A parking lot differentiated value guard method based on user portrait

Abstract

This invention relates to the field of parking lot monitoring systems, specifically to a differentiated parking lot monitoring method based on user profiles. By collecting behavioral data from each registered vehicle, a user behavior profile and credit scoring system are established. When the parking lot monitoring AI handles passage issues, differentiated decisions are made to improve the passage efficiency of high-frequency compliant users and reduce the overall manual intervention cost of the system. The interaction results are also updated to the user profile database to update the user credit score in real time, thereby improving the accuracy of the monitoring AI interaction.

Description

Technical Field

[0001] This invention relates to the field of parking lot monitoring systems, and more specifically to a differentiated parking lot monitoring method based on user profiles. Background Technology

[0002] A parking lot security system is an intelligent integrated management platform for managing and monitoring vehicles entering and exiting parking lots. Through the combination of hardware and software, it achieves 24 / 7 automated management of parking lots, reducing or even replacing the need for manual monitoring. The system typically consists of core components such as license plate recognition cameras, barriers, inductive loop detectors, booth displays, voice broadcasting equipment, and backend management software. When a vehicle enters the parking lot, the license plate recognition camera automatically captures and identifies the license plate number. The system compares this information with its database to determine if the vehicle is authorized. If it is a monthly pass or temporarily authorized vehicle, the barrier will automatically open to allow passage. If it is an unfamiliar vehicle, the system will automatically record the entry time and issue a temporary pass, while simultaneously collecting parking fees at the exit using a time-based billing method.

[0003] Existing unmanned parking lot systems frequently encounter issues such as license plate recognition failures and payment irregularities, causing vehicles to linger at exits and impacting traffic efficiency. For example, the utility model patent application CN202021032101.4, entitled "An AI-Powered Parking Lot Management Monitoring and Data Analysis System," utilizes an information storage module, an AI acquisition module, a central control module, and an execution module to achieve AI recognition functionality. However, it employs a uniform processing strategy for different users, failing to dynamically allocate resources based on user priority, resulting in overall low operational efficiency. Summary of the Invention

[0004] The purpose of this invention is to provide a parking lot differentiated duty method based on user profiles, which aims to improve the problem that the existing parking lot AI duty system adopts a uniform processing strategy for different users and cannot dynamically allocate according to user priority, resulting in low overall operational efficiency.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A parking lot differentiated monitoring method based on user profiles includes the following steps: S01. Collect behavioral data for each registered vehicle; S02. Set scoring rules and calculate user credit scores based on behavioral data; S03. Set credit scoring levels and divide users into n levels from trust to distrust based on their creditworthiness, and establish a user profile database. S04. Configure the monitoring AI and preset differentiated processing strategies for different groups at different levels; S05. If license plate recognition fails or the gate fails to rise, the monitoring AI is triggered. The monitoring AI uses the vehicle license plate number as an index to query the user level. S06. The AI ​​on duty selects and executes the corresponding strategy from the preset differentiated processing strategy according to the user level; S07. Update the user profile database with the results of this interaction.

[0006] Furthermore, step S03 also includes the following steps: S31. Obtain user credit scores and behavioral data. S32. If the credit score is greater than or equal to 90 and the number of entries and exits this month is greater than 20, then classify the user into the S-level user tier; otherwise, proceed to step S33. S33. Is the credit score between 80 and 89, or is the user a monthly card user with a credit score greater than or equal to 70? If so, classify the user as a Level A user; otherwise, proceed to step S34. S34. If the credit score is between 60 and 79 and the vehicle is a temporary vehicle with no abnormal records, then classify the user as a B-level user; otherwise, proceed to step S35. S35. Check if the credit score is between 40 and 59 or if there is a record of overdue payment or assistance. If so, classify the user as a C-level user; otherwise, proceed to step S36. S36. Users with a credit score of less than 40, who are blacklisted, or who have a history of disputes of 2 or more are classified as Level D users.

[0007] Furthermore, the behavioral data includes cumulative entry and exit data, historical assistance data, payment records, vehicle type, violation records, and dispute records; The cumulative inflow and outflow data includes inflow and outflow data for this month, inflow and outflow data for this year, and inflow and outflow data for historical years. The historical help data includes the number of help requests, the type of help request, and the resolution method; The payment records include the number of overdue payments and the on-time payment rate; The vehicle types include monthly pass vehicles, temporary vehicles, and special vehicles; The violation records include the number of times the gate was breached and blacklist markings; The dispute record includes the number of complaints and the results of their handling.

[0008] Furthermore, the setting of scoring rules includes the following steps: S21. Divide the scoring dimensions and set the dimensions of entry and exit frequency, historical assistance rate, payment compliance, violation record and usage duration; S22. Set weights: set the weight for entry / exit frequency to 20%; set the weight for historical assistance rate to 25%; set the weight for payment compliance to 30%; set the weight for violation records to 15%; and set the weight for usage duration to 10%. S23. Set scoring rules The scoring rule for the frequency of entry and exit is that a person who enters and exits 20 or more times this month will receive full marks, and a person who enters and exits less than 20 times will receive progressively lower scores. The scoring rules for the historical help-seeking rate dimension are as follows: a help-seeking rate of less than or equal to 1% receives full marks, a help-seeking rate of more than 10% receives 0 marks, and the help-seeking rate is calculated by decreasing the score from 1% to 10%. The scoring rules for payment compliance are: full marks for no outstanding payments, and 10 points are deducted for each outstanding payment, until the score reaches 0. The scoring rules for the violation record dimension are as follows: full marks are awarded for no violations, and 20 points are deducted for each violation, until the score reaches 0. The scoring criteria for the vehicle usage duration dimension is that full marks are awarded for registration and usage exceeding 6 months, with scores increasing from 0 to 6 months of registration duration. S24. Output the user's credit score according to the scoring rules.

[0009] Furthermore, the differentiated processing strategy includes, Credit-based clearance allows vehicles of S-level and A-level users to pass directly even if their license plates are occasionally obscured. The system records this information in the background, skips the secondary verification process after payment, and prioritizes their passage. Fast track reduces confirmation steps for S-level and A-level user vehicles, lowering the confidence threshold by 5%; payment anomalies are prioritized for remote verification. The standard procedure applies to vehicles belonging to Class B users, and the standard handling strategy is executed accordingly. Strengthen verification by raising the confidence threshold for C-level user vehicles, adding payment verification steps, and prioritizing manual processing. The system will automatically start recording audio for vehicles belonging to Class D users and simultaneously notify management personnel to prevent automatic passage by raising the barrier. Vehicles will be allowed to pass only after manual approval.

[0010] Furthermore, the credit release specifically includes the following steps: P01. The AI-controlled query user is categorized as either Level S or Level A; P02. Query the confidence level of vehicle identification this time. If the confidence level of S-level user license plate is greater than or equal to 60% and the confidence level of A-level user vehicle is greater than or equal to 70%, then proceed to step P03; otherwise, execute the standard processing strategy. P03. Query the number of times the user's credit has been granted in the past 7 days. If the number of times the credit has been granted to an S-level user is less than or equal to 3 times, and the number of times the credit has been granted to an A-level user is less than or equal to 1 time, then proceed to step P04; otherwise, execute the standard processing strategy. P04. Raise the barrier directly, generate a release record in the background, and push it to the management personnel simultaneously.

[0011] Furthermore, step S07 includes the following steps: When the AI-assisted solution is successfully resolved, the score for the historical help rate dimension increases by 0.5 points; When the AI ​​interaction fails and human intervention is required, the score for the historical help rate dimension will be reduced by 1 point. When the number of consecutive successful credit approvals reaches a preset value, the license plate confidence threshold is lowered.

[0012] Furthermore, in step S5, when license plate recognition fails and user information cannot be obtained, the current time and entrance / exit number are used as temporary indexes, and the standard processing strategy is executed.

[0013] By adopting the above technical solution, the present invention has the following advantages compared with the prior art: Behavioral data of each registered vehicle is collected to establish user behavior profiles and a credit scoring system. When parking lot AI handles passage issues, it enables differentiated decision-making, improves the passage efficiency of high-frequency compliant users, and reduces the overall cost of manual intervention in the system. The interaction results are also updated to the user profile database, and the user credit score is updated in real time to improve the accuracy of AI interaction. Attached Figure Description

[0014] Figure 1 This is a flowchart of the parking lot differentiated monitoring method based on user profiles as described in this invention. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0016] Additionally, it should be noted that the terms "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer" are all based on the orientation or positional relationship shown in the accompanying drawings. They are merely for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element of the present invention must have a specific orientation. Therefore, they should not be construed as limitations on the present invention.

[0017] When an element is referred to as being "fixed to," "set on," or "contained on" another element, it can be directly on or indirectly on that other element. When an element is referred to as being "connected to," it can be directly connected to or indirectly connected to that other element.

[0018] Unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication between two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances. Example

[0019] Please refer to Figure 1 As shown, this embodiment provides a parking lot differentiated monitoring method based on user profiles, including the following steps: S01. Collect behavioral data for each registered vehicle; S02. Set scoring rules and calculate user credit scores based on behavioral data; S03. Set credit scoring levels and divide users into n levels from trust to distrust based on their creditworthiness, and establish a user profile database. S04. Configure the monitoring AI and preset differentiated processing strategies for different groups at different levels; S05. If license plate recognition fails or the gate fails to rise, the monitoring AI is triggered. The monitoring AI uses the vehicle license plate number as an index to query the user level. S06. The AI ​​on duty selects and executes the corresponding strategy from the preset differentiated processing strategy according to the user level; S07. Update the user profile database with the results of this interaction.

[0020] Behavioral data of each registered vehicle is collected to establish user behavior profiles and a credit scoring system. When parking lot AI handles passage issues, it enables differentiated decision-making, improves the passage efficiency of high-frequency compliant users, and reduces the overall cost of manual intervention in the system. The interaction results are also updated to the user profile database, and the user credit score is updated in real time to improve the accuracy of AI interaction.

[0021] The behavioral data includes cumulative entry / exit data, historical request data, payment records, vehicle type, violation records, and dispute records. Specifically, cumulative entry / exit data includes data for this month, this year, and historical years; this month's data is used to determine vehicle activity levels and differentiate processing based on these levels. Historical request data includes the number of requests, request type, and resolution method; obtaining the number of requests from users improves the efficiency of manual processing, ensuring that users can leave the parking lot as quickly as possible. Payment records include the number of overdue payments and the on-time payment rate; obtaining users' payment habits and summarizing overdue payments improves the efficiency of fee collection after AI intervention. Vehicle types include monthly pass vehicles, temporary vehicles, and special vehicles. Violation records include the number of times a vehicle has violated the gate and blacklisting; dispute records include the number of complaints and their resolution. This is used to mark special users, and manual processing is primarily used for these users to avoid the AI ​​mistakenly allowing vehicles to pass.

[0022] Specifically, setting scoring rules includes the following steps: S21. Divide the scoring dimensions and set the dimensions of entry and exit frequency, historical assistance rate, payment compliance, violation record and usage duration; S22. Set weights: set the weight for entry / exit frequency to 20%; set the weight for historical assistance rate to 25%; set the weight for payment compliance to 30%; set the weight for violation records to 15%; and set the weight for usage duration to 10%. S23. Set scoring rules The scoring rule for the frequency of entry and exit is that a person who enters and exits 20 or more times this month will receive full marks, and a person who enters and exits less than 20 times will receive progressively lower scores. The scoring rules for the historical help-seeking rate dimension are as follows: a help-seeking rate of less than or equal to 1% receives full marks, a help-seeking rate of more than 10% receives 0 marks, and the help-seeking rate is calculated by decreasing the score from 1% to 10%. The scoring rules for payment compliance are: full marks for no outstanding payments, and 10 points are deducted for each outstanding payment, until the score reaches 0. The scoring rules for the violation record dimension are as follows: full marks are awarded for no violations, and 20 points are deducted for each violation, until the score reaches 0. The scoring criteria for the vehicle usage duration dimension is that full marks are awarded for registration and usage exceeding 6 months, with scores increasing from 0 to 6 months of registration duration. S24. Output the user's credit score according to the scoring rules.

[0023] In this embodiment, the entry / exit frequency dimension is weighted at 20%, a relatively high weight level, to ensure a large amount of user behavior data, thereby improving the accuracy of post-intervention processing by the AI. The standing assistance dimension is also given a high weight level to identify situations requiring human intervention, ensuring timely intervention during AI interactions and improving passage efficiency. Payment compliance is given the highest weight, as outstanding payments significantly impact a user's credit score, reducing the probability of the AI ​​mistakenly allowing passage.

[0024] Specifically, step S03 also includes the following steps: S31. Obtain user credit scores and behavioral data. S32. If the credit score is greater than or equal to 90 and the number of entries and exits this month is greater than 20, then classify the user into the S-level user tier; otherwise, proceed to step S33. S33. Is the credit score between 80 and 89, or is the user a monthly card user with a credit score greater than or equal to 70? If so, classify the user as a Level A user; otherwise, proceed to step S34. S34. If the credit score is between 60 and 79 and the vehicle is a temporary vehicle with no abnormal records, then classify the user as a B-level user; otherwise, proceed to step S35. S35. Check if the credit score is between 40 and 59 or if there is a record of overdue payment or assistance. If so, classify the user as a C-level user; otherwise, proceed to step S36. S36. Users with a credit score of less than 40, who are blacklisted, or who have a history of disputes of 2 or more are classified as Level D users.

[0025] Level S represents Super Trusted Users, comprising approximately 5% of the total user base; Level A represents High Trusted Users, comprising approximately 20% of the total user base; Level B represents Standard Users, comprising approximately 60% of the total user base; Level C represents Users of Interest, likely with outstanding payments and violation records, comprising approximately 12% of the total user base; Level D represents High-Risk Users, likely with multiple violations, outstanding payments, or being blacklisted, requiring close monitoring to prevent recurrence of toll evasion, comprising approximately 3% of the total user base. By setting reasonable conditions to adjust the proportion of users at each level, it avoids an excessive number of highly trusting users, which could affect the efficiency of AI-powered toll collection, and similarly, an excessive number of low-trusting users, which could affect vehicle traffic efficiency.

[0026] Specifically, the differentiated processing strategy includes, Credit-based clearance allows vehicles of S-level and A-level users to pass directly even if their license plates are occasionally obscured. The system records this information in the background, skips the secondary verification process after payment, and prioritizes their passage. Fast track reduces confirmation steps for S-level and A-level user vehicles, lowering the confidence threshold by 5%; payment anomalies are prioritized for remote verification. The standard procedure applies to vehicles belonging to Class B users, and the standard handling strategy is executed accordingly. Strengthen verification by raising the confidence threshold for C-level user vehicles, adding payment verification steps, and prioritizing manual processing. The system will automatically start recording audio for vehicles belonging to Class D users and simultaneously notify management personnel to prevent automatic passage by raising the barrier. Vehicles will be allowed to pass only after manual approval.

[0027] For high-trust users, proactive credit-based passage and fast-track processing are adopted to improve their passage efficiency and reduce their waiting time, aiming for seamless passage. For low-trust users, a payment verification step is added to ensure they complete the payment process, and manual processing is prioritized to avoid accidental passage by the AI ​​monitoring system. For high-risk users, recordings are proactively recorded to facilitate subsequent dispute resolution. Passing through the barrier is prohibited, and manual verification is conducted to check for any outstanding fees, improving the rate of debt recovery.

[0028] Furthermore, credit release specifically includes the following steps: P01. The AI-controlled query user is categorized as either Level S or Level A; P02. Query the confidence level of vehicle identification this time. If the confidence level of S-level user license plate is greater than or equal to 60% and the confidence level of A-level user vehicle is greater than or equal to 70%, then proceed to step P03; otherwise, execute according to the fast track strategy. P03. Query the number of times the user's credit has been granted in the past 7 days. If the number of times the credit has been granted to an S-level user is less than or equal to 3 times, and the number of times the credit has been granted to an A-level user is less than or equal to 1 time, then proceed to step P04; otherwise, execute the standard processing strategy. P04. Raise the barrier directly, generate a release record in the background, and push it to the management personnel simultaneously.

[0029] For S-level users, a relatively low license plate confidence threshold is set to avoid missed identifications and negatively impact their user experience. The confidence level is also greater than 60% to prevent false identifications, such as misclassifying other users as S-level, which would affect the payment collection rate. For A-level users, a 70% confidence threshold is set to effectively prevent false identifications, while a relatively high threshold further minimizes misjudgments. After license plate recognition, the number of credit approvals granted to each user over the past 7 days is recorded to prevent frequent credit approvals that could negatively impact the payment collection rate.

[0030] Furthermore, the standard processing strategy includes the following steps: Q01. The AI ​​on duty checks whether the user's tier is B. If so, proceed to step Q02. Q02. Query the confidence level of this license plate recognition. If the confidence level is greater than or equal to 80%, proceed to step Q03; otherwise, perform manual verification. Q03. Check if the user has any unpaid records. If so, prompt the user to complete the payment and then allow passage. If not, raise the barrier normally and allow passage. Q04. Generate interaction records and update them to the user profile database.

[0031] Step S07 includes the following steps: When the AI-assisted solution is successfully resolved, the score for the historical help rate dimension increases by 0.5 points; When the AI ​​interaction fails and human intervention is required, the score for the historical help rate dimension will be reduced by 1 point. When an overdue payment is made during the AI-assisted interaction process, the payment compliance score increases by 1 point. When the number of consecutive successful credit approvals reaches a preset value, the license plate confidence threshold is lowered.

[0032] After the interaction with the AI ​​is completed, the user profile is updated so that the AI ​​can process the user's subsequent requests to leave the room more efficiently.

[0033] In step S5, when license plate recognition fails and user information cannot be obtained, the current time and entrance / exit number are used as temporary indexes, and the standard processing strategy is followed. Establishing a temporary index provides an anchor point for subsequent data acquisition and merging.

[0034] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A parking lot differentiated monitoring method based on user profiles, characterized in that, Includes the following steps: S01. Collect behavioral data for each registered vehicle; S02. Set scoring rules and calculate user credit scores based on behavioral data; S03. Set credit scoring levels and divide users into n levels from trust to distrust based on their creditworthiness, and establish a user profile database. S04. Configure the monitoring AI and preset differentiated processing strategies for different groups at different levels; S05. If license plate recognition fails or the gate fails to rise, the monitoring AI is triggered. The monitoring AI uses the vehicle license plate number as an index to query the user level. S06. The AI ​​on duty selects and executes the corresponding strategy from the preset differentiated processing strategy according to the user level; S07. Update the user profile database with the results of this interaction.

2. The parking lot differentiated monitoring method based on user profiles according to claim 1, characterized in that, Step S03 also includes the following steps: S31. Obtain user credit scores and behavioral data. S32. If the credit score is greater than or equal to 90 and the number of entries and exits this month is greater than 20, then classify the user into the S-level user tier; otherwise, proceed to step S33. S33. Is the credit score between 80 and 89, or is the user a monthly card user with a credit score greater than or equal to 70? If so, classify the user as a Level A user; otherwise, proceed to step S34. S34. If the credit score is between 60 and 79 and the vehicle is a temporary vehicle with no abnormal records, then classify the user as a B-level user; otherwise, proceed to step S35. S35. Check if the credit score is between 40 and 59 or if there is a record of overdue payment or assistance. If so, classify the user as a C-level user; otherwise, proceed to step S36. S36. Users with a credit score of less than 40, who are blacklisted, or who have a history of disputes of 2 or more are classified as Level D users.

3. The parking lot differentiated monitoring method based on user profiles according to claim 1, characterized in that, The behavioral data includes cumulative entry and exit data, historical assistance data, payment records, vehicle type, violation records, and dispute records; The cumulative inflow and outflow data includes inflow and outflow data for this month, inflow and outflow data for this year, and inflow and outflow data for historical years. The historical help data includes the number of help requests, the type of help request, and the resolution method; The payment records include the number of overdue payments and the on-time payment rate; The vehicle types include monthly pass vehicles, temporary vehicles, and special vehicles; The violation records include the number of times the gate was breached and blacklist markings; The dispute record includes the number of complaints and the results of their handling.

4. The parking lot differentiated monitoring method based on user profiles according to claim 1, characterized in that, The setting of scoring rules includes the following steps. S21. Divide the scoring dimensions and set the dimensions of entry and exit frequency, historical assistance rate, payment compliance, violation record and usage duration; S22. Set weights: set the weight for entry / exit frequency to 20%; set the weight for historical assistance rate to 25%; set the weight for payment compliance to 30%; set the weight for violation records to 15%; and set the weight for usage duration to 10%. S23. Set scoring rules The scoring rule for the frequency of entry and exit is that a person who enters and exits 20 or more times this month will receive full marks, and a person who enters and exits less than 20 times will receive progressively lower scores. The scoring rules for the historical help-seeking rate dimension are as follows: a help-seeking rate of less than or equal to 1% receives full marks, a help-seeking rate of more than 10% receives 0 marks, and the help-seeking rate is calculated by decreasing the score from 1% to 10%. The scoring rules for payment compliance are: full marks for no outstanding payments, and 10 points are deducted for each outstanding payment, until the score reaches 0. The scoring rules for the violation record dimension are as follows: full marks are awarded for no violations, and 20 points are deducted for each violation, until the score reaches 0. The scoring criteria for the vehicle usage duration dimension is that full marks are awarded for registration and usage exceeding 6 months, with scores increasing from 0 to 6 months of registration duration. S24. Output the user's credit score according to the scoring rules.

5. The parking lot differentiated monitoring method based on user profiles according to claim 4, characterized in that, The differentiated processing strategy includes, Credit-based clearance allows vehicles of S-level and A-level users to pass directly even if their license plates are occasionally obscured. The system records this information in the background, skips the secondary verification process after payment, and prioritizes their passage. Fast track reduces confirmation steps for S-level and A-level user vehicles, lowering the confidence threshold by 5%; payment anomalies are prioritized for remote verification. The standard procedure applies to vehicles belonging to Class B users, and the standard handling strategy is executed accordingly. Strengthen verification by raising the confidence threshold for C-level user vehicles, adding payment verification steps, and prioritizing manual processing. The system will automatically start recording audio for vehicles belonging to Class D users and simultaneously notify management personnel to prevent automatic passage by raising the barrier. Vehicles will be allowed to pass only after manual approval.

6. The parking lot differentiated monitoring method based on user profiles according to claim 5, characterized in that, The credit release specifically includes the following steps: P01. The AI-controlled query user is categorized as either Level S or Level A; P02. Query the confidence level of vehicle identification this time. If the confidence level of S-level user license plate is greater than or equal to 60% and the confidence level of A-level user vehicle is greater than or equal to 70%, then proceed to step P03; otherwise, execute the standard processing strategy. P03. Query the number of times the user's credit has been granted in the past 7 days. If the number of times the credit has been granted to an S-level user is less than or equal to 3 times, and the number of times the credit has been granted to an A-level user is less than or equal to 1 time, then proceed to step P04; otherwise, execute the standard processing strategy. P04. Raise the barrier directly, generate a release record in the background, and push it to the management personnel simultaneously.

7. The parking lot differentiated monitoring method based on user profiles according to claim 5, characterized in that, Step S07 includes the following steps: When the AI-assisted solution is successfully resolved, the score for the historical help rate dimension increases by 0.5 points; When the AI ​​interaction fails and human intervention is required, the score for the historical help rate dimension will be reduced by 1 point. When the number of consecutive successful credit approvals reaches a preset value, the license plate confidence threshold is lowered.

8. The parking lot differentiated monitoring method based on user profiles according to claim 5, characterized in that, In step S5, when license plate recognition fails and user information cannot be obtained, the current time and entrance / exit number are used as temporary indexes, and the standard processing strategy is executed.

Images