low-potential redevelopment areas of facade appearance, the potential of building age characteristics is optimized. If the proportion of buildings in the block unit that are 50 years or more older than the time of determination reaches 50%, it is determined to be a high-potential redevelopment area of ​​facade appearance, and its assigned value is marked as 1. For low-potential regeneration areas of planar path order, the potential of community public space characteristics is optimized. If the community public space coverage rate is lower than the standard, it is determined to be a high-potential regeneration area of ​​planar path order, and its assigned value is marked as 1. The community public space coverage rate refers to the ratio of public green space area to the number of permanent residents in the community; the standard refers to a green space rate of not less than 30% for residential blocks or a per capita public green space area of ​​not less than 70% according to the building climate zoning. The three types of spatial entropy values ​​are superimposed to obtain the final public space patch assignment result. The rules for classifying the update potential level are as follows: public space patches with a final assignment value of 3 are determined as high-potential update spaces, public space patches with a final assignment value of 2 are determined as medium-potential update spaces, public space patches with a final assignment value of 1 are determined as low-potential update spaces, and public space patches with a final assignment value of 0 are deleted. The orientation type of public space renewal is determined based on the spatial entropy standard values ​​of three types: "planar path order - facade appearance - landscape facilities".

6. The method for assessing the renewal potential of community public spaces based on spatial entropy according to claim 1, characterized in that, The process of linking the visualized display of update potential content with various external databases to export a community public space update potential evaluation report includes the following steps: In the geographic information platform, the results of the regeneration potential identification of the public spatial patches are visualized by type and level through the output device. The regeneration potential identification results include the spatial entropy value distribution of the three types of each public spatial patch, the regeneration potential level attribute, and the regeneration potential type information. The obtained results of the renewal potential are linked to the databases of various planning and management departments, and a community public space renewal potential evaluation report is printed out to guide the site selection for public space renewal in each street and community.

7. A computer device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When one or more of the programs are executed by one or more of the processors, the one or more processors implement a method for assessing the potential for updating community public spaces based on spatial entropy as described in any one of claims 1-6.

8. A storage medium containing computer-executable instructions, characterized in that, The computer-executable instructions, when executed by a computer processor, are used to perform a method for assessing the potential for renewal of community public spaces based on spatial entropy as described in any one of claims 1-6.

Images