Visitor travel path analysis method based on 3D map

Abstract

The invention discloses a visitor travel path analysis method based on a 3D map, and relates to the technical field of security systems, and the method comprises the following steps: S1, system demand analysis and planning; s2, modeling and optimizing a 3D map; s3, designing a visitor identity verification mechanism; s4, generation and management of visitors' access certificates; s5, setting an electronic fence in the key area; s6, collecting and processing travel path data in real time; s7, performing travel path data analysis and anomaly detection; s8, modeling a visitor behavior mode; and S9, designing and realizing a visual instrument panel. According to the visitor travel path analysis method based on the 3D map, through real-time data acquisition, an intelligent algorithm and behavior pattern modeling, fine management and anomaly detection of visitor activities are realized. The method not only can track the traveling track of visitors in the office space in real time and timely recognize and respond to potential safety risks, but also classifies the visitors according to behavior characteristics through clustering analysis, and provides powerful data support for safety monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of security systems, and in particular to a visitor behavior analysis method based on a 3D map. Background Art

[0002] As the scale of power grid enterprises continues to expand and the complexity of their business increases, visitor management faces increasingly severe security challenges. At present, although there are existing systems to manage external visitors through strict review and registration, there is still a deficiency in real-time monitoring of visitor activity trajectories, which may lead to unauthorized personnel entering sensitive areas such as machine rooms and monitoring rooms, thus posing potential safety hazards. Traditional visitor management methods often rely on manual records and static monitoring, which are difficult to adapt to dynamically changing visitor behaviors and diverse security needs. In addition, the lack of in-depth analysis of visitor behavior makes it difficult to form accurate security strategies and response mechanisms. Therefore, the introduction of intelligent and digital management methods, combined with 3D maps and advanced positioning technologies, is particularly important for improving the efficiency and safety of visitor behavior management. Summary of the invention

[0003] In view of the deficiencies of the prior art, the present invention provides a visitor behavior analysis method based on a 3D map to solve the problems raised in the above background technology.

[0004] To achieve the above object, the present invention provides the following technical solution: a visitor behavior analysis method based on a 3D map, comprising the following steps: S1. System requirements analysis and planning; Conduct a comprehensive analysis of the needs for visitor behavior management, and through communication with various departments, understand the specific needs and concerns of different positions for visitor management, including visitor identity authentication, behavior monitoring, and data statistics functions; S2, 3D map modeling and optimization; Establish an accurate 3D map of the office space. During the modeling process, high-precision laser scanning and photogrammetry technology are used to generate a 3D model of the building and mark important areas such as the computer room, monitoring room and office; S3,Design of visitor authentication mechanism; Design multi-factor authentication mechanisms to increase the security of visitor entry; S4. Generation and management of visitor passes; Automatically generate active tag visitor pass based on visitor identity information; S5. Setting up electronic fences in key areas; Set up electronic fences in the 3D map to identify important production sites, offices and other sensitive areas; S6. Real-time collection and processing of behavior data; After a visitor enters the office building, the visitor's activity trajectory is tracked in real time through positioning base stations and active tags. During the data collection process, the visitor's real-time location is associated with the 3D map, and its travel path is recorded. The collected location information is cleaned and corrected through data processing algorithms to ensure the accuracy of the data; S7, behavior data analysis and anomaly detection; Use data mining and machine learning algorithms to conduct in-depth analysis of visitor behavior data and identify common visitor behavior patterns; by learning from historical data, quickly identify abnormal behaviors, such as a visitor frequently entering sensitive areas or staying too long, and generate alerts in a timely manner; S8, visitor behavior pattern modeling; Based on historical data, the visitor behavior patterns are modeled to identify different types of visitor behaviors. Cluster analysis methods are used to classify visitors according to their behavior characteristics to form different behavior labels, such as regular visitors and risky visitors. S9. Visual dashboard design and implementation; Design a visual dashboard to display real-time data and behavior analysis results of visitors; the dashboard includes information on visitor flow, visitor activity trajectory, abnormal alarms, and provides intuitive charts combined with 3D maps.

[0005] To further optimize the technical solution, in step S2, the model's textures and materials are optimized to improve the model's interactivity and visual appeal in practical applications, and the map is optimized to ensure smooth operation on different devices, providing basic support for subsequent visitor behavior analysis.

[0006] To further optimize the technical solution, in step S3, the visitor provides a valid certificate when registering, scans the certificate information through OCR technology, and quickly checks the blacklist information in the database; Biometric identification based on face recognition or fingerprint recognition ensures the uniqueness and accuracy of visitor identity.

[0007] To further optimize the technical solution, in step S4, the pass contains not only the basic information of the visitor, but also an embedded wireless communication module for real-time tracking of the visitor's movements; The pass also has a dynamic generation function, which updates information in real time based on the visitor's behavior and automatically cancels the card when it expires to prevent security risks caused by abuse or loss.

[0008] To further optimize the technical solution, in step S5, the setting of the electronic fence is based on actual security requirements, taking into account the access rights and monitoring requirements of different areas; When a visitor enters a critical area set by the electronic fence, an alarm is automatically sent to notify security personnel so that they can intervene in time to prevent unauthorized persons from entering sensitive areas and enhance security protection capabilities.

[0009] To further optimize the technical solution, in step S6, a visitor behavior real-time tracking model is built in, and the visitor's activity trajectory is tracked in real time based on the model, including position calculation, path length calculation, distance calculation and state determination; The position calculation is as follows: ; in, Indicates the position information of the last moment, and updates the current position information in combination with the current speed and time interval; The path length calculation is as follows: ; Used to calculate the current path length as the path length at the previous moment plus the distance between the current position and the previous position; The distance calculation is as follows: ; in, is the target location coordinate, which is used to calculate the distance between the visitor's current location and the target location; The status determination is as follows: ; Determine the visitor's activity status based on the visitor's speed, the distance between the current location and the target, and the status of the electronic fence; In the above formula, For visitors at time The position coordinates ( ),in is the plane coordinate, is the height coordinate; It is a function of the visitor's path, indicating the length of the path from the starting point to the current point; The distance between the visitor's current location and the target location; For visitors at time The speed is defined as ,in is the change in path length, is the time interval; The activity status of visitors is defined as different behavior patterns, such as walking, staying, and moving quickly; The status of the electronic fence, indicating whether the visitor is in the allowed area.

[0010] To further optimize the technical solution, in step S7, based on the residence time anomaly detection model, the entry frequency anomaly detection model, the overall behavior anomaly index model and the alarm trigger condition model, the behavior data analysis and anomaly detection are performed; Set the following model parameters: For visitors at time The position coordinates ( ); For visitors in the area Duration of stay within Entrance area for visitors frequency; The visitor's behavior status is combined with location, speed, and regional attributes; is the abnormality index, which reflects the abnormality of visitor behavior; and is the mean and standard deviation of the regular behavior pattern, which is used to measure the difference between normal behavior and abnormal behavior.

[0011] To further optimize the technical solution, in step S7: The dwell time anomaly detection model is as follows: ; in, Indicates that the visitor is in a certain area The residence time within and Respectively represent the mean and standard deviation of the normal residence time in the area; This model is used to measure whether the length of time a visitor spends in a certain area is abnormal; The entry frequency anomaly detection model is as follows: ; This model represents the degree to which the frequency of visitors entering a certain area deviates from the normal pattern. Enter an area for visitors The frequency, and are the mean and standard deviation of the frequency of regular visitors in the area; The overall behavior abnormality index model is as follows: ; in, is the current visitor's location, is the center point of the safe area. represents the typical distance distribution within the safe area, is the weight coefficient of each factor; The model combines multiple behavioral factors to calculate the visitor’s overall anomaly index; The alarm trigger condition model is as follows: ; in, For the set threshold, when the visitor's behavior index is abnormal Exceeding the threshold , the alarm will be triggered automatically.

[0012] To further optimize the technical solution, in step S8, when modeling, the following model parameters are set: For the The behavioral feature vector of each visitor includes multiple dimensions, such as residence time, visit frequency, and path complexity; For the The center point of the class group, i.e., the centroid, represents the cluster center; For Visitors With cluster center The distance between is the total number of clusters, indicating the number of visitor behavior patterns; is the weight vector, defining the importance of each feature in cluster analysis; and For the The mean and standard deviation of the class are used for subsequent anomaly detection and label generation, respectively; Based on the above model parameters, the distance calculation formula, clustering update formula and behavior label generation formula are constructed.

[0013] To further optimize this technical solution, the distance calculation formula is as follows: ; The formula calculates the visitor With cluster center The distance is the number of feature dimensions; by introducing the weight vector , adjust the distance calculation according to the importance of different features; The cluster update formula is as follows: ; in, Indicates that it belongs to Number of visitors to the class; This formula is used to update the cluster centers to ensure that the cluster centers reflect the characteristics of the current cluster groups; The behavior label generation formula is as follows: ; in, and It is the set thresholds that define labels for regular, risky, and abnormal visitors.

[0014] Compared with the prior art, the present invention provides a visitor behavior analysis method based on 3D maps, which has the following beneficial effects: This visitor behavior analysis method based on 3D maps achieves refined management and anomaly detection of visitor activities through real-time data collection, intelligent algorithms and behavior pattern modeling. This method can not only track the movement of visitors in the office in real time, identify and respond to potential security risks in a timely manner, but also classify visitors according to their behavioral characteristics through cluster analysis, providing strong data support for security monitoring. Ultimately, it improves the safety protection level of power grid companies, effectively reduces safety hazards caused by improper visitor management, and ensures the security of internal facilities and data. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of a flow chart of a visitor behavior analysis method based on a 3D map proposed by the present invention; Figure 2 A schematic diagram of the composition and flow of a visitor behavior real-time tracking model in a visitor behavior analysis method based on a 3D map proposed by the present invention; Figure 3 The formula in step S8 of the visitor behavior analysis method based on 3D map proposed by the present invention and its flow chart; Figure 4 This is a formula and a flow chart of step S7 of a visitor behavior analysis method based on a 3D map proposed by the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] Example: See also Figure 1, a visitor behavior analysis method based on 3D map, comprising the following steps: S1. System requirements analysis and planning Conduct a comprehensive analysis of the needs for visitor behavior management, and through communication with various departments, understand the specific needs and concerns of different positions for visitor management, including visitor identity verification, behavior monitoring, and data statistics functions.

[0018] In this embodiment, the scalability of the system should be considered during the planning phase to adapt to possible future technology upgrades and changes in business requirements. At the same time, the hardware architecture and software environment of the system should be determined to provide a basis for subsequent system development and implementation.

[0019] S2. 3D map modeling and optimization Create accurate 3D office space maps. During the modeling process, high-precision laser scanning and photogrammetry technology are used to generate a three-dimensional model of the building, marking important areas such as computer rooms, monitoring rooms and offices.

[0020] In this embodiment, in step S2, the model's textures and materials are optimized to improve the model's interactivity and visual appeal in practical applications, and the map is optimized to ensure smooth operation on different devices, providing basic support for subsequent visitor behavior analysis.

[0021] S3, Visitor Authentication Mechanism Design Design multi-factor authentication mechanisms to increase the security of visitor entry.

[0022] In this embodiment, in step S3, the visitor provides a valid certificate when registering, and the certificate information is scanned by OCR technology to quickly check the blacklist information in the database; Biometric identification methods based on face recognition or fingerprint recognition ensure the uniqueness and accuracy of visitor identities. The identity verification process should be simplified to avoid inconvenience to visitors while ensuring the efficiency and security of the system.

[0023] S4. Generation and management of visitor passes Automatically generate active tag visitor pass based on visitor's identity information.

[0024] In this embodiment, in step S4, the pass contains not only the basic information of the visitor, but also an embedded wireless communication module for real-time tracking of the visitor's movements; The pass also has a dynamic generation function, which updates information in real time based on the visitor's behavior and automatically cancels the card when it expires to prevent security risks caused by abuse or loss.

[0025] S5. Setting up electronic fences in key areas Set up electronic fences in the 3D map to identify important production sites, offices and other sensitive areas.

[0026] In this embodiment, in step S5, the setting of the electronic fence is based on actual security requirements, taking into account the access rights and monitoring requirements of different areas; When a visitor enters a critical area set by the electronic fence, an alarm is automatically sent to notify security personnel so that they can intervene in time to prevent unauthorized persons from entering sensitive areas and enhance security protection capabilities.

[0027] S6. Real-time collection and processing of behavior data After visitors enter the office building, their activities are tracked in real time through positioning base stations and active tags. During the data collection process, the visitor's real-time location is associated with the 3D map and their travel path is recorded. The collected location information is cleaned and corrected through data processing algorithms to ensure the accuracy of the data.

[0028] like Figure 2 As shown, in this embodiment, in step S6, a visitor behavior real-time tracking model is built in, and the visitor's activity trajectory is tracked in real time based on the model, including position calculation, path length calculation, distance calculation and state determination.

[0029] The position calculation is as follows: ; in, Indicates the position information of the last moment, and updates the current position information in combination with the current speed and time interval; The path length calculation is as follows: ; Used to calculate the current path length as the path length at the previous moment plus the distance between the current position and the previous position; The distance calculation is as follows: ; in, is the target location coordinate, which is used to calculate the distance between the visitor's current location and the target location; The status determination is as follows: ; Determine the visitor's activity status based on the visitor's speed, the distance between the current location and the target, and the status of the electronic fence; In the above formula, For visitors at time The position coordinates ( ),in is the plane coordinate, is the height coordinate; It is a function of the visitor's path, indicating the length of the path from the starting point to the current point; The distance between the visitor's current location and the target location; For visitors at time The speed is defined as ,in is the change in path length, is the time interval; The activity status of visitors is defined as different behavior patterns, such as walking, staying, and moving quickly; The status of the electronic fence, indicating whether the visitor is in the allowed area.

[0030] When the model is used, it includes the following processes: Real-time location update: When a visitor enters the office building, the system obtains their location data in real time through the positioning base station and active tags, using the location calculation formula Update the visitor's current coordinates. This process needs to be updated multiple times per second in the system to ensure the real-time location.

[0031] Path tracking and analysis: Using the path length calculation formula The system can record the visitor's path in real time and obtain the complete path by continuously accumulating the displacement of each step, which provides data support for subsequent behavior analysis.

[0032] Behavioral state monitoring: combined with state determination formula , the system can identify the different activity states of visitors. If a visitor stays in a certain area for too long (for example, the speed is 0), an alarm may be triggered to alert security personnel to intervene.

[0033] Electronic fence monitoring: by distance calculation and status determination , the system can determine in real time whether the visitor has entered a key area or a sensitive area. Once a visitor is detected entering a key area set by the electronic fence, the system will immediately sound an alarm to ensure safety protection.

[0034] By combining the visitor's real-time location, path length, distance calculation and activity status, a systematic solution is provided for real-time monitoring of visitor behavior. Based on the multi-dimensional analysis of visitor behavior, visitor management is more refined and potential safety hazards can be discovered and handled in a timely manner.

[0035] S7. Behavior data analysis and anomaly detection Use data mining and machine learning algorithms to conduct in-depth analysis of visitor behavior data and identify regular visitor behavior patterns. By learning from historical data, we can quickly identify abnormal behaviors, such as a visitor frequently entering sensitive areas or staying too long, and generate alerts in a timely manner.

[0036] like Figure 4 As shown, in this embodiment, in step S7, based on the residence time anomaly detection model, the entry frequency anomaly detection model, the overall behavior anomaly index model and the alarm trigger condition model, the behavior data analysis and anomaly detection are performed; Set the following model parameters: For visitors at time The position coordinates ( ); For visitors in the area Duration of stay within Entrance area for visitors frequency; The visitor's behavior status is combined with location, speed, and regional attributes; is the abnormality index, which reflects the abnormality of visitor behavior; and is the mean and standard deviation of the regular behavior pattern, which is used to measure the difference between normal behavior and abnormal behavior.

[0037] Furthermore, in step S7: The dwell time anomaly detection model is as follows: ; in, Indicates that the visitor is in a certain area The residence time within and Respectively represent the mean and standard deviation of the normal residence time in the area; This model is used to measure whether the length of time a visitor spends in a certain area is abnormal; The entry frequency anomaly detection model is as follows: ; This model represents the degree to which the frequency of visitors entering a certain area deviates from the normal pattern. Enter an area for visitors The frequency, and are the mean and standard deviation of the frequency of regular visitors in the area; The overall behavior abnormality index model is as follows: ; in, is the current visitor's location, is the center point of the safe area. represents the typical distance distribution within the safe area, is the weight coefficient of each factor; The model combines multiple behavioral factors to calculate the visitor’s overall anomaly index; The alarm trigger condition model is as follows: ; in, For the set threshold, when the visitor's behavior index is abnormal Exceeding the threshold , the alarm will be triggered automatically.

[0038] Based on the above three models, when using: Data collection and behavior modeling: First, the system uses historical behavior data to calculate the time visitors spend in each area. and entry frequency Perform statistics to calculate the normal behavior pattern of each area, that is, the mean and standard deviation This step is the basis of anomaly detection and ensures that the model can distinguish between regular and abnormal behavior.

[0039] Real-time behavior monitoring and anomaly scoring: In actual operation, the system collects visitor behavior data in real time and first uses the residence time anomaly detection model and the entry frequency anomaly detection model , calculates the degree to which a visitor's behavior deviates in a specific area. If a visitor's dwell time and frequency of entry are outside the normal range, these formulas generate a high anomaly score.

[0040] Comprehensive abnormality index calculation: Comprehensive visitor's stay time, entry frequency, and current location information, using the overall behavior abnormality index model , the system can dynamically calculate the comprehensive abnormal index of visitor behavior. By adding the location deviation item , the system can detect whether visitors are in sensitive areas or have entered unauthorized spaces.

[0041] Alarm trigger mechanism: When the abnormal index Exceeding the set threshold , the system immediately triggers an alarm and notifies security personnel to intervene. For example, if a visitor enters sensitive areas such as the computer room or monitoring room multiple times, or stays in certain key areas for too long, the system will automatically determine these abnormal behaviors based on the formula and issue an early warning.

[0042] System self-learning and adjustment: the system will be updated regularly and The system continuously learns and adjusts the value of the visitor based on new data. This enables the system to dynamically adapt to different visitor behavior patterns and continuously improve the accuracy of anomaly detection. By analyzing visitor behavior in different time periods and scenarios, the system can gradually optimize the weights. , thereby more effectively balancing the influence of various factors.

[0043] S8. Visitor Behavior Modeling Based on historical data, the visitor behavior patterns are modeled to identify different types of visitor behaviors. Using cluster analysis methods, visitors are classified according to their behavior characteristics to form different behavior labels, such as regular visitors and risky visitors.

[0044] The behavior characteristics of visitors are classified by cluster analysis method to identify different types of visitor behaviors. Table 1 is a classification table of visitor behavior characteristics.

[0045] Table 1 This table helps to clarify the behavioral characteristics of visitors, making subsequent clustering analysis more precise and thus generating accurate behavioral labels.

[0046] By analyzing the characteristics in the table, visitors can be classified in detail, providing more targeted support for subsequent security monitoring and visitor management strategies. For example, visitors who stay in sensitive areas for a long time can be marked as "risky visitors", while visitors who frequently enter public areas and stay for a short time can be marked as "regular visitors". These labels will help security personnel quickly identify potential security threats in real-time monitoring.

[0047] In this embodiment, in step S8, when modeling, the following model parameters are set: For the The behavioral feature vector of each visitor includes multiple dimensions, such as residence time, visit frequency, and path complexity; For the The center point of the class group, i.e., the centroid, represents the cluster center; For Visitors With cluster center The distance between is the total number of clusters, indicating the number of visitor behavior patterns; is the weight vector, defining the importance of each feature in cluster analysis; and For the The mean and standard deviation of the class are used for subsequent anomaly detection and label generation, respectively; like Figure 3 As shown, based on the above model parameters, a distance calculation formula, a clustering update formula, and a behavior label generation formula are constructed.

[0048] Furthermore, the distance calculation formula is as follows: ; The formula calculates the visitor With cluster center The distance is the number of feature dimensions; by introducing the weight vector , adjust the distance calculation according to the importance of different features; The cluster update formula is as follows: ; in, Indicates that it belongs to Number of visitors to the class; This formula is used to update the cluster centers to ensure that the cluster centers reflect the characteristics of the current cluster groups; The behavior label generation formula is as follows: ; in, and It is the set thresholds that define labels for regular, risky, and abnormal visitors.

[0049] Based on the above three formulas, when using: Data collection and feature extraction: At the beginning, the system needs to collect visitor behavior data and construct a feature vector for each visitor. These features can include dwell time, visit frequency, path complexity, type of area entered, etc. These features are standardized to ensure that the impact of each feature on distance calculation is relatively balanced in subsequent analysis.

[0050] Distance calculation and preliminary clustering: using the distance calculation formula , calculate each visitor To different cluster centers In the initial stage, cluster centers can be randomly selected, or the intelligent initialization method of K-means algorithm can be used. By calculating the distance, the system assigns each visitor to the nearest cluster center.

[0051] Cluster center update: using cluster update formula , recalculate the center of each cluster according to the allocation results. This step is the core of cluster analysis. Through iterative calculation, it gradually converges to a stable cluster center to ensure that each type of behavior pattern can accurately reflect the characteristics of the visitor.

[0052] Behavior tag generation: Once the cluster center is stable, the system uses the behavior tag generation formula to generate a behavior tag for each visitor. Assign labels. By setting appropriate thresholds and , the system can automatically identify regular visitors, risky visitors, and abnormal visitors. This process ensures the interpretability and actionability of visitor behavior patterns.

[0053] Dynamic update and self-learning: In actual applications, visitor behavior is dynamic, and the system should regularly update feature vectors and cluster centers. Use new data to recalculate distances and clusters to ensure that the model always reflects the trend of current visitor behavior. At the same time, the system can also automatically adjust the threshold and , optimize the definition of behavior labels according to actual conditions.

[0054] Data support and decision-making: Through in-depth analysis of visitor behavior patterns, management can formulate more accurate security management strategies based on visitor classification. For example, for risky visitors, it may be necessary to increase monitoring or issue instant alerts when they enter sensitive areas to improve overall security protection capabilities.

[0055] Combining the multi-dimensional behavior characteristics of visitors with distance calculation, and through adaptive clustering algorithms and label generation mechanisms, we can achieve refined management of visitor behavior. It can not only effectively identify different types of visitor behavior, but also provide important data support for subsequent security monitoring and management decisions, thus improving the overall security management efficiency.

[0056] S9. Visual dashboard design and implementation Design a visual dashboard to display real-time data and behavior analysis results of visitors; the dashboard includes information on visitor flow, visitor activity trajectory, abnormal alarms, and provides intuitive charts combined with 3D maps.

[0057] In this embodiment, data visualization is used to help security personnel quickly understand the current visitor situation and take corresponding countermeasures in a timely manner to improve the overall security level.

[0058] The beneficial effects of the present invention are: This visitor behavior analysis method based on 3D maps achieves refined management and anomaly detection of visitor activities through real-time data collection, intelligent algorithms and behavior pattern modeling. This method can not only track the movement of visitors in the office in real time, identify and respond to potential security risks in a timely manner, but also classify visitors according to their behavioral characteristics through cluster analysis, providing strong data support for security monitoring. Ultimately, it improves the safety protection level of power grid companies, effectively reduces safety hazards caused by improper visitor management, and ensures the security of internal facilities and data.

[0059] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0060] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A visitor behavior analysis method based on 3D maps, characterized in that: The following steps are involved: S1. System requirements analysis and planning; Conduct a comprehensive analysis of the needs for visitor behavior management, and through communication with various departments, understand the specific needs and concerns of different positions for visitor management, including visitor identity authentication, behavior monitoring, and data statistics functions; S2, 3D map modeling and optimization; Establish an accurate 3D map of the office space. During the modeling process, high-precision laser scanning and photogrammetry technology are used to generate a 3D model of the building and mark important areas such as the computer room, monitoring room and office; S3,Design of visitor identity authentication mechanism; Design multi-factor authentication mechanisms to increase the security of visitor entry; S4. Generation and management of visitor passes; Automatically generate active tag visitor pass based on visitor identity information; S5. Setting up electronic fences in key areas; Set up electronic fences in the 3D map to identify important production sites, offices and other sensitive areas; S6. Real-time collection and processing of behavior data; After a visitor enters the office building, the visitor's activity trajectory is tracked in real time through positioning base stations and active tags. During the data collection process, the visitor's real-time location is associated with the 3D map, and its travel path is recorded. The collected location information is cleaned and corrected through data processing algorithms to ensure the accuracy of the data; S7, behavior data analysis and anomaly detection; Use data mining and machine learning algorithms to conduct in-depth analysis of visitor behavior data and identify common visitor behavior patterns; by learning from historical data, quickly identify abnormal behaviors, such as a visitor frequently entering sensitive areas or staying too long, and generate alerts in a timely manner; S8, visitor behavior pattern modeling; Based on historical data, the visitor behavior patterns are modeled to identify different types of visitor behaviors. Cluster analysis methods are used to classify visitors according to their behavior characteristics to form different behavior labels, such as regular visitors and risky visitors. S9. Visual dashboard design and implementation; Design a visual dashboard to display real-time data and behavior analysis results of visitors; the dashboard includes information on visitor flow, visitor activity trajectory, abnormal alarms, and provides intuitive charts combined with 3D maps.

2. A visitor behavior analysis method based on 3D map according to claim 1, characterized in that: In step S2, the model's textures and materials are optimized to improve the model's interactivity and visual appeal in practical applications, and the map is optimized to ensure smooth operation on different devices, providing basic support for subsequent visitor behavior analysis.

3. The visitor behavior analysis method based on 3D map according to claim 1, characterized in that: In step S3, the visitor provides a valid certificate when registering, scans the certificate information through OCR technology, and quickly checks the blacklist information in the database; Biometric identification based on face recognition or fingerprint recognition ensures the uniqueness and accuracy of visitor identity.

4. The visitor behavior analysis method based on 3D map according to claim 1, characterized in that: In step S4, the pass contains not only the basic information of the visitor, but also an embedded wireless communication module for real-time tracking of the visitor's movements; The pass also has a dynamic generation function, which updates information in real time based on the visitor's behavior and automatically cancels the card when it expires to prevent security risks caused by abuse or loss.

5. The visitor behavior analysis method based on 3D map according to claim 1, characterized in that: In step S5, the setting of the electronic fence is based on actual security requirements, taking into account the access rights and monitoring requirements of different areas; When a visitor enters a critical area set by the electronic fence, an alarm is automatically sent to notify security personnel so that they can intervene in time to prevent unauthorized persons from entering sensitive areas and enhance security protection capabilities.

6. The visitor behavior analysis method based on 3D map according to claim 1, characterized in that: In step S6, a visitor behavior real-time tracking model is built in, and the activity trajectory of the visitor is tracked in real time based on the model, including position calculation, path length calculation, distance calculation and state determination; The position calculation is as follows: ； in, Indicates the position information of the last moment, and updates the current position information in combination with the current speed and time interval; The path length calculation is as follows: ； Used to calculate the current path length as the path length at the previous moment plus the distance between the current position and the previous position; The distance calculation is as follows: ； in, is the target location coordinate, which is used to calculate the distance between the visitor's current location and the target location; The status determination is as follows: ； Determine the visitor's activity status based on the visitor's speed, the distance between the current location and the target, and the status of the electronic fence; In the above formula, For visitors at time The position coordinates ( ),in is the plane coordinate, is the height coordinate; It is a function of the visitor's path, indicating the length of the path from the starting point to the current point; The distance between the visitor's current location and the target location; For visitors at time The speed is defined as ,in is the change in path length, is the time interval; The activity status of visitors is defined as different behavior patterns, such as walking, staying, and moving quickly; The status of the electronic fence, indicating whether the visitor is in the allowed area.

7. The visitor behavior analysis method based on 3D map according to claim 1, characterized in that: In step S7, based on the residence time anomaly detection model, the entry frequency anomaly detection model, the overall behavior anomaly index model and the alarm triggering condition model, behavior data analysis and anomaly detection are performed; Set the following model parameters: For visitors at time The position coordinates ( ); For visitors in the area Duration of stay within Entrance area for visitors frequency; The visitor's behavior status is combined with location, speed, and regional attributes; is the abnormality index, which reflects the abnormality of visitor behavior; and is the mean and standard deviation of the regular behavior pattern, which is used to measure the difference between normal behavior and abnormal behavior.

8. The visitor behavior analysis method based on 3D map according to claim 7, characterized in that: In step S7: The dwell time anomaly detection model is as follows: ； in, Indicates that the visitor is in a certain area The residence time within and Respectively represent the mean and standard deviation of the normal residence time in the area; This model is used to measure whether the length of time a visitor spends in a certain area is abnormal; The entry frequency anomaly detection model is as follows: ； This model represents the degree to which the frequency of visitors entering a certain area deviates from the normal pattern. Enter an area for visitors The frequency, and are the mean and standard deviation of the frequency of regular visitors in the area; The overall behavior anomaly index model is as follows: ； in, is the current visitor's location, is the center point of the safe area. represents the typical distance distribution within the safe area, is the weight coefficient of each factor; The model combines multiple behavioral factors to calculate the visitor’s overall anomaly index; The alarm trigger condition model is as follows: ； in, For the set threshold, when the visitor's behavior index is abnormal Exceeding the threshold , the alarm will be triggered automatically.

9. The visitor behavior analysis method based on 3D map according to claim 1, characterized in that: In step S8, when modeling, the following model parameters are set: For the The behavioral feature vector of each visitor includes multiple dimensions, such as residence time, visit frequency, and path complexity; For the The center point of the class group, i.e., the centroid, represents the cluster center; For Visitors With cluster center The distance between is the total number of clusters, indicating the number of visitor behavior patterns; is the weight vector, defining the importance of each feature in cluster analysis; and For the The mean and standard deviation of the class are used for subsequent anomaly detection and label generation, respectively; Based on the above model parameters, the distance calculation formula, clustering update formula and behavior label generation formula are constructed.

10. The visitor behavior analysis method based on 3D map according to claim 9, characterized in that: The distance calculation formula is as follows: ； The formula calculates the visitor With cluster center The distance is the number of feature dimensions; By introducing the weight vector , adjust the distance calculation according to the importance of different features; The cluster update formula is as follows: ； in, Indicates that it belongs to Number of visitors to the class; This formula is used to update the cluster centers to ensure that the cluster centers reflect the characteristics of the current cluster groups; The behavior label generation formula is as follows: ； in, and It is the set thresholds that define labels for regular, risky, and abnormal visitors.

Images