An intelligent monitoring method and device for port travel inspection risk identification

Abstract

The application provides an intelligent monitoring method and device for port travel inspection risk identification, comprising the following steps: constructing a multi-dimensional data acquisition system, realizing synchronous acquisition and full-chain association of multiple types of data such as video stream, identity information, shopping record, and customs track, forming a complete data link of “identity-track-behavior-equipment state”, and solving the problem of insufficient integration of existing multi-source data. Relying on the full-chain data association relationship and the special risk model, the target objects such as passengers needing verification, non-conventional customs behavior, and abnormal flow of island tax-free goods are identified, the risk judgment ability of island tax-free anti-backflow, key personnel control and other scenes is improved, and the supervision accuracy is improved. Through the hierarchical push and classified verification and disposal mechanism of the hierarchical risk early warning signal, the accurate docking of the early warning information and the disposal terminal is realized, the problem of low efficiency of front-end monitoring and back-end disposal cooperation is solved, and the risk response and on-site disposal cycle is shortened.

Description

Technical Field

[0001] This application relates to the field of system control technology, and in particular to an intelligent monitoring method and device for identifying risks in port passenger inspection. Background Technology

[0002] During the implementation of the border closure, the number of people and their luggage leaving the island through the second-line ports has surged, placing higher demands on the efficiency and accuracy of port passenger inspection and supervision. The existing manual passenger inspection and supervision model is limited by manpower allocation and operating mode, resulting in problems such as low supervision efficiency and delayed risk identification, making it difficult to adapt to the current supervision load of the second-line ports. Summary of the Invention

[0003] The purpose of this application is to at least partially solve one of the technical problems in the related art.

[0004] Therefore, the first objective of this application is to propose an intelligent monitoring method for identifying risks in port passenger inspection.

[0005] The second objective of this application is to propose an intelligent monitoring device for identifying risks in port passenger inspection.

[0006] The third objective of this application is to propose an electronic device.

[0007] The fourth objective of this application is to provide a computer-readable storage medium.

[0008] The fifth objective of this application is to provide a computer program product.

[0009] To achieve the above objectives, the first aspect of this application proposes an intelligent monitoring method for identifying risks in port passenger inspection, comprising: A multi-dimensional data collection system is constructed for port passenger inspection scenarios. Based on various regulatory nodes of the smart checkpoint, video stream data, passenger identity information, island shopping records, customs clearance trajectory information, associated tag data of preset verification personnel, transportation information and equipment operation status data are collected synchronously. The equipment operation status data is the software and hardware operating parameter data of the regulatory nodes. Based on the preset port passenger inspection data association rules, the data collected synchronously by various regulatory nodes of the smart checkpoint are mapped to form a full-chain data association relationship of identity-trajectory-behavior-equipment status; Using the established risk model, analysis is performed based on the data correlation of the entire chain to identify target objects including passengers requiring verification, irregular customs clearance behavior, abnormal circulation of duty-free goods on the island, and means of transportation requiring verification, and trigger graded risk warning signals; The aforementioned risk warning signals are implemented through tiered push notifications, categorized verification and handling, and result retrospective analysis to achieve full-process control of port passenger inspection risks.

[0010] To achieve the above objectives, a second aspect of this application provides an intelligent monitoring device for identifying risks in port passenger inspection, comprising: The data acquisition module is used to build a multi-dimensional data acquisition system. For port passenger inspection scenarios, it synchronously collects video stream data, passenger identity information, island shopping records, customs clearance trajectory information, associated tag data of preset verification personnel, transportation information and equipment operation status data based on various regulatory nodes of the smart checkpoint. The equipment operation status data is the software and hardware operating parameter data of the regulatory node. The association mapping module is used to perform association mapping on the data synchronously collected by various regulatory nodes of the smart checkpoint based on the preset port passenger inspection data association rules, so as to form a full-chain data association relationship of identity-trajectory-behavior-equipment status; The analysis and early warning module is used to perform analysis based on the established risk model and the full-chain data correlation to identify target objects including passengers requiring verification, irregular customs clearance behavior, abnormal circulation of duty-free goods on the island, and means of transportation requiring verification, and trigger graded risk early warning signals. The control module is used to implement graded push, classified verification and handling, and result backtracking of the graded risk warning signals to achieve full-process control of port passenger inspection risks.

[0011] To achieve the above objectives, a third aspect of this application provides an electronic device, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions to implement the steps of the intelligent monitoring method for port passenger inspection risk identification proposed in the first aspect of this application.

[0012] To achieve the above objectives, a fourth aspect of this application provides a computer-readable storage medium that, when the instructions in the storage medium are executed by a processor of an electronic device, enables the electronic device to perform the steps of the intelligent monitoring method for identifying passenger inspection risks at ports of entry, as proposed in the first aspect of this application.

[0013] To achieve the above objectives, a fifth aspect of this application provides a computer program product, including a computer program that, when executed by a processor in a communication device, implements the steps of the intelligent monitoring method for identifying passenger inspection risks at ports of entry proposed in the first aspect of this application.

[0014] In this embodiment, a multi-dimensional data collection system is constructed to achieve synchronous collection and full-chain association of various data types, including video streams, identity information, shopping records, and customs clearance trajectories. This forms a complete data link of "identity-trajectory-behavior-device status," addressing the problem of insufficient integration of existing multi-source data and providing accurate data support for risk identification. Based on the full-chain data association and a dedicated risk model, targeted identification is achieved for passengers requiring verification, irregular customs clearance behaviors, and abnormal circulation of duty-free goods. This improves risk assessment capabilities in scenarios such as preventing the return of duty-free goods to the mainland and managing key personnel, thereby enhancing regulatory accuracy. Through a tiered risk warning signal push and categorized verification and handling mechanism, precise connection between warning information and handling terminals is achieved, resolving the problem of inefficient coordination between front-end monitoring and back-end handling, shortening the risk response and on-site handling cycle, and balancing customs clearance efficiency and regulatory effectiveness.

[0015] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0016] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 A flowchart illustrating an intelligent monitoring method for identifying risks at ports of entry, provided as an embodiment of this application; Figure 2 A schematic diagram of the structure of an intelligent monitoring device for identifying risks in port passenger inspection provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application. Detailed Implementation

[0017] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0018] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the embodiments of this application. The singular forms “a” and “the” as used in the embodiments of this application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0019] It should be understood that although the terms first, second, third, etc., may be used to describe various information in the embodiments of this application, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words "if" and "suppose" as used herein can be interpreted as "when," "when," or "in response to a determination."

[0020] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0021] With the ongoing construction of the closed-off operation, the second-line port, as a hub for the flow of people, baggage, and various taxable goods leaving the island, has seen a surge in customs clearance. This includes an increase in the number of tourists, encompassing ordinary travelers, business people, and duty-free shoppers; baggage is characterized by its diverse categories, complex attributes, and increased concealment, including not only regular personal items but also duty-free goods and cross-border trade-related items; and the implementation of customs clearance models such as staggered departures and centralized declarations has further increased the complexity of the regulatory landscape.

[0022] However, the current manual passenger inspection and supervision model, still used in some scenarios, is no longer sufficient to meet the increasing regulatory workload and requirements. In terms of manpower allocation, supervision in some scenarios heavily relies on fixed personnel input, while surges in customs clearance traffic are random and sudden. During peak periods, a mismatch between manpower and regulatory needs can easily occur, leading to an imbalance in the allocation of regulatory resources.

[0023] In terms of operational mode, manual passenger inspection supervision requires following the process of "verifying identity for each person and inspecting baggage piece by piece." Staff need to manually verify identity information and manually check baggage items. This process is not only cumbersome but also has a limited capacity for processing per unit of time, directly resulting in low supervision efficiency and difficulty in coping with the impact of large-scale customs clearance traffic. Regarding risk identification, manual passenger inspection supervision mainly relies on the personal experience and subjective judgment of staff, lacking systematic and standardized risk identification support. It has significant limitations in accurately screening key personnel, identifying concealed illegal items, and capturing abnormal behavior in real time, leading to a lag in risk identification and a tendency for missed or misjudgments. This makes it impossible to form a comprehensive, end-to-end risk control system.

[0024] The following description, with reference to the accompanying drawings, describes an intelligent monitoring method and apparatus for identifying risks at ports of entry.

[0025] Figure 1 This is a flowchart illustrating an intelligent monitoring method for identifying risks at ports of entry, provided as an embodiment of this application.

[0026] like Figure 1 As shown, the intelligent monitoring method for identifying risks in port passenger inspection includes, but is not limited to, the following steps: S101 constructs a multi-dimensional data collection system. Targeting port passenger inspection scenarios, it synchronously collects video stream data, passenger identity information, island shopping records, customs clearance trajectory information, associated tag data of preset verification personnel, transportation information, and equipment operation status data based on various regulatory nodes of the smart checkpoint. Among them, the equipment operation status data is the hardware and software operating parameter data of the regulatory nodes.

[0027] In one feasible implementation, the various regulatory nodes of the smart checkpoint include at least one of the following: a first-line entry checkpoint, a second-line exit checkpoint, a duty-free shopping area verification node, a baggage security checkpoint, and a transportation inspection node. Among these, the first-line entry checkpoint and the second-line exit checkpoint, as important control nodes for personnel and transportation entering and leaving the island, primarily collect passenger identity information, transportation passage information, video stream data, and customs clearance trajectory information. The duty-free shopping area verification node, as a control node related to the offshore duty-free business, primarily collects offshore shopping records, passenger identity verification information, and associated tag data of preset verification personnel. The baggage security checkpoint, as a goods security control node, primarily collects baggage security check video stream data and equipment operating status data. The transportation inspection node, as a compliance control node for transportation vehicles, primarily collects vehicle appearance images, license plate information, operational condition-related data, and passage trajectory information.

[0028] In one feasible implementation, video stream data is synchronously collected by intelligent surveillance cameras deployed at monitoring nodes such as port security checkpoints, isolation areas, turnstiles, and checkpoints. These intelligent surveillance cameras support the GB / T28181 and Onvif protocols and are compatible with SDK interfaces for non-standard equipment. The collected content includes passenger facial images, dynamic video of their behavior, and images of the exterior of vehicles.

[0029] In one feasible implementation, after collecting video stream data, the process further includes: performing full-target recognition on the collected video stream data, extracting passenger facial feature parameters, and identifying abnormal behaviors in the clearance scenario. Passenger facial feature parameters include at least one of hairstyle, whether a hat is worn, whether glasses are worn, and whether a mask is worn; abnormal behaviors include at least one of abnormal running, lingering, crowd gathering, leaving items behind, and vigorous exercise.

[0030] In some embodiments, the full-target recognition processing is executed in parallel by the analysis module of the smart surveillance camera. By parsing the video stream data of at least 200 smart surveillance cameras, structured human image data and structured behavioral data are obtained.

[0031] In one feasible implementation, passenger identity information is collected collaboratively by document reading devices and biometric collection devices. This includes structured data such as passenger name, document number, gender, age, and document validity period. The collection process meets relevant compliance requirements for personal information protection, and data transmission uses an encryption protocol.

[0032] In one feasible implementation, offshore shopping records are collected by connecting with the offshore duty-free shopping supervision system. Specifically, this includes related data such as shopping time, product category, product quantity, consumption amount, pickup status, and duty-free quota usage. The data update frequency is consistent with the transaction confirmation time of the duty-free shopping system.

[0033] In one feasible implementation, customs clearance trajectory information is recorded synchronously with timestamps through the positioning module of the regulatory node. This includes time-series data such as the time when the passenger enters the port area, the time when they pass through each security checkpoint, the time when they arrive at the gate for verification, and the time when they complete customs clearance and release, as well as the physical location coordinates corresponding to each step, forming a complete customs clearance time-series trajectory chain.

[0034] In one feasible implementation, the pre-set associated tag data of verification personnel includes tags for key supervision personnel, tags for whitelisted personnel, tags for tax-free repatriation risk, and tags associated with violation records. The tag data is synchronized in real time with the customs key personnel database and the supervision business system, and supports the dynamic addition, modification, and expiration management of tags.

[0035] In one feasible implementation, vehicle passage information includes vehicle license plate number, model, color, entry time, exit time, and passage route, which is collected collaboratively by smart checkpoint data collection equipment and photo recognition modules.

[0036] In one feasible implementation, the equipment operation status data includes hardware operation data of the monitoring node hardware device and software operation data of the software system. The hardware operation data includes CPU utilization, memory usage, equipment temperature and number of mechanical component operations, while the software operation data includes process operation status, interface call success rate and data transmission latency.

[0037] S102, based on preset port passenger inspection data association rules, performs correlation mapping on the data synchronously collected by various regulatory nodes of the smart checkpoint, forming a full-chain data association relationship of identity-trajectory-behavior-equipment status.

[0038] In one feasible implementation, the pre-defined port passenger inspection data association rules are based on passenger identity information, which includes at least one of ID card number, entry / exit documents, and customs clearance certificates. A one-to-one mapping is established between passenger identity information and customs clearance trajectory information, passenger facial feature parameters, abnormal behavior identification results, island shopping records, and the associated tag data of pre-defined verification personnel.

[0039] In one feasible implementation, passenger identity information is statically associated with island shopping records and the associated tag data of preset verification personnel to form a basic association chain of "identity-attribute". The passenger identity information, the associated tag data of preset verification personnel and the customs clearance trajectory information are bound according to timestamps to ensure that the identity attributes and trajectories of the same passenger are dynamically and accurately matched at different time points.

[0040] In one feasible implementation, based on timestamps and location information of various regulatory nodes at smart checkpoints, the basic association chain of "identity-attribute" is dynamically associated with customs clearance trajectory information, passenger facial feature parameters in video stream data, and abnormal behavior identification results to form a dynamic association chain of "identity-trajectory-behavior". Among them, the island shopping record and transportation passage information are associated according to the verification node to realize the binding of shopping behavior and transportation.

[0041] In one feasible implementation, the hardware and software operating condition parameter data (i.e., equipment operating status data) of various monitoring nodes of the smart checkpoint are associated with the dynamic association chain of "identity-trajectory-behavior" in a scenario-based manner to form a full-chain association relationship of "identity-trajectory-behavior-equipment status". Among them, video stream data and equipment operating status data are associated according to spatial location to ensure that video stream data and hardware and software operating condition parameter data in the same monitoring area correspond one-to-one.

[0042] In one feasible implementation, the correlation mapping is executed through the correlation scheduling module of the data fusion layer, which supports the visual configuration and dynamic updating of correlation rules (including timestamp binding rules, verification node correlation rules, and spatial location correlation rules). During the mapping process, a data legality verification mechanism is adopted to automatically remove duplicate and incorrectly correlated data, ensuring the accuracy of the data correlation relationship across the entire chain. The correlated data is stored in a structured data format, supporting multi-dimensional retrieval and trajectory tracing based on passenger identity information, time range, regulatory node type, and spatial location information.

[0043] S103 utilizes the established risk model to perform analysis based on the full-chain data correlation, identifying target objects including passengers requiring verification, irregular customs clearance behaviors, abnormal circulation of duty-free goods on the island, and means of transportation requiring verification, triggering graded risk warning signals.

[0044] In one feasible implementation, the constructed risk model includes a key passenger identification sub-model, an abnormal behavior early warning sub-model, an offshore duty-free goods anti-repatriation sub-model, and a risky transportation vehicle identification sub-model. Each sub-model is trained using multiple sets of data through machine learning. The multiple sets of data include a first type of data and a second type of data. Each set of data in the first type includes the historical normal status of the controlled object in the port passenger inspection scenario and the first label corresponding to the historical normal status. Each set of data in the second type includes the historical abnormal status of the controlled object in the port passenger inspection scenario and the second label corresponding to the historical abnormal status. Among them, the controlled objects include passengers, customs clearance behavior, offshore duty-free goods, and transportation vehicles.

[0045] In some embodiments, for the "passenger" control object, the first type of data includes the passenger's historical normal identity information, normal customs clearance trajectory, compliant shopping records and corresponding first label ("normal passenger"); the second type of data includes the historical identity label of key supervision personnel, multiple records of illegal customs clearance, false identity verification records and corresponding second label ("problem passenger").

[0046] In some embodiments, for the "customs clearance behavior" control object, the first type of data includes historical normal passage speed, compliance verification process trajectory, no abnormal behavior records and corresponding first tags ("routine customs clearance behavior"); the second type of data includes historical abnormal running, lingering, multiple trips to verification nodes and other behavior records and corresponding second tags ("non-routine customs clearance behavior").

[0047] In some embodiments, for the "offshore duty-free goods" control targets, the first type of data includes historical compliant shopping frequency, normal pick-up and departure records, compliant use records of duty-free quota and corresponding first labels ("normal circulation"); the second type of data includes historical multiple shopping without leaving the island, bulk splitting and carrying, and purchase records exceeding the quota and corresponding second labels ("abnormal circulation").

[0048] In some embodiments, for the "vehicle" control object, the first type of data includes historical compliant travel trajectories, no records of illegal passenger transport, normal inspection results and corresponding first labels ("normal vehicle"); the second type of data includes historical risk vehicle registration information, records of illegal transportation, records of carrying high-risk passengers and corresponding second labels ("risk vehicle").

[0049] In one feasible implementation, performing analysis based on the entire chain of data relationships includes: The key passenger identification sub-model compares passenger identity information, preset verification personnel associated tag data with key personnel information database, and combines the frequency and time interval characteristics of customs clearance trajectory to identify key supervision categories of passengers who need to be verified. By using the abnormal behavior early warning sub-model, the abnormal behavior data parsed from the video stream is correlated with the temporal characteristics of the clearance trajectory to identify combined unconventional clearance behaviors such as "abnormal running + rapid clearance" and "lingering and wandering + multiple trips to verification nodes".

[0050] By using the offshore duty-free anti-reflow sub-model, we can link offshore shopping records, customs clearance routes, and transportation information. Based on shopping frequency, pickup status, boarding / flight records, and departure intervals, we can determine whether there are abnormal circulation situations such as "multiple reflows", "pickup without leaving the island", or "batch of goods split and carried".

[0051] By using a risky transportation vehicle identification sub-model, the vehicle's travel information is compared with a risk vehicle database. Combined with the nodes along the vehicle's travel trajectory and the associated tag data of the passengers, high-risk transportation vehicles that require verification are identified.

[0052] In one feasible implementation, the tiered risk warning signal includes Level 1, Level 2, and Level 3 warnings, with each level corresponding to different triggering conditions and handling priorities: Level 1 warning is a high-priority warning. The triggering condition is the identification of a correlation between "key monitored passengers requiring verification + abnormal circulation of bulk duty-free goods + risky means of transportation". The warning signal is pushed to the command center's large screen and the on-site mobile individual terminal in real time, requiring on-site handling to be completed in the shortest possible time (e.g., 3 minutes).

[0053] Level 2 warning is a medium-priority warning. The triggering conditions are the identification of "unconventional customs clearance behavior + abnormal flow of duty-free goods in a single transaction", or only the identification of "high-risk passengers / transportation vehicles requiring verification". The warning signal is pushed to the on-site handling terminal, requiring the on-site handling to be completed as soon as possible (e.g., within 10 minutes).

[0054] Level 3 warnings are low-priority warnings. The trigger condition is the identification of a "single minor abnormal feature (such as a single instance of lingering or wandering, or a slight mismatch between shopping records and customs clearance trajectory)". The warning signal is only pushed to the customer service platform for sorting and processing.

[0055] In some embodiments, the encapsulated content of the graded risk warning signal includes the warning level, passenger ID number / vehicle license plate number, associated data evidence (video clips / images / structured data), the name of the sub-risk model triggered by the warning, the location and timestamp of the regulatory node where the warning occurred, to ensure that the personnel handling the situation have complete evidence for handling the situation.

[0056] In one feasible implementation, the analysis based on the full-chain data association relationship further includes: The facial feature parameters of passengers extracted from video stream data are compared with the collected passenger identity information to verify consistency. The verification results are recorded, including the verification difference items and the degree of difference. The verification difference items are features that are inconsistent with the baseline data, such as hairstyle, whether a hat is worn, whether glasses are worn, and whether a mask is worn. The degree of difference is represented by feature matching similarity measurement. The similarity threshold can be configured through the system's visual interface, with a range of 85%-95%.

[0057] Based on timestamps and spatial coordinates, abnormal behaviors and customs clearance trajectory information extracted from video stream data are identified and recorded to determine the spatiotemporal characteristics of the abnormal behaviors. These spatiotemporal characteristics include the customs clearance point, the time period during which the customs clearance occurs, and the duration of the customs clearance behavior. Specifically, the spatiotemporal characteristics include the customs clearance point where the abnormal behavior occurs (such as security checkpoints, turnstiles, duty-free verification areas), the time period during which the customs clearance occurs (accurate to the minute), and the duration of the customs clearance behavior (accurate to the second), and these characteristics correspond one-to-one with the time sequence nodes of the customs clearance trajectory.

[0058] The verification results, customs clearance spatiotemporal characteristics, island shopping records, associated tag data of preset verification personnel, transportation information, and equipment operation status data are integrated from multiple dimensions to form a judgment basis covering identity, trajectory, behavior, and equipment status. Analysis of the entire data chain's relationships is then performed based on this judgment basis.

[0059] In one feasible implementation, the target objects are identified, including passengers requiring verification, irregular customs clearance behaviors, abnormal circulation of duty-free goods, and vehicles requiring verification. This includes: passengers requiring verification include whether passenger identity information matches the associated tag data of preset verification personnel, and whether the passenger identity information is consistent with the facial feature parameters extracted from the video stream data; irregular customs clearance behaviors include abnormal jumps across nodes in the customs clearance trajectory; abnormal circulation of duty-free goods includes inconsistencies between the offshore shopping record and the declared information; and vehicles requiring verification include whether the vehicle identification matches the preset verification list, and whether there are discrepancies between the declared information of the transported goods and the offshore shopping record.

[0060] In some embodiments, the verification results in the judgment criteria are used as the initial screening condition, linking passenger identity information with the entire chain of data to complete the initial stratification of "compliance / risk". If the verification result shows that the similarity is greater than or equal to the similarity threshold, the identity verification is deemed to have passed; if the verification result shows that the similarity is less than the similarity threshold, the identity verification is deemed to be abnormal, and the verification difference items, such as "no glasses in the ID photo vs. glasses in the video capture" and "difference in hairstyle", are extracted and marked as key analysis objects, triggering subsequent multi-level verification. By linking the passenger identity information with the preset verification personnel's associated tag data in the entire chain of data, if the person is on the whitelist, even if there are slight differences, the analysis priority can be temporarily reduced; if the person is a key regulatory tag, the analysis priority is directly upgraded to high priority.

[0061] In some embodiments, based on timestamps, the occurrence time and duration of abnormal behavior (such as loitering) are compared with the "node passage time" and "stay time at each stage" in the customs clearance trajectory information to verify whether the abnormal behavior matches the customs clearance process (such as whether "staying in the duty-free verification area for 30 minutes" corresponds to the "shopping verification not completed" trajectory). If there is no corresponding customs clearance trajectory for the time period of the abnormal behavior (such as "staying outside the gate" but not recording the time of entry into the port), it is marked as "trajectory abnormal association, triggering subsequent vehicle passage information verification (whether it is an unregistered accompanying person).

[0062] In some embodiments, the device operating status data (such as camera voltage, network bandwidth, and software response time) associated with the abnormal behavior at the critical point of occurrence is used. If the device has fault records (such as abnormal voltage or data transmission delay), the abnormal behavior is determined to be "device misidentification," and the weight of this abnormal feature is reduced. If the device operating parameters are normal, the abnormal behavior is confirmed to be "real occurrence," and the risk weight of this feature is strengthened.

[0063] In some embodiments, the island shopping records (shopping time, quantity of goods, and pickup status) are matched with the transportation information (license plate, travel trajectory, and association of boarding personnel) using passenger identity information and verification node ID. If there are situations such as "multiple shopping records but no corresponding island boarding / flight records", "bulk purchase of goods does not match the vehicle's carrying capacity", or "high-risk passengers are associated with high-risk vehicles", they are marked as "combined risk characteristics".

[0064] In some embodiments, the associated tags of the pre-set verification personnel (such as "duty-free repatriation risk tag" or "violation record tag") are matched with abnormal behavior types and access points. If a traveler with the "duty-free repatriation risk tag" exhibits abnormal behavior such as "lingering and loitering + multiple trips" at the verification point in the duty-free shopping area, and the shopping record shows "multiple purchases of high-value goods in a short period of time", then a strong risk association of "tag + behavior + scenario" is formed.

[0065] S104 implements tiered push, classified verification and handling, and result traceability of graded risk warning signals to achieve full-process control of port passenger inspection risks.

[0066] In one feasible implementation, for passengers requiring verification, the first-level warning process involves "joint verification by two people + baggage inspection + secondary identity verification"; the second-level warning process involves "single-person verification + random baggage inspection"; and the third-level warning process involves "back-end data review + continuous trajectory monitoring".

[0067] In one feasible implementation, for unconventional customs clearance behavior, the first-level warning follows the process of "on-site interception + behavior tracing + investigation of related personnel"; the second-level warning follows the process of "on-site inquiry + situation recording"; and the third-level warning follows the process of "data traceability + subsequent review".

[0068] In one feasible implementation, for abnormal circulation of duty-free goods on outlying islands, a Level 1 warning will implement a process of "goods seizure + transaction tracing + means of transport verification"; a Level 2 warning will implement a process of "goods verification + shopping record verification".

[0069] In one feasible implementation, for vehicles requiring verification, the Level 1 warning process involves "vehicle impoundment + comprehensive inspection + verification of drivers and passengers"; the Level 2 warning process involves "vehicle spot checks + verification of travel routes".

[0070] In summary, the intelligent monitoring method for port passenger inspection risk identification provided in this application constructs a multi-dimensional data collection system to achieve synchronous collection and full-chain association of multiple types of data, such as video streams, identity information, shopping records, and customs clearance trajectories. This forms a complete data link of "identity-trajectory-behavior-device status," solving the problem of insufficient integration of existing multi-source data and providing accurate data support for risk identification. Based on the full-chain data association and a dedicated risk model, it specifically identifies target objects such as passengers requiring verification, irregular customs clearance behaviors, and abnormal circulation of duty-free goods, improving risk assessment capabilities in scenarios such as preventing the return of duty-free goods and managing key personnel, thereby enhancing regulatory accuracy. Through a tiered risk warning signal push and classified verification and handling mechanism, it achieves precise connection between warning information and handling terminals, solving the problem of inefficient coordination between front-end monitoring and back-end handling, shortening the risk response and on-site handling cycle, and balancing customs clearance efficiency and regulatory effectiveness.

[0071] Corresponding to the aforementioned intelligent monitoring method for identifying risks in port passenger inspection, this application also provides an intelligent monitoring device for identifying risks in port passenger inspection. Since the embodiments of the intelligent monitoring device for identifying risks in port passenger inspection in this application correspond to the embodiments of the aforementioned intelligent monitoring method for identifying risks in port passenger inspection, details not disclosed in the embodiments of the intelligent monitoring device for identifying risks in port passenger inspection can be referred to the embodiments of the aforementioned intelligent monitoring method for identifying risks in port passenger inspection, and will not be repeated here.

[0072] Figure 2 This is a schematic diagram of the structure of an intelligent monitoring device for identifying risks in port passenger inspection, provided as an embodiment of this application.

[0073] like Figure 2 As shown, the intelligent monitoring device 200 for identifying risks in port passenger inspection includes: The data acquisition module 201 is used to build a multi-dimensional data acquisition system. For the port passenger inspection scenario, it synchronously collects video stream data, passenger identity information, island shopping records, customs clearance trajectory information, associated tag data of preset verification personnel, transportation information and equipment operation status data based on various regulatory nodes of the smart checkpoint. Among them, the equipment operation status data is the software and hardware operating parameter data of the regulatory node. The association mapping module 202 is used to perform association mapping on the data synchronously collected by various regulatory nodes of the smart checkpoint based on the preset port passenger inspection data association rules, so as to form a full-chain data association relationship of identity-trajectory-behavior-equipment status; The analysis and early warning module 203 is used to utilize the constructed risk model to perform analysis based on the data correlation of the entire chain, identify target objects including passengers requiring verification, irregular customs clearance behavior, abnormal circulation of duty-free goods on the island, and means of transportation requiring verification, and trigger graded risk early warning signals. Control module 204 is used to implement graded push, classified verification and handling, and result backtracking of graded risk warning signals to achieve full-process control of port passenger inspection risks.

[0074] The methods and apparatus provided in the embodiments of this application have been described above. To achieve the functions of the methods provided in the embodiments of this application, the methods and apparatus can be further refined using electronic devices.

[0075] Figure 3 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application. Figure 3 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0076] like Figure 3 As shown, the electronic device 300 includes a processor 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from memory 306 into a random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the electronic device 300. The processor 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0077] The following components are connected to I / O interface 305: memory 306 including hard disks, etc.; and communication section 307 including network interface cards such as LAN (Local Area Network) cards, modems, etc., which performs communication processing via a network such as the Internet; and driver 308 is also connected to I / O interface 305 as needed.

[0078] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 307. When the computer program is executed by processor 301, it performs the functions defined in the methods of this application.

[0079] In an exemplary embodiment, a storage medium including instructions is also provided, such as a memory including instructions, which can be executed by a processor 301 of an electronic device 300 to perform the above-described method. Optionally, the storage medium may be a non-transitory computer-readable storage medium, such as a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device.

[0080] In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can transmit, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wireline, optical fiber, RF, etc., or any suitable combination thereof.

[0081] In this embodiment of the disclosure, a computer program product is also provided, including a computer program that, when executed by a processor, implements the methods disclosed in the above embodiments.

[0082] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0083] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. An intelligent monitoring method for identifying risks in passenger inspection at ports of entry, characterized in that, include: A multi-dimensional data collection system is constructed for port passenger inspection scenarios. Based on various regulatory nodes of the smart checkpoint, video stream data, passenger identity information, island shopping records, customs clearance trajectory information, associated tag data of preset verification personnel, transportation information and equipment operation status data are collected synchronously. The equipment operation status data is the software and hardware operating parameter data of the regulatory nodes. Based on the preset port passenger inspection data association rules, the data collected synchronously by various regulatory nodes of the smart checkpoint are mapped to form a full-chain data association relationship of identity-trajectory-behavior-equipment status; Using the established risk model, analysis is performed based on the data correlation of the entire chain to identify target objects including passengers requiring verification, irregular customs clearance behavior, abnormal circulation of duty-free goods on the island, and means of transportation requiring verification, and trigger graded risk warning signals; The aforementioned risk warning signals are implemented through tiered push notifications, categorized verification and handling, and result retrospective analysis to achieve full-process control of port passenger inspection risks.

2. The method according to claim 1, characterized in that, The various monitoring nodes of the smart checkpoint include at least one of the following: first-line entry checkpoint, second-line exit checkpoint, duty-free shopping area verification node, baggage security check node, and transportation inspection node.

3. The method according to claim 2, characterized in that, After acquiring the video stream data, the process also includes: Perform full-target recognition on the collected video stream data, extract passenger facial feature parameters, and identify abnormal behaviors in the customs clearance scenario; The passenger's facial feature parameters include at least one of the following: hairstyle, whether wearing a hat, whether wearing glasses, and whether wearing a mask; The abnormal behavior includes at least one of the following: abnormal running, lingering or loitering, crowd gathering, leaving behind objects, or strenuous exercise.

4. The method according to claim 1, characterized in that, The device operating status data includes hardware operating data and software operating data. The hardware operating data includes CPU utilization, memory usage, device temperature, and number of times mechanical components have been used. The software operating data includes process running status, interface call success rate, and data transmission latency.

5. The method according to claim 1, characterized in that, The preset port passenger inspection data association rules are association rules established based on passenger identity information, which includes at least one of ID card number, entry and exit documents, and customs clearance certificate.

6. The method according to claim 1, characterized in that, The association mapping includes: Passenger identity information, associated tag data of preset verification personnel, and customs clearance trajectory information are bound together by timestamps. Offshore shopping records and transportation passage information are linked by verification nodes. Video stream data and equipment operation status data are linked by spatial location.

7. The method according to claim 1, characterized in that, The risk model is derived by training multiple sets of data through machine learning. The multiple sets of data include a first type of data and a second type of data. Each set of data in the first type of data includes the historical normal status of the controlled object in the port passenger inspection scenario and the first label corresponding to the historical normal status. Each set of data in the second type of data includes the historical abnormal status of the controlled object in the port passenger inspection scenario and the second label corresponding to the historical abnormal status. The controlled object includes passengers, customs clearance behavior, duty-free goods leaving the island, and means of transportation.

8. The method according to claim 1, characterized in that, The analysis based on the entire chain of data relationships includes: The facial feature parameters of passengers extracted from video stream data are compared with the collected passenger identity information to verify consistency. The verification results, including the verification discrepancies and the degree of discrepancy, are recorded. Based on timestamps and spatial coordinates, abnormal behaviors and clearance trajectory information extracted from video stream data are identified and recorded to determine the spatiotemporal characteristics of clearance to which the abnormal behaviors belong. The spatiotemporal characteristics of clearance include clearance points, clearance time periods, and clearance behavior durations. The verification results, the customs clearance spatiotemporal characteristics, the island shopping records, the associated tag data of the preset verification personnel, the transportation information and the equipment operation status data are integrated in multiple dimensions to form a judgment basis covering identity, trajectory, behavior and equipment status. The data relationship of the entire chain is analyzed based on the judgment criteria.

9. The method according to claim 1, characterized in that, The identification includes target objects such as passengers requiring verification, irregular customs clearance behavior, abnormal circulation of duty-free goods leaving the island, and means of transportation requiring verification, including: Passenger verification includes checking whether passenger identity information matches the associated tag data of preset verification personnel, and whether passenger identity information is consistent with facial feature parameters extracted from video stream data; abnormal customs clearance behavior includes abnormal jumps across nodes in the customs clearance trajectory; abnormal circulation of duty-free goods is when the shopping record for leaving the island is inconsistent with the declared information; and verification of the means of transportation is required to check whether the means of transportation identification matches the preset verification list, and whether there are differences between the declared information of the transported goods and the shopping record for leaving the island.

10. An intelligent monitoring device for identifying risks in port passenger inspection, characterized in that, include: The data acquisition module is used to build a multi-dimensional data acquisition system. For port passenger inspection scenarios, it synchronously collects video stream data, passenger identity information, island shopping records, customs clearance trajectory information, associated tag data of preset verification personnel, transportation information and equipment operation status data based on various regulatory nodes of the smart checkpoint. The equipment operation status data is the software and hardware operating parameter data of the regulatory node. The association mapping module is used to perform association mapping on the data synchronously collected by various regulatory nodes of the smart checkpoint based on the preset port passenger inspection data association rules, so as to form a full-chain data association relationship of identity-trajectory-behavior-equipment status; The analysis and early warning module is used to perform analysis based on the established risk model and the full-chain data correlation to identify target objects including passengers requiring verification, irregular customs clearance behavior, abnormal circulation of duty-free goods on the island, and means of transportation requiring verification, and trigger graded risk early warning signals. The control module is used to implement graded push, classified verification and handling, and result backtracking of the graded risk warning signals to achieve full-process control of port passenger inspection risks.

Images