Intelligent updating method and system for old residential area based on spatial anchor point crowd profiling

Abstract

The application provides an old residential area intelligent updating method and system based on spatial anchor point crowd portrait, and relates to the field of urban planning. The method comprises the following steps: constructing an urban renewal strategy database, determining a spatial anchor point of an old residential area problem space, collecting spatial anchor point resident information, intelligently matching an updating strategy, and intelligently updating the old residential area. The method realizes the intelligent updating of the old residential area by collecting crowd portrait data of the old residential area problem space, matching urban renewal strategy items, and connecting to a concrete 3D printer and a hoisting mechanical arm. The application can cope with the updating of old residential areas in the field of urban planning, realize the matching of updating strategies based on resident crowd portraits, collect more data on the current problems of old residential areas, have higher accuracy of updating strategy matching, have a shorter cycle of "research-design-construction", and have higher qualified rates of standardized operations in updating construction.

Description

Technical Field

[0001] This invention relates to the field of urban planning technology, specifically to a method and system for intelligent renewal of old residential areas based on spatial anchor point population profiles. Background Technology

[0002] The renewal of old residential areas is an important part of urban renewal in the field of urban planning. This invention primarily refers to improving the outdoor space environment of old residential areas. Renewal of old residential areas helps improve residents' quality of life, enhances the overall image of the city, and achieves the intrinsic development of old urban areas. Traditionally, the renewal of old urban residential areas generally requires relevant planning and design personnel to conduct surveys to understand the situation of the residential areas and residents' willingness to renew, formulate renewal strategies based on subjective judgment, and then carry out construction with the help of construction workers and some small machinery. The entire renewal process is relatively slow and consumes a lot of manpower and resources. On the other hand, the development of technologies such as facial recognition, image recognition, and modular assembly provides a more scientific and efficient approach to urban renewal. Combined with spatial anchor point-based crowd profiling technology, public participation in the renewal of old residential areas can be achieved more efficiently.

[0003] Currently, common methods for renovating old residential areas include two approaches. One involves relevant staff conducting on-site assessments of the spatial environment, investigating factors such as green space ratio, damage to paved areas and plazas, the number of public facilities, and the age of buildings, then comparing these assessments with existing residential construction standards to identify and renovate any non-compliant elements. This method fails to consider the residents' wishes regarding renovation, relying solely on construction standards to determine the target for renovation, which contradicts the people-centered approach to residential area renewal. The other approach involves understanding public opinion through questionnaires and interviews, and then incorporating resident feedback into the renovation process. However, this requires significant time and manpower and lacks specificity regarding the specific problems within the residential area, resulting in low efficiency. Summary of the Invention

[0004] (a) Technical problems to be solved

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for intelligent renewal of old residential areas based on spatial anchor point population profiles. This solves the problems of existing methods not considering the renewal wishes of residents in old residential areas, determining renewal targets solely through construction standards, which does not conform to the "people-oriented" concept of residential area renewal, or requiring a large investment of time and manpower, and lacking specificity for the problems in the residential areas, thus resulting in low efficiency.

[0006] (II) Technical Solution

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] Firstly, a method for intelligent renewal of old residential areas based on spatial anchor point population profiles is provided, including:

[0009] The renewal strategies in the texts of urban renewal projects in old residential areas are extracted and integrated to form urban renewal strategy entries; based on the current situation survey information in the texts of urban renewal projects, the entries are labeled with population profile tags to form an urban renewal strategy database;

[0010] Build a 3D model of old residential area, identify the spatial range of entrances and exits, unit entrances, green spaces, squares and facilities in the 3D model, mark the space type, identify the frequency of use and mood changes, and determine the spatial anchor points of problem spaces in old residential area;

[0011] Collect resident information to construct population profiles at spatial anchor points;

[0012] The user profile dimensional preferences for spatial anchor points and urban renewal strategy items are calculated separately to intelligently match urban renewal strategy items for spatial anchor points. The formula for calculating dimensional preferences is as follows:

[0013]

[0014] Among them, G a For creating a user profile, consider the information gain value (I) in dimension a. a Create a user profile using the a-dimensional information entropy.

[0015] Construct a model for urban renewal in old residential areas and complete the intelligent renovation and construction.

[0016] Preferably, the extraction and fusion of renewal strategies from the text of urban renewal projects in old residential areas to form urban renewal strategy entries specifically includes:

[0017] The text of urban renewal projects is scanned, and OCR text recognition technology is used to recognize characters. The spatial type and renewal strategy of the urban renewal project are obtained. The project text and the recognized information are uniformly numbered, entered into the computer, and fused to form urban renewal strategy entries.

[0018] Preferably, the step of tagging entries with population profile labels based on the current status survey information of the urban renewal project text specifically includes:

[0019] Based on the current situation survey information in the urban renewal project documents, six types of population profile information were extracted through keyword identification: age, health status, occupation type, family income, living area, and type of outdoor activities. Then, population profile tags were labeled for each urban renewal strategy item.

[0020] Preferably, the process of constructing a 3D model of the old residential area, identifying the spatial range within the 3D model, marking the spatial type, usage frequency and mood changes, and determining the spatial anchor points of the problem spaces in the old residential area specifically includes:

[0021] A 3D model of the target old residential area was collected and built using oblique photography drones, and the spatial range of the residential area entrances, unit entrances, green spaces, squares and facilities in the 3D model was identified by image recognition technology, and the spatial types were marked.

[0022] Temporary cameras capable of covering various spaces are installed, with a built-in micro-expression mood recognition module trained through deep learning. These cameras record the frequency of space usage and the mood scores of users when entering and leaving the space. Spaces with a usage frequency less than 20% of the average usage frequency, or spaces where the mood score drops to less than 70% of the entry score when leaving, are identified as problem spaces.

[0023] The location closest to the geometric centroid of the old residential area that is within walking distance is identified as the spatial anchor point of the problem space.

[0024] The construction of the micro-expression emotion recognition module specifically includes:

[0025] Miniature devices with facial recognition and expression analysis capabilities were installed in temporary cameras. A raw dataset of micro-expressions was obtained by randomly selecting volunteers to score photos of human facial expressions. Ten regions where micro-expression muscle movements frequently occur were located, and after vectorization, the corresponding scores were used as machine learning labels. The dataset was divided into a training set: validation set: test set = 6:2:2. A supervised classification machine learning algorithm was used for machine learning training to generate a cluster of automatic micro-expression mood score recognition models, thus constructing a micro-expression mood recognition module.

[0026] Preferably, the collection of resident information and the construction of spatial anchor point population profiles specifically include:

[0027] Temporary touchscreen devices with camera functions are installed at spatial anchor points. Residents fill in their basic information, including age, health status, occupation type, family income, and living area, through the touchscreen devices. They can then select their needs for updating the problematic space at the spatial anchor point from four options: "Repair," "Demolition," "Add Facilities," and "Other." The camera function identifies the number of residents, their stay time, and their frequency of stay in the problematic space at the spatial anchor point, thus determining the type of outdoor activities of the residents.

[0028] The touch screen device with camera function is specifically a temporary fixed body device with camera and touch screen functions and a built-in processor. The processor has a built-in algorithm program that automatically identifies and calculates the number of people staying, the stay time, and the stay frequency, and inputs a questionnaire with basic information such as age, health status, occupation type, family income, living area, and update request options.

[0029] The number of users staying refers to the number of users who are identified as the same person through facial recognition and who appear continuously within the spatial anchor point range for more than 5 minutes throughout the day.

[0030] The dwell time refers to the cumulative dwell time of all users throughout the day, calculated using the following formula:

[0031]

[0032] Where T represents the dwell time, and i represents the user ID of the user who is identified as the same person through facial recognition and appears continuously within the spatial anchor point range for more than 5 minutes. This indicates the time point at which user numbered i left the spatial anchor point range. This indicates the time point at which the user with ID i entered the spatial anchor point range;

[0033] Dwell frequency refers to the average number of times all users stay on the platform during the day, calculated using the following formula:

[0034]

[0035] Where C represents the dwell frequency, and Ci represents the number of times a user identified as the same person through facial recognition dwells in a day;

[0036] By integrating data on residents' age, health status, occupation type, family income, living area, and outdoor activity type, six dimensions of population profile labels are generated for each spatial anchor point.

[0037] Preferably, the calculation of the population profile dimension preferences for spatial anchor points and urban renewal strategy items respectively includes:

[0038] The importance of each dimension is determined by calculating the information gain value of each dimension of the population profile of spatial anchor points and urban renewal strategy items. Three key dimensions of the population profile are selected, and the formula for calculating the dimension preference of the population profile of spatial anchor points is obtained by using the information gain value of each dimension as a weight.

[0039] Preferably, the determination of the importance of each dimension specifically includes:

[0040] Using the ID3 decision tree algorithm, calculate the total information entropy I for whether spatial anchor points match urban renewal strategy entries:

[0041]

[0042] Where n is the number of spatial anchor point samples collected, and P i This represents the probability that spatial anchor point i matches an urban renewal strategy entry;

[0043] For a certain dimension 'a' of the user profile, the data sample S containing all spatial anchor points is divided into k subsets {S1, S2, ..., S3} using the category of 'a'. k}, where S k Includes set S with dimension a. k A sample of values, S ik For S k The number of samples where dimension 'a' is the key dimension; the information entropy value of the samples divided according to dimension 'a' is:

[0044]

[0045] Calculate the information gain value of dimension a:

[0046] G a =I(s1, s2, ..., s n )-E a

[0047] Information gain reflects the probability that dimension a is a key dimension. Based on the calculated information gain values ​​of each dimension, the importance of each dimension is determined and used as weights in the dimension preference calculation formula.

[0048] Preferably, the construction of the urban renewal work model for old residential areas and the completion of intelligent renewal construction specifically include:

[0049] In the 3D model of the target old residential area, the spatial anchor point coordinates are used to connect to the matching urban renewal strategy entries. Through semantic analysis, entries that do not meet the spatial anchor point renewal requirements are eliminated, and an urban renewal work sandbox for the target old residential area is constructed.

[0050] Large-scale concrete 3D printers, hoisting robotic arms, and connection devices containing intelligent recognition and analysis modules are installed at spatial anchor points; repair-type spaces are intelligently repaired by connecting large-scale concrete 3D printers to scan the damaged shapes of roads, walls, and hard surfaces; new-type spaces are equipped with modular facilities by connecting intelligent hoisting equipment, and the updated facilities' service life and green space maintenance instructions are output to guide the later maintenance of old residential area renewal spaces.

[0051] The connection device has a built-in intelligent recognition and analysis module, which intelligently connects the urban renewal strategy in the urban renewal work sandbox to a large concrete 3D printer or hoisting robotic arm for intelligent updating.

[0052] The identification module divides the sample data into a training set: validation set: test set = 6:2:2, and uses a deep learning model built with a pre-trained DenseNet161 backbone network and backend modules on the ImageNet dataset for image recognition training to identify the damaged locations and shapes of roads, walls, and open spaces to be updated; the analysis module combines the urban renewal strategy in the urban renewal work sandbox to analyze the modular facilities that need to be installed.

[0053] Preferably, the modular facility includes modular units and connecting components. The modular units include tables and chairs, trash cans, pavilions, signs, and planting boxes, made of wood or wood-concrete composite materials, which are easy to reuse, maintain, and disassemble and recycle after disposal. The plant species are herbaceous plants, shrubs, and small trees that are suitable for the local climate.

[0054] Secondly, a smart renovation system for old residential areas based on spatial anchor point population profiles is provided, including:

[0055] The urban renewal strategy database construction module is used to extract and integrate renewal strategies from urban renewal project texts to form urban renewal strategy entries; and to tag the entries with population profile labels based on the current status survey information of the urban renewal project texts to form an urban renewal strategy database.

[0056] The spatial anchor point determination module for the problem space of old residential areas is used to build a 3D model of old residential areas, identify the spatial range in the 3D model, mark the spatial type, use frequency and mood changes, and determine the spatial anchor points of the problem space of old residential areas.

[0057] The spatial anchor point resident information collection module is used to collect resident information and construct a spatial anchor point population profile.

[0058] The updated strategy intelligent matching module is used to update the strategy intelligent matching, and calculate the population profile dimension preferences for spatial anchor points and urban renewal strategy items respectively.

[0059] The intelligent renovation and construction module for old residential areas is used to match urban renewal strategy items based on the spatial anchor point population profile dimensions and preferences, form a renovation work sandbox, and complete the intelligent renovation and construction of old residential areas.

[0060] (III) Beneficial Effects

[0061] (1) The present invention provides a method and system for intelligent renewal of old residential areas based on spatial anchor point population profiles. By identifying the problem spaces of old residential areas, combining the intelligent matching and renewal strategy of population profiles, and carrying out intelligent renewal construction, the work cycle of "research-design-construction" is shortened to 60%.

[0062] (2) The present invention provides an intelligent renewal method and system for old residential areas based on spatial anchor point population profiles. It uses semantic recognition to extract and integrate renewal strategies in urban renewal project texts to form urban renewal strategy entries including renewal space types. This enables large-scale data sharing of renewal strategies and helps to carry out renewal work.

[0063] (3) The present invention provides a method and system for intelligent renewal of old residential areas based on spatial anchor point population profiles. By labeling urban renewal strategy items with population profile tags of spatial anchor points in old residential areas, the information gain values ​​of each dimension of the ID3 decision tree algorithm are used to determine the importance of each dimension, thereby achieving update strategy matching based on population profiles. This breaks through the traditional, relatively subjective expert judgment, making the whole process more objective, intelligent, and automated, with a matching accuracy of over 85%.

[0064] (4) The present invention provides an intelligent renewal method and system for old residential areas based on spatial anchor point population profiles. By identifying the problem spaces in old residential areas, temporary cameras are set up at the anchor points of the problem spaces to record the space usage characteristics in a targeted manner, making the entire collection process more automated and increasing the amount of data collected on the current status of old residential areas by 400%, while also improving the targeting of data collection.

[0065] (5) The present invention provides a method and system for intelligent renewal of old residential areas based on spatial anchor point population profiles. It utilizes a large concrete 3D printer and a hoisting robotic arm, combined with modular facilities, to make the renewal construction more standardized and reduce the rework rate of substandard construction by 50%. Attached Figure Description

[0066] Figure 1 This is a flowchart of the method of the present invention;

[0067] Figure 2 This is a 3D model of an old residential area constructed using oblique photography technology in an embodiment of the present invention;

[0068] Figure 3 This is a diagram showing facial recognition and the number of people staying in the area, as described in this embodiment of the invention.

[0069] Figure 4 This is an example diagram of the spatial status analysis before the update in an embodiment of the present invention;

[0070] Figure 5 This is an example diagram illustrating the analysis of spatial update results in an embodiment of the present invention. Detailed Implementation

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

[0072] Example

[0073] like Figure 1-5 As shown, this embodiment of the invention provides a method for intelligent renewal of old residential areas based on spatial anchor point population profiles, characterized by comprising:

[0074] Urban renewal strategies are extracted and integrated from urban renewal project texts to form urban renewal strategy entries; and based on the current status survey information in the urban renewal project texts, population profile tags are labeled for the entries to form an urban renewal strategy database.

[0075] Build a 3D model of old residential areas, identify the spatial range in the 3D model, mark the spatial type, use frequency and mood changes, and determine the spatial anchor points of problem spaces in old residential areas;

[0076] Collect resident information to construct population profiles at spatial anchor points;

[0077] The updated strategy intelligent matching calculates the preferences of the demographic profile dimensions for spatial anchor points and urban renewal strategy items, respectively. The calculation formula is as follows:

[0078]

[0079] Among them, G a For creating a user profile, consider the information gain value (I) in dimension a. a Create a user profile using the a-dimensional information entropy.

[0080] Based on the spatial anchor point population profile dimensions and preferences, update urban renewal strategy items, form a renewal work sandbox, and complete the intelligent renewal construction of old residential areas.

[0081] Specifically:

[0082] (1) Constructing an urban renewal strategy database, specifically including:

[0083] (1.1) Download and input the texts of urban renewal projects published on the official websites of the natural resources and planning bureaus of cities across the country in the past three years, and scan the images.

[0084] (1.2) Character recognition is performed using OCR text recognition technology. Then, through the technical solution of "word segmentation" + "keyword matching" + "template matching", the spatial type and renewal strategy of each urban renewal project are obtained. The project text and the recognized information are uniformly numbered and entered into the computer and fused to form urban renewal strategy entries.

[0085] Table 1. Keywords for Semantic Recognition of Project Spatial Types

[0086]

[0087] Table 2. Project Update Strategy: Semantic Recognition Keywords

[0088]

[0089] Table 3. Illustration of Urban Renewal Strategy Items

[0090]

[0091] (1.3) Based on the current status survey information in the urban renewal project documents, six population profiles—resident age, health status, occupation type, family income, living area, and outdoor activity type—are extracted from each urban renewal strategy item through keyword identification. Population profile tags are then assigned to each urban renewal strategy item to form an urban renewal strategy database. This database is updated in real-time based on information released on the official websites of the natural resources and planning bureaus of each city.

[0092] Table 4. Categorization of Audience Profile Tags

[0093]

[0094]

[0095] (2) Determine the spatial anchor points of the problem space in this old residential area, specifically including:

[0096] (2.1) A three-dimensional model of the target old residential area was collected and built by oblique photography drone, and the spatial range of the residential area entrance, unit entrance, green space, square and facilities in the three-dimensional model was identified by image recognition technology, and the spatial type was marked.

[0097] (2.2) Install temporary cameras that can cover various spaces, equip the temporary cameras with miniature devices with facial recognition and expression analysis functions, obtain the original dataset of micro-expressions by randomly selecting volunteers to score human facial expression photos, locate 10 areas where micro-expression muscle movement occurs frequently, vectorize them and use the corresponding scores as machine learning labels, and divide them into training set: validation set: test set = 6:2:2. Use supervised classification machine learning algorithm to perform machine learning training, generate micro-expression mood score automatic recognition model cluster, and thus construct micro-expression mood recognition module.

[0098] (2.3) Record the frequency of space use and the mood score of users when entering and leaving the space. Spaces with a usage frequency of less than 20% of the average usage frequency, and spaces where the mood score drops to less than 70% of the entry score when leaving, are identified as problem spaces.

[0099] (2.4) The location closest to the geometric centroid of the problem space that is within walking distance is determined as the spatial anchor point of the problem space of the old residential area.

[0100] (3) Collection of resident information at spatial anchor points, specifically including:

[0101] (3.1) Install a temporary touch screen device with camera function at the spatial anchor point and a temporary fixed body device with built-in processor. The processor has built-in algorithm program to automatically identify and calculate the number of residents n, the stay time T, and the stay frequency C, and enters a questionnaire with basic information such as age, health status, occupation type, family income, living area and update request options.

[0102] The number of users staying, n, refers to the number of users who are identified as the same person through facial recognition and who appear continuously within the spatial anchor point range for more than 5 minutes throughout the day.

[0103] Dwell time T refers to the cumulative dwell time of all users throughout the day, and its calculation formula is as follows:

[0104]

[0105] Where i represents the user ID that is identified as the same person through facial recognition and appears continuously within the spatial anchor point range for more than 5 minutes, and T Ai T represents the time point at which user numbered i leaves the spatial anchor point range. Bi This indicates the time point at which the user with ID i entered the spatial anchor point range.

[0106] Dwell frequency C refers to the average number of times all users dwell in a day, and its calculation formula is as follows:

[0107]

[0108] Ci represents the number of times a user identified as the same person through facial recognition stays on the site in a day.

[0109] (3.2) Residents fill in their basic information, including age, health status, occupation type, family income and living area, through a touch screen device, and select their update requests for the problem space at the space anchor point from four options: “Repair”, “Demolition”, “Add Facilities”, and “Other”.

[0110] (3.3) By using the camera function to identify the number of residents, the duration of their stay, and the frequency of their stay in the problem space at the spatial anchor point, the type of outdoor activities of the residents can be determined.

[0111] (4) Intelligent matching of update strategies for spatial anchor points in the update strategy database, specifically including:

[0112] (4.1) Integrate the data identified in step three, including age, health status, occupation type, family income, living area, and outdoor activity type, and label the six dimensions of the population profile at each spatial anchor point.

[0113] (4.2) Select spatial anchor point samples and urban renewal strategy item samples, with a ratio of 1:10. Use the ID3 decision tree algorithm to calculate the total information entropy I for whether the spatial anchor point matches the urban renewal strategy item.

[0114]

[0115] Taking the age dimension of a user profile as an example, the data sample containing all spatial anchors is divided into six subsets {S1, S2, ..., S6} using the six age categories. Each subset contains samples with the age attributes of infants (0-6), children (7-12), teenagers (13-17), young adults (18-45), middle-aged (46-69), and elderly (69 and above). Based on the number of samples with age as the key dimension, the information entropy value of the samples divided by age dimension is calculated as follows:

[0116]

[0117] Further calculation of the information gain value for the age dimension:

[0118] G(age) = I(s1, s2) - E(age) = 0.7218

[0119] Repeat the above steps to obtain the information gain values ​​of the six dimensions of the population profile, as shown in Table 5. The three items with the highest values ​​(age, health status, and type of outdoor activity) are used as the key dimensions for subsequent matching and screening.

[0120] Table 5 Information Gain Values ​​of Audience Profiles

[0121]

[0122] (4.3) Substitute the calculated information gain values ​​for each dimension as weights into the dimension preference calculation formula to obtain:

[0123] P = 0.7218I 年龄 +0.6538I 健康状况 +0.5392I 职业类型 +0.3759I 家庭收入 +0.2864I 居住面积 +0.8728I 室外活动类型

[0124] Based on the specific I values ​​of different spatial anchors or strategy databases, their respective dimensional preferences are calculated and matched. Urban renewal strategy entries that share the same spatial type as the spatial anchor, and whose three key dimensions (age, health status, and outdoor activity type) have a similarity of over 70% and a difference in dimensional preferences of less than 10% are selected.

[0125] (5) Based on the matching update strategy, construct intelligent updates for old residential areas, specifically including:

[0126] (5.1) In the three-dimensional model of old residential area, according to the spatial anchor point coordinates, connect the matching urban renewal strategy items of repair and addition, and through semantic analysis, remove the items that do not meet the spatial anchor point renewal requirements to form the urban renewal work sand table of the target old residential area.

[0127] Table 6 Update Strategy Entry Matching Table

[0128]

[0129]

[0130] (5.2) Install connection devices, a large-scale concrete 3D printer, and a hoisting robotic arm at each spatial anchor point. The connection device has a built-in intelligent recognition and analysis module. The recognition module divides the sample data into a training set: validation set: test set = 6:2:2, and uses a deep learning model built with a pre-trained DenseNet161 backbone network and backend modules on the ImageNet dataset for image recognition training to identify the damaged locations and shapes of roads, walls, and open spaces to be updated. The analysis module, combined with the connection update strategy, extracts the category information of the identified spaces.

[0131] (5.3) Based on the identification of repair and new spaces by the connection device, different devices are intelligently connected for updates. Repair spaces are connected to a large-scale concrete 3D printer for intelligent repair of damaged roads, walls, and hard surfaces. New spaces are installed using intelligent hoisting equipment, incorporating modular facilities including tables, chairs, trash cans, pavilions, signs, and planting boxes, along with herbs, shrubs, and small trees suitable for the local climate. Furthermore, the updated facilities' service life and green space maintenance instructions are output to guide the later maintenance of renovated spaces in old residential areas.

[0132] Another embodiment of the present invention provides a smart renovation system for old residential areas based on spatial anchor point population profiles, including:

[0133] The urban renewal strategy database construction module is used to extract and integrate renewal strategies from the texts of urban renewal projects in old residential areas to form urban renewal strategy entries; and to label the entries with population profile tags based on the current status survey information of the urban renewal project texts to form an urban renewal strategy database.

[0134] The spatial anchor point determination module for problem spaces in old residential areas is used to build a 3D model of old residential areas, identify the spatial range of entrances and exits, unit entrances, green spaces, squares and facilities in the 3D model, mark the space type, identify the frequency of use and mood changes, and determine the spatial anchor points of problem spaces in old residential areas.

[0135] The spatial anchor point resident information collection module is used to collect resident information and construct a spatial anchor point population profile.

[0136] The updated strategy intelligent matching module is used to calculate the demographic preferences of spatial anchors and urban renewal strategy items, respectively, so as to intelligently match urban renewal strategy items for spatial anchors.

[0137] The intelligent renovation and construction module for old residential areas is used to build a sand table of urban renewal work for old residential areas and complete the intelligent renovation and construction.

[0138] Embodiments of this application may be provided as methods or computer program products. Therefore, this application may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of this application may be implemented in various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0139] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0140] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0141] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0142] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. An old residential area intelligent updating method based on spatial anchor point crowd profiling, characterized in that, include: The renewal strategies in the text of urban renewal projects in old residential areas are extracted and integrated to form urban renewal strategy entries; Based on the current status survey information of urban renewal project texts, the entries are labeled with population profile tags to form an urban renewal strategy database; Build a 3D model of old residential area, identify the spatial range of entrances and exits, unit entrances, green spaces, squares and facilities in the 3D model, mark the space type, identify the frequency of use and mood changes, and determine the spatial anchor points of problem spaces in old residential area; Collect resident information to construct population profiles at spatial anchor points; The user profile dimensional preferences for spatial anchor points and urban renewal strategy items are calculated separately to intelligently match urban renewal strategy items for spatial anchor points. The formula for calculating dimensional preferences is as follows: Wherein, G a is the a-dimensional information gain value of the crowd portrait, I a is the a-dimensional information entropy of the crowd portrait; Construct a model for urban renewal in old residential areas and complete the intelligent renovation and construction. 2.The old residential area intelligent updating method based on spatial anchor point crowd profiling according to claim 1, characterized in that, The extraction and integration of renewal strategies from urban renewal project texts of old residential areas to form urban renewal strategy entries specifically include: The text of urban renewal projects is scanned, and OCR text recognition technology is used to recognize characters. The spatial type and renewal strategy of the urban renewal project are obtained. The project text and the recognized information are uniformly numbered, entered into the computer, and fused to form urban renewal strategy entries. 3.The old residential area intelligent updating method based on spatial anchor point crowd profiling according to claim 1, characterized in that, The step of tagging entries with population profile labels based on the current status survey information from urban renewal project texts specifically includes: Based on the current situation survey information in the urban renewal project documents, six types of population profile information were extracted through keyword identification: age, health status, occupation type, family income, living area, and type of outdoor activities. Then, population profile tags were labeled for each urban renewal strategy item. 4.The old residential area intelligent updating method based on spatial anchor point crowd profiling according to claim 1, characterized in that, The process of constructing a 3D model of the old residential area, identifying the spatial range of entrances and exits, unit entrances, green spaces, plazas, and facilities within the 3D model, marking space types, identifying usage frequency and mood changes, and determining spatial anchor points for problem spaces in the old residential area, specifically includes: A 3D model of the target old residential area was collected and built using oblique photography drones, and the spatial range of the residential area entrances, unit entrances, green spaces, squares and facilities in the 3D model was identified by image recognition technology, and the spatial types were marked. Temporary cameras capable of covering various spaces are installed, with a built-in micro-expression mood recognition module trained through deep learning. These cameras record the frequency of space usage and the mood scores of users when entering and leaving the space. Spaces with a usage frequency less than 20% of the average usage frequency, or spaces where the mood score drops to less than 70% of the entry score when leaving, are identified as problem spaces. The location closest to the geometric centroid of the old residential area that is within walking distance is identified as the spatial anchor point of the problem space. The construction of the micro-expression emotion recognition module specifically includes: Miniature devices with facial recognition and expression analysis capabilities were installed in temporary cameras. A raw dataset of micro-expressions was obtained by randomly selecting volunteers to score photos of human facial expressions. Ten regions where micro-expression muscle movements occur frequently were located, and after vectorization, the corresponding scores were used as machine learning labels. The dataset was divided into a training set: validation set: test set = 6:2:

2. A supervised classification machine learning algorithm was used for machine learning training to generate a cluster of automatic micro-expression mood score recognition models, thus constructing a micro-expression mood recognition module.

5. The method according to claim 1, wherein, The collection of resident information to construct spatial anchor point population profiles specifically includes: Temporary touchscreen devices with camera functions are installed at spatial anchor points. Residents fill in their basic information, including age, health status, occupation type, family income, and living area, through the touchscreen devices. They can then select their needs for updating the problematic space at the spatial anchor point from four options: "Repair," "Demolition," "Add Facilities," and "Other." The camera function identifies the number of residents, their stay time, and their frequency of stay in the problematic space at the spatial anchor point, thus determining the type of outdoor activities of the residents. The touch screen device with camera function is specifically a temporary fixed body device with camera and touch screen functions and a built-in processor. The processor has a built-in algorithm program that automatically identifies and calculates the number of people staying, the stay time, and the stay frequency, and inputs a questionnaire with basic information such as age, health status, occupation type, family income, living area, and update request options. Wherein, the number of resident users n refers to the number of users who are identified as the same person through facial recognition and who appear continuously within the spatial anchor point range for more than 5 minutes in a day; The dwell time refers to the cumulative dwell time of all users throughout the day, calculated using the following formula: wherein T denotes a residence time, i denotes a user number who is the same person by face recognition and continuously appears in a space anchor point range for > 5 min, denotes a time point at which the user numbered i leaves the space anchor point range, denotes a time point at which the user numbered i enters the space anchor point range; Dwell frequency refers to the average number of times all users stay on the platform during the day, calculated using the following formula: where C denotes the residence frequency, C i denotes the number of times of residence of the user who is the same person through face recognition in a day; By integrating data on residents' age, health status, occupation type, family income, living area, and outdoor activity type, six dimensions of population profile labels are generated for each spatial anchor point. 6.The old residential district intelligent updating method based on spatial anchor point crowd profiling according to claim 1, characterized in that, The calculation of the population profile dimensions and preferences for spatial anchor points and urban renewal strategy items, respectively, specifically includes: The importance of each dimension is determined by calculating the information gain value of each dimension of the population profile of spatial anchor points and urban renewal strategy items. Three key dimensions of the population profile are selected, and the formula for calculating the dimension preference of the population profile of spatial anchor points is obtained by using the information gain value of each dimension as a weight. 7.The old residential district intelligent updating method based on spatial anchor point crowd profiling according to claim 1, characterized in that, The aforementioned urban renewal work model for old residential areas, which completes the intelligent renewal construction, specifically includes: In the 3D model of the target old residential area, the spatial anchor point coordinates are used to connect to the matching urban renewal strategy entries. Through semantic analysis, entries that do not meet the spatial anchor point renewal requirements are eliminated, and an urban renewal work sandbox for the target old residential area is constructed. Large-scale concrete 3D printers, hoisting robotic arms, and connection devices containing intelligent recognition and analysis modules are installed at spatial anchor points; repair-type spaces are intelligently repaired by connecting large-scale concrete 3D printers to scan the damaged shapes of roads, walls, and hard surfaces; new-type spaces are equipped with modular facilities by connecting intelligent hoisting equipment, and the updated facilities' service life and green space maintenance instructions are output to guide the later maintenance of old residential area renewal spaces. The connection device has a built-in intelligent recognition and analysis module, which intelligently connects the urban renewal strategy in the urban renewal work sandbox to a large concrete 3D printer or hoisting robotic arm for intelligent updating. The identification module divides the sample data into a training set: validation set: test set = 6:2:2, and uses a deep learning model built with a pre-trained DenseNet161 backbone network and backend modules on the ImageNet dataset for image recognition training to identify the location and shape of damage to roads, walls and open spaces to be updated; the analysis module combines the urban renewal strategy in the urban renewal work sandbox to analyze the modular facilities that need to be installed. 8.The old residential area intelligent updating method based on spatial anchor point crowd profiling according to claim 7, characterized in that, The modular facility includes modular units and connecting components. The modular units include tables and chairs, trash cans, pavilions, signs, and planting boxes. They are made of wood or wood-concrete composite materials, which are convenient for reuse, maintenance, and disassembly and recycling after disposal. The plant species in the planting boxes are herbaceous plants, shrubs, and small trees that are suitable for the local climate.

9. An old residential area intelligent updating system based on spatial anchor point crowd profiling, characterized in that, include: The urban renewal strategy database construction module is used to extract and integrate renewal strategies from the texts of urban renewal projects in old residential areas to form urban renewal strategy entries; and to label the entries with population profile tags based on the current status survey information of the urban renewal project texts to form an urban renewal strategy database. The spatial anchor point determination module for problem spaces in old residential areas is used to build a 3D model of old residential areas, identify the spatial range of entrances and exits, unit entrances, green spaces, squares and facilities in the 3D model, mark the space type, identify the frequency of use and mood changes, and determine the spatial anchor points of problem spaces in old residential areas. The spatial anchor point resident information collection module is used to collect resident information and construct a spatial anchor point population profile. The updated strategy intelligent matching module is used to calculate the dimensional preferences of the demographic profiles for spatial anchor points and urban renewal strategy items, respectively, thereby intelligently matching urban renewal strategy items for spatial anchor points. The formula for calculating dimensional preferences is as follows: Wherein, G a is the a-dimensional information gain value of the crowd portrait, I a is the a-dimensional information entropy of the crowd portrait; The intelligent renovation and construction module for old residential areas is used to build a sand table of urban renewal work for old residential areas and complete the intelligent renovation and construction.

Images