Behavior intervention system for traffic safety education of minors

Abstract

The application discloses a behavior intervention system for juvenile traffic safety education and relates to the technical field of traffic safety education, and comprises the following steps: generating a multi-dimensional behavior DNA image based on operation data of juveniles in a virtual training scene; matching user resident geographic location information according to high-risk scene features in the multi-dimensional behavior DNA image and automatically labeling spatial anchor points in a real road map; and completing a micro intervention task by a user through a gesture action, instantly comparing the micro intervention task with historical data of the multi-dimensional behavior DNA image, and recording a safety behavior digital certificate of the micro intervention task; and adding the safety behavior digital certificate to a regional safety behavior integral pool and setting a family reward mechanism. The application constructs a multi-dimensional behavior DNA image by using micro behavior data such as a dangerous identification time difference and a decision accuracy rate collected in a virtual training scene, and realizes accurate characterization of individual risk features.

Description

Technical Field

[0001] This invention relates to the field of traffic safety education technology, and in particular to behavioral intervention systems for traffic safety education of minors. Background Technology

[0002] With the widespread adoption of smart devices and the development of augmented reality technology, traffic safety education for minors is gradually shifting from traditional classroom instruction to digital immersive training. Early education models rely on static text and image textbooks and video cases, making it difficult to establish real-world scenario cognition. While driving simulators that have emerged in recent years can reproduce typical traffic situations, they suffer from issues such as scenario rigidity and delayed feedback. Although AR navigation applications that have emerged in recent years attempt to combine real-time traffic conditions for risk warnings, their intervention strategies are still based on a general rule base and lack dynamic adaptation to individual behavioral characteristics.

[0003] Existing technological solutions have significant shortcomings in the field of traffic safety education for minors: risk assessment systems based on population statistical models struggle to accurately characterize individual behavioral traits, resulting in insufficient specificity in risk warnings. Traditional intervention methods employ fixed teaching content, failing to dynamically adjust educational strategies based on real-time user operation data, leading to low efficiency in the cognitive-behavioral conversion process. Particularly noteworthy is the failure of existing systems to establish a closed-loop feedback mechanism from virtual training to real-world scenarios, making it difficult to translate educational outcomes into sustained behavioral improvement. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a behavioral intervention system for traffic safety education for minors, which solves the problem in the prior art that traffic safety education for minors lacks personalized real-time intervention driven by dynamic behavioral characteristics.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: This invention provides a behavioral intervention system for traffic safety education of minors, comprising: a profile generation module, which generates a multi-dimensional behavioral DNA profile based on the minor's operational data in a virtual training scenario; an anchor point labeling module, which automatically labels spatial anchor points on a real road map based on the high-risk scenario characteristics in the multi-dimensional behavioral DNA profile and the user's permanent geographical location information; an AR display module, which captures real-time road images and overlays AR visual labels matching the current risk mode when the user's mobile terminal meets the spatial anchor point triggering conditions; a certificate recording module, which records digital certificates of safe behavior for micro-intervention tasks by instantly comparing the micro-intervention tasks with historical data of the multi-dimensional behavioral DNA profile after the user completes micro-intervention tasks through gestures; and a points reward module, which accumulates the digital certificates of safe behavior into a regional safe behavior points pool and sets up a family reward mechanism.

[0007] As a preferred embodiment of the behavioral intervention system for traffic safety education of minors described in this invention, the virtual training scenario constructs a highly realistic road environment using Unreal Engine, and the user completes virtual intervention tasks through first-person interaction; the operation data includes hazard identification time difference, decision accuracy rate, and virtual behavior execution trajectory.

[0008] As a preferred embodiment of the behavioral intervention system for traffic safety education of minors described in this invention, the specific steps for generating a multi-dimensional behavioral DNA profile are as follows: The operational data is used as the basic feature and then normalized. Based on the normalized basic features, risk preference index, scenario adaptability index and behavioral stability coefficient are calculated and integrated to form a multi-dimensional behavioral DNA profile.

[0009] As a preferred embodiment of the behavioral intervention system for traffic safety education of minors described in this invention, the high-risk scenario characteristics refer to a set of quantitative indicators in the operational data that reflect the risk of user behavior; the set of quantitative indicators includes behavioral frequency indicators in road environments, behavioral severity indicators in traffic situations, and cross-scenario correlation characteristic indicators.

[0010] As a preferred embodiment of the behavioral intervention system for traffic safety education of minors described in this invention, the user's permanent geographical location information includes road type, distribution of traffic facilities, and accident-prone road section markings; The specific steps for automatically marking spatial anchor points on a real road map are as follows. Filter out real road areas from the user's frequently visited geographical location information that match the set of quantitative indicators; Extract road elements associated with high-risk scene features within the real road area, and determine the spatial coordinates and coverage of each road element; Mark the spatial coordinates and coverage area of ​​each road element in the real road map as spatial anchor points.

[0011] As a preferred embodiment of the behavioral intervention system for traffic safety education of minors described in this invention, the spatial anchor point triggering condition refers to the mobile terminal's real-time positioning coordinates entering the coverage area of ​​the spatial anchor point, and the user's movement speed being lower than the preset safety intervention speed threshold range. The AR visualization tags matching the current risk pattern are generated by retrieving AR tag content corresponding to high-risk scene features in the multi-dimensional behavioral DNA profile and associating it with the spatial coordinates and coverage of each road element at the current spatial anchor point.

[0012] As a preferred embodiment of the behavioral intervention system for traffic safety education for minors described in this invention, the micro-intervention task refers to a short behavioral correction exercise corresponding to the characteristics of a high-risk scenario, which is completed by the user through gestures on a real road map.

[0013] As a preferred embodiment of the behavioral intervention system for traffic safety education of minors described in this invention, the specific steps for recording the digital credentials for safe behaviors in micro-intervention tasks are as follows: Capture user gestures as they complete micro-intervention tasks, and identify the accuracy and completion of the gestures. The accuracy and completion rate are dynamically time-normalized and matched with historical data in the multidimensional behavioral DNA profile to generate digital credentials of safe behavior that include task type, completion time and achievement status.

[0014] As a preferred embodiment of the behavioral intervention system for traffic safety education of minors described in this invention, the regional safety behavior points pool refers to a set of regional safety behavior data formed by summarizing the digital credentials of safety behaviors of all users within the region to which the user's permanent geographical location information belongs, and calculating them according to a preset points accumulation rule.

[0015] As a preferred embodiment of the behavioral intervention system for traffic safety education of minors described in this invention, the specific steps for establishing a family reward mechanism are as follows: Link user family accounts with regional security behavior points pools and set points redemption rules; The system updates family points in real time and synchronizes them to the family account. When the family account points reach the preset redemption threshold, the reward distribution process is triggered.

[0016] The beneficial effects of this invention are as follows: It constructs a multi-dimensional behavioral DNA profile based on micro-behavioral data such as hazard identification time difference and decision accuracy collected in virtual training scenarios, achieving precise characterization of individual risk features. Through this multi-dimensional behavioral DNA profile, spatial anchor points matching high-risk scenarios are automatically marked on real road maps. When a user enters an anchor point area, personalized corrective prompts are dynamically presented using augmented reality, combined with gesture recognition to complete real-time behavioral intervention. This not only achieves a seamless transition from virtual training to real-world scenarios but also forms a continuously optimized safety education system through the accumulation of digital safety behavior credentials and regional point feedback. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A schematic diagram of a behavioral intervention system for traffic safety education for minors.

[0019] Figure 2 Flowchart for labeling spatial anchor points.

[0020] Figure 3 Flowchart for generating security behavior credentials.

[0021] Figure 4 Flowchart for rewarding safe behaviors with points. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0025] Reference Figures 1-4 As one embodiment of the present invention, this embodiment provides a behavioral intervention system for traffic safety education of minors, including the following modules: The profile generation module generates a multi-dimensional behavioral DNA profile based on the operational data of minors in virtual training scenarios.

[0026] The virtual training scenario constructs a highly realistic road environment using Unreal Engine, allowing users to complete virtual intervention tasks through first-person interaction; operational data includes hazard identification time difference, decision accuracy, and virtual behavior execution trajectory.

[0027] Specifically, launch the Unreal Engine editor, create a new project, select the "Game" template, and enable multi-platform support (including PC, mobile devices, and VR headsets). Note that because cross-platform operation is required, enable the "Mobile Preview" and "VR Mode" options when configuring project settings to ensure the scene can adapt its rendering accuracy across different hardware.

[0028] Import or create road models (such as city intersections, zebra crossings, and traffic lights) using Unreal Engine's Asset Library. Apply physically based rendering (PBR) to simulate materials and lighting, assigning metallic and roughness parameters to road surfaces; simulate asphalt materials, and add high dynamic range (HDR) lighting to vehicles and NPCs (non-player characters) to achieve realistic shadows and reflections.

[0029] The Chaos physics system, powered by Unreal Engine, simulates vehicle motion, collisions, and destruction effects. It adds physical collider and rigid body properties to each dynamic object (such as a car or bicycle), setting mass, friction, and gravity parameters. Notably, because the Chaos physics system supports real-time physics calculations, it generates realistic accident animations when NPCs violate traffic rules (such as running a red light) by calculating the consequences of collisions through continuous collision detection (CCD).

[0030] Furthermore, use Blueprint visual programming to define user interaction flows (such as first-person perspective operation and dialogue intervention). Create an Event Graph to define logic: add "keyboard input" or "VR controller action" events to trigger user responses; connect "conditional branch" nodes to handle NPC behavior (such as the NPC stopping or continuing the violation after the user inputs traffic safety regulations text).

[0031] Operational data is used as the basic feature and normalized. Based on the normalized basic feature, risk preference index, scenario adaptability index and behavioral stability coefficient are calculated and integrated to form a multi-dimensional behavioral DNA profile.

[0032] Specifically, a min-max normalization method is applied to each basic feature to scale the values ​​to... Range. Calculate the global minimum and global maximum values ​​for each basic feature (based on historical datasets); Based on the normalized hazard identification time difference and decision accuracy, a weighted average method is used to calculate the risk preference index. The risk preference index reflects the user's tendency to choose high-risk behaviors. Based on the normalized decision accuracy and virtual behavior execution trajectory, a feature fusion method is used to calculate the scenario fitness index. The scenario fitness index measures the user's performance adaptability in different traffic scenarios.

[0033] Based on the normalized time difference for hazard identification and the virtual behavior execution trajectory, a behavior stability coefficient is calculated using variance. The behavior stability coefficient assesses the consistency of user behavior. A higher behavior stability coefficient indicates stronger stability. Risk preference index, scenario adaptability index, and behavior stability coefficient are integrated into a multi-dimensional behavioral DNA profile using vector representation, serving as three dimensions.

[0034] The anchor point annotation module automatically annotates spatial anchor points on real road maps by matching high-risk scenario characteristics from the multi-dimensional behavioral DNA profile with the user's frequently used geographical location information.

[0035] It should be noted that high-risk scenario characteristics refer to a set of quantitative indicators in the operational data that reflect the risk of user behavior. This set of quantitative indicators includes behavioral frequency indicators in road environments (reflecting the frequency of user errors in road environments), behavioral severity indicators in traffic scenarios (reflecting the severity of the consequences of errors in traffic scenarios), and cross-scenario correlation characteristic indicators (reflecting the risk correlation between different scenarios). User's permanent geographic location information includes road type, distribution of traffic facilities, and accident-prone road section markers, obtained through the mobile terminal's GPS positioning function combined with electronic map API services.

[0036] Filter out real road areas from the user's frequently visited geographical location information that match the set of quantitative indicators.

[0037] Specifically, a mapping rule is established between a set of quantitative indicators and geographic location information. Since a high behavior frequency indicator indicates that users often make mistakes on specific road types, accident-prone road sections are matched; a high behavior severity indicator indicates that the consequences of the mistake are serious, so areas with dense traffic facilities are matched; and a high cross-scenario correlation feature indicator indicates that the risk crosses scenarios, so areas with complex road types are matched.

[0038] The rules are preset as follows: if the behavior frequency index value is greater than the high-frequency risk matching threshold (preset 0.7), then the accident-prone road sections will be matched first; if the behavior severity index value is greater than the severe risk matching threshold (preset 0.6), then the traffic facility distribution area will be matched first; if the cross-scenario association feature index value is greater than the cross-scenario risk matching threshold (preset 0.5), then the area with diverse road types will be matched.

[0039] It should be noted that the high-frequency risk matching threshold, the severe risk matching threshold, and the cross-scenario risk matching threshold are all based on the statistical distribution characteristics of historical training data. By analyzing the behavioral data of a large number of minors in virtual training scenarios, the statistical quantiles of the behavior frequency index, behavior severity index, and cross-scenario correlation characteristic index at the corresponding risk level are selected as the threshold benchmark values, and then determined after verification and adjustment in actual scenarios.

[0040] Use SQL queries to filter real road areas based on user's frequent geographic location information and a set of quantitative indicators. The query conditions are based on mapping rules: for example, `SELECT * FROM geographic location information table WHERE accident-prone road segment identifier = 'High' AND road type IN ('urban road', 'rural road')`. Since the quantitative indicator set values ​​are normalized, numerical comparisons are performed directly. The filtered results are output as a list of real road areas (e.g., specific road segment names or IDs).

[0041] Extract road elements associated with high-risk scene features within the real road area, and determine the spatial coordinates and coverage of each road element.

[0042] Specifically, road element types are determined based on a set of quantitative indicators. Since behavior frequency indicators are associated with high-frequency risk points, elements corresponding to accident-prone road sections (such as intersections) are extracted; behavior severity indicators are associated with severe risk points, so elements corresponding to traffic facility distribution (such as traffic lights) are extracted; cross-scenario association feature indicators are associated with complex areas, so elements corresponding to road types (such as multi-lane intersections) are extracted. A GIS database query is used, for example, by calling API functions to obtain the element list.

[0043] For each identified road element, a GPS coordinate extraction method is used. A map API (such as the Google Geocoding API) is called, inputting the element name (e.g., "XX Road Traffic Light"), and the system returns latitude and longitude coordinates. Since the coordinates are standard geographic data, they are stored in (latitude, longitude) format.

[0044] The precise latitude and longitude coordinates of road elements are obtained by calling standard geographic information system APIs (such as Google Maps API). Based on the element type (intersection, traffic light, etc.) and associated risk characteristics (such as inner wheel difference risk), circular or rectangular coverage areas are defined respectively. The size of the area is determined according to the simulated impact distance of various risk behaviors in the virtual training scenario, and the specific boundary range is calculated using basic geometric methods.

[0045] Mark the spatial coordinates and coverage area of ​​each road element in the real road map as spatial anchor points.

[0046] Specifically, initialize the map view using a digital map service (such as Google Maps or Baidu Maps API), setting the center point to the user's usual geographic location. Simultaneously, for each road element, call the map API's marker function to add spatial anchor points. Input the spatial coordinates (latitude, longitude) and set the marker icon to a risk type icon (e.g., a red icon indicates high risk).

[0047] Based on coverage data, draw geometric shapes on the map. For example, if the coverage area is a circle with a radius of 50 meters, call the API's circle drawing function and input the coordinates and radius; if it is a rectangle, call the rectangle drawing function and input the boundary coordinates. The default fill color for the shapes is semi-transparent red to indicate risk areas.

[0048] Save the marker data (coordinates and coverage area) to a database (such as a local database on a mobile device or in the cloud) as spatial anchor points for output.

[0049] The AR display module captures real-time road images and overlays AR visual labels that match the current risk mode when the user's mobile terminal meets the spatial anchor trigger conditions.

[0050] The triggering condition for a spatial anchor point refers to the mobile terminal's real-time positioning coordinates entering the coverage area of ​​the spatial anchor point, and the user's movement speed being lower than the preset safety intervention speed threshold range.

[0051] Specifically, spatial anchor point data is read from the database, including the spatial coordinates (latitude and longitude) and coverage area (such as radius or boundary) of each road element. At the same time, standard APIs (such as Android's LocationManager or iOS's CoreLocation) are used to call the mobile terminal's GPS module to obtain real-time location coordinates (current latitude and longitude).

[0052] Calculate the distance between the real-time location coordinates and the spatial anchor point coordinates. Use the Haversine formula to calculate the distance between the two points (in meters). If the calculated distance is less than or equal to the coverage radius, the coordinates are included in the coverage area. Use the mobile terminal's GPS speed sensor or accelerometer to obtain the real-time movement speed (in meters per second or kilometers per hour). The preset safety intervention speed threshold range refers to the range of movement speeds allowed to trigger AR traffic safety education interventions. Specifically, it is based on the statistical average walking speed of minors and is set as a walking speed range (0 to 1.4 meters per second, corresponding to 0 to 5 kilometers per hour) to avoid distraction during high-speed movement.

[0053] If the location coordinates enter the coverage area and the movement speed is within the safe intervention speed threshold range, the spatial anchor point triggering condition is met. A trigger signal is output to initiate real-time road image capture. If the condition detection fails, no trigger is initiated, and the system returns to a waiting state.

[0054] The AR visualization tags for matching the current risk pattern are generated by retrieving AR tag content corresponding to the high-risk scene features in the multi-dimensional behavioral DNA profile and associating it with the spatial coordinates and coverage of each road element at the current spatial anchor point.

[0055] Specifically, once the spatial anchor trigger conditions are met, the mobile terminal's rear camera is activated to capture a real-time video stream. The camera is initialized using a standard API (such as Android's Camera2 or iOS's AVFoundation), and the API is called to capture the image. Retrieve multidimensional behavioral DNA profiles to extract features of high-risk scenarios. Based on the current spatial anchor point, extract associated road elements (such as intersections or traffic lights). Since the spatial anchor point contains road element types and coordinates, the element data is read directly. For example, if the spatial anchor point is an intersection, the road element is "intersection".

[0056] AR visual labels are generated based on the characteristics of high-risk scenarios and road elements. The AR visual labels are based on the "risk pattern matching" mentioned in the document: a high behavior frequency index indicates a high-frequency risk, generating a warning icon (such as an exclamation mark); a high behavior severity index indicates a serious risk, generating a text prompt (such as "Caution: Inner wheel difference risk"); a high cross-scenario association feature index indicates a complex risk, generating an animated prompt (such as a flashing arrow).

[0057] Use an AR framework (such as ARKit or ARCore) to overlay the generated label content onto the live camera feed. Call the API's rendering function: input the label content (icon, text) and spatial anchor coordinates, and set the label's position in the image (e.g., overlaying on road elements). The overlay process ensures the label dynamically adjusts as the camera moves, maintaining a stable position.

[0058] The credential recording module allows users to complete micro-intervention tasks through gestures, instantly comparing the micro-intervention tasks with historical data from multi-dimensional behavioral DNA profiles, and recording digital credentials of safe behaviors for the micro-intervention tasks.

[0059] It should be noted that micro-intervention tasks refer to brief behavioral correction exercises that users complete using gestures on real road maps, corresponding to the characteristics of high-risk scenarios (such as a stop gesture corresponding to the risk of running a red light in a high-risk scenario). The purpose is to capture the gesture images and quantify their quality.

[0060] Capture user gestures as they complete micro-intervention tasks, and identify the accuracy and completion of the gestures.

[0061] Specifically, the system calls the rear camera API of the mobile terminal to capture real-time video streams. Gesture templates are preset based on high-risk scenario characteristics. These characteristics include behavior frequency indicators, behavior severity indicators, and cross-scenario correlation feature indicators, each corresponding to a specific gesture: for example, when the behavior severity indicator is high, the gesture template is a "hand outstretched stop gesture"; when the behavior frequency indicator is high, the gesture template is a "waving warning gesture." The gesture templates are stored in a local database and contain the key point coordinate sequence of the gesture (such as finger positions).

[0062] Gesture recognition APIs (such as MediaPipe Hands) are used to process real-time video and detect key points of user gestures (such as wrist and fingertip coordinates). Gesture recognition accuracy is determined by comparing the differences in key point positions between the user's gesture and a preset template. Spatial distance measurement methods are used to evaluate the degree of matching between the two, quantifying the spatial position difference between the actual gesture and the standard gesture as an accuracy score.

[0063] Gesture completion is determined by analyzing the completeness of the gesture execution, primarily examining whether the duration of the gesture from start to finish meets the expected standard. The evaluation process is based on time series analysis, comparing the actual execution time with the standard duration.

[0064] The accuracy and completion rate are dynamically time-normalized and matched with historical data in the multidimensional behavioral DNA profile to generate digital credentials of safe behavior that include task type, completion time and achievement status.

[0065] Specifically, it receives accuracy and completion data to form the current time series. Historical data sequences (accuracy and completion values ​​of past gestures, stored as a time series array) are extracted from the multidimensional behavioral DNA profile. Since the multidimensional behavioral DNA profile contains historical operation data, relevant sequences are retrieved from the database.

[0066] The gesture sequence matching process employs a dynamic time warping algorithm. This algorithm effectively addresses the challenge of matching gestures with varying execution speeds and durations by calculating the minimum cumulative difference between the current and historical gesture sequences. After obtaining the sequence difference calculation result, it is compared with a preset matching benchmark value. If the sequence difference calculation result does not exceed the preset matching benchmark value, the gesture matching is considered successful; otherwise, the gesture matching is considered unsuccessful.

[0067] It should be noted that the preset matching baseline value is a quantile value calculated by analyzing the difference distribution characteristics of successfully matched samples in the historical gesture dataset and using statistical methods, which ensures that the critical state of matching success and failure can be effectively distinguished.

[0068] The digital credential for safe behavior contains three elements: task type, completion time, and gesture matching status. The task type is obtained from the micro-intervention task definition (e.g., "stop gesture practice"); the completion time is recorded using the mobile terminal system timestamp; and the gesture matching status is determined from the evaluation conclusion output by the gesture matching algorithm, marked as "gesture matching successful" or "gesture matching failed".

[0069] The points reward module accumulates digital credentials for safe behaviors into a regional safe behavior points pool, and sets up a family reward mechanism.

[0070] The regional security behavior points pool refers to a collection of digital credentials for the security behaviors of all users within the region to which a user's permanent geographical location information is located, calculated according to a preset points accumulation rule, forming a regional security behavior data set.

[0071] Specifically, the preset points accumulation rules use an accumulation method: for digital credentials that successfully match gestures, 1 point is assigned; for digital credentials that fail to match gestures, no point is assigned; the total points for the region are the sum of the points for all user credentials.

[0072] The regional security behavior dataset includes regional name, total points, number of credentials, etc. Create a database table to store the dataset: fields include regional name, total points, and last update time. Use INSERT or UPDATE statements to write the data.

[0073] Link user family accounts with regional security behavior points pools and set points redemption rules.

[0074] Specifically, a user's family account is defined as a unique identifier for each user's family (such as a family ID). The user's family account is read from the user profile, for example, from the family ID field in the database. An SQL statement is used to establish the association; after association, each family account is mapped to the total points in the regional security behavior points pool.

[0075] The points redemption system employs a tiered reward structure, setting different redemption levels based on the accumulated points. For example, 10 points can be redeemed for a basic reward (such as a safety manual), while 50 points can be redeemed for a premium reward (such as a VR device trial voucher). The redemption rules are stored in a database table, with fields including redemption thresholds and reward types.

[0076] The system updates family points in real time and synchronizes them to the family account. When the family account points reach the preset redemption threshold, the reward distribution process is triggered.

[0077] Family points are displayed in real-time on the family account interface. The total points are read from the regional security behavior points pool using an API. Synchronization process: When points change (e.g., a new credential is added), a database event listener is triggered, updating the family account display interface with the latest points value.

[0078] The preset redemption threshold is obtained from the redemption rules (e.g., 10 or 50). The current points of the family account are compared with the redemption threshold. When the points value is detected to meet or exceed the redemption standard, the reward distribution process is immediately initiated. The reward items or benefits are automatically allocated by calling the predefined reward distribution interface, and a redemption success notification is sent to the family account at the same time.

[0079] In summary, this invention constructs a multi-dimensional behavioral DNA profile based on micro-behavioral data such as hazard identification time difference and decision accuracy collected in virtual training scenarios, achieving precise characterization of individual risk features. By automatically marking spatial anchor points matching high-risk scenarios on real road maps using this multi-dimensional behavioral DNA profile, personalized corrective prompts are dynamically presented using augmented reality when a user enters an anchor point area. Combined with gesture recognition, real-time behavioral intervention is achieved. This not only achieves a seamless transition from virtual training to real-world scenarios but also forms a continuously optimized safety education system through the accumulation of digital safety behavior credentials and regional point feedback.

[0080] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A behavioral intervention system for traffic safety education of minors, characterized in that: include, The profile generation module generates a multi-dimensional behavioral DNA profile based on the operation data of minors in virtual training scenarios; The anchor point annotation module automatically annotates spatial anchor points on real road maps by matching the high-risk scenario characteristics in the multi-dimensional behavioral DNA profile with the user's permanent geographical location information. The AR display module captures real-time road images and overlays AR visual labels that match the current risk mode when the user's mobile terminal meets the spatial anchor trigger conditions. The credential recording module allows users to complete micro-intervention tasks through gestures, instantly comparing the micro-intervention tasks with historical data from the multi-dimensional behavioral DNA profile, and recording digital credentials of the safe behavior of the micro-intervention tasks. The points reward module accumulates digital credentials for safe behaviors into a regional safe behavior points pool, and sets up a family reward mechanism.

2. The behavioral intervention system for traffic safety education of minors as described in claim 1, characterized in that: The virtual training scenario uses Unreal Engine to construct a highly realistic road environment, and users complete virtual intervention tasks through first-person interaction; the operation data includes hazard identification time difference, decision accuracy, and virtual behavior execution trajectory.

3. The behavioral intervention system for traffic safety education of minors as described in claim 2, characterized in that: The specific steps for generating a multidimensional behavioral DNA profile are as follows: The operational data is used as the basic feature and then normalized. Based on the normalized basic features, risk preference index, scenario adaptability index and behavioral stability coefficient are calculated and integrated to form a multi-dimensional behavioral DNA profile.

4. The behavioral intervention system for traffic safety education of minors as described in claim 3, characterized in that: The high-risk scenario characteristics refer to a set of quantitative indicators in the operational data that reflect the risk of user behavior; the set of quantitative indicators includes behavior frequency indicators in road environments, behavior severity indicators in traffic situations, and cross-scenario correlation characteristic indicators.

5. The behavioral intervention system for traffic safety education of minors as described in claim 4, characterized in that: The user's permanent location information includes road type, distribution of traffic facilities, and signs of accident-prone road sections; The specific steps for automatically marking spatial anchor points on a real road map are as follows. Filter out real road areas from the user's frequently visited geographical location information that match the set of quantitative indicators; Extract road elements associated with high-risk scene features within the real road area, and determine the spatial coordinates and coverage of each road element; Mark the spatial coordinates and coverage area of ​​each road element in the real road map as spatial anchor points.

6. The behavioral intervention system for traffic safety education of minors as described in claim 5, characterized in that: The spatial anchor point triggering condition refers to the mobile terminal's real-time positioning coordinates entering the coverage area of ​​the spatial anchor point, and the user's movement speed being lower than the preset safety intervention speed threshold range. The AR visualization tags matching the current risk pattern are generated by retrieving AR tag content corresponding to high-risk scene features in the multi-dimensional behavioral DNA profile and associating it with the spatial coordinates and coverage of each road element at the current spatial anchor point.

7. The behavioral intervention system for traffic safety education of minors as described in claim 6, characterized in that: The micro-intervention task refers to a brief behavioral correction exercise that users complete using gestures on a real road map, corresponding to the characteristics of high-risk scenarios.

8. The behavioral intervention system for traffic safety education of minors as described in claim 7, characterized in that: The specific steps for recording the security behavior digital credentials for micro-intervention tasks are as follows. Capture user gestures as they complete micro-intervention tasks, and identify the accuracy and completion of the gestures. The accuracy and completion rate are dynamically time-normalized and matched with historical data in the multidimensional behavioral DNA profile to generate digital credentials of safe behavior that include task type, completion time and achievement status.

9. The behavioral intervention system for traffic safety education of minors as described in claim 8, characterized in that: The regional security behavior points pool refers to a set of regional security behavior data formed by aggregating the digital credentials of all users within the region to which a user's permanent geographical location information belongs, and calculating them according to a preset points accumulation rule.

10. The behavioral intervention system for traffic safety education of minors as described in claim 9, characterized in that: The specific steps for setting up a family reward mechanism are as follows: Link user family accounts with regional security behavior points pools and set points redemption rules; The system updates family points in real time and synchronizes them to the family account. When the family account points reach the preset redemption threshold, the reward distribution process is triggered.

Images