An airport passenger whole-process customs clearance method based on face recognition technology

Abstract

The application discloses an airport passenger whole-process customs clearance method based on a face recognition technology, establishes a whole customs clearance platform system, constructs a cloud end and various independent airport nodes, forms a cloud end identity database of various process service points in the cloud end and forms a local identity database of various process service points in each airport node, combines passenger certificate information and face biological feature information to form unique identity information data during customs clearance, each service point collects on-site face photos, extracts face photo feature codes, uploads the face photo feature codes to the cloud end for comparison, perfects the cloud end identity database or sends the airport node for comparison and perfects the local identity database, and thus sequentially clears the whole process of check-in, security check and boarding. The platform system is a passenger identity information management and control center with unified service, unified standard and unified operation and maintenance, realizes real-time scheduling operation, and achieves the application target of "once registration, lifelong use, smooth travel and worry-free travel".

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of air passenger customs clearance, and particularly relates to an airport passenger whole-process customs clearance method based on face recognition technology. BACKGROUND

[0002] When passengers clear customs at an airport, they need to go through the processes of necessary check-in, security check, boarding and arrival at a station. In the traditional customs clearance process, passengers need to frequently show their certificates and boarding passes, for example, they need to show their valid identity certificates and boarding passes in the security check process, show their valid identity certificates in the check-in process and show their boarding passes in the boarding process. For passengers, the frequent showing of certificates and boarding passes is easy to cause loss and affect their mood, and for the airport, there is always the problem of low inspection efficiency.

[0003] Nowadays, the Internet of Everything has shown a great development trend. Although some airports and airlines have successively launched identity card and face comparison self-service security check, mobile one-code customs clearance and other solutions to facilitate passenger travel in combination with Internet technology, and some airports have launched single-airport one-face whole-process customs clearance solutions, defects are inevitable.

[0004] 1. The identity card customs clearance has gradually fallen behind, and the mobile one-code customs clearance needs to frequently show the mobile phone and has the risk of being screenshot and stolen;

[0005] 2. The face recognition only exists in individual service points, the coverage is low, and the whole-process customs clearance cannot be realized;

[0006] 3. The single-airport one-face whole-process customs clearance only covers a single airport, passengers cannot enjoy the same service at other airports, or passengers need to frequently register to enjoy the same service at different airports;

[0007] 4. The airport passenger flow in different places is different, and can be divided into large, medium and small airports. The airport passenger flow of a large airport such as Beijing Airport is more than 100 million per year, and the airport passenger flow of a small airport is only a few hundred thousand per year. The information database configuration required by such a large passenger flow gap needs to be different in different places.

[0008] To overcome the above problems, the present application is thus created. SUMMARY

[0009] The purpose of the present application is to provide an airport passenger whole-process customs clearance method based on face recognition technology, and to build a unified, fast, convenient and safe passenger unique identity recognition "platform system" for travel in large and small airports across the country.

[0010] To achieve the above purpose, the solution of the present application is: an airport passenger whole-process customs clearance method based on face recognition technology, comprising the following steps:

[0011] S1: Establish an airport passenger whole-process customs clearance platform system, which includes a cloud and airport nodes, and each airport is an independent airport node;

[0012] S2: First, the passenger registers on the client, the client collects passenger certificate information and face live photo to the cloud, the cloud extracts the face live photo feature code, combines the feature code with the certificate information, and forms a passenger identity information library;

[0013] S3: The cloud interfaces with the booking system and the departure system, compares the booking system and the departure system with the passenger identity information library, filters out the passenger flight information to be taken, and the flight information includes booking information, check-in information, security information and boarding information;

[0014] S4: The cloud extracts the corresponding passenger identity information from the passenger identity information library according to the filtered flight information, and the cloud divides the extracted passenger identity information according to the departure airport and the arrival airport, and each forms a cloud passenger travel big library, wherein the departure airport is further divided into a cloud check-in identity library, a cloud security identity library and a cloud boarding identity library; the arrival airport forms a cloud arrival identity library;

[0015] S5: The cloud distributes the cloud passenger travel big library, the cloud check-in identity library, the cloud security identity library and the cloud boarding identity library to the corresponding departure airport node according to the flight information, each departure airport node stores itself to form a local passenger travel big library, a local check-in identity library, a local security identity library and a local boarding identity library, and each check-in service point, security service point and boarding service point of the departure airport is connected to the corresponding departure airport node; the cloud distributes the cloud arrival identity library to the corresponding arrival airport node according to the flight information, forms a local arrival identity library, and each arrival airport arrival service point is connected to the corresponding arrival airport node;

[0016] S6: The initial state of the passenger is the check-in state; the check-in service point collects the on-site face photo and extracts the face photo feature code, then the check-in service point uploads the on-site face photo feature code to the cloud or the airport node,

[0017] If uploaded to the cloud, the cloud will identify and authenticate the feature code with the cloud check-in identity library to get the authentication result, if the authentication result is inconsistent, the passenger cannot check in, if the authentication result is consistent, the passenger will check in, complete the check-in, the cloud synchronously updates the passenger state to the check-in state, and the cloud migrates the passenger identity information from the cloud check-in identity library to the cloud security identity library;

[0018] If the airport node is uploaded, the airport node performs identity recognition authentication on the feature code with the local check-in identity library, obtains an authentication result, if the authentication result is inconsistent, the passenger cannot check in, if the authentication result is consistent, the passenger performs check-in operation, completes check-in, the airport node returns the passenger check-in state to the cloud, and migrates the passenger identity information from the local check-in identity library to the local security identity library; the cloud synchronously updates the passenger state to the check-in state, and the cloud migrates the passenger identity information from the cloud check-in identity library to the cloud security identity library;

[0019] S7: The security service point collects the on-site face photo, extracts the face photo feature code, and then uploads the on-site face photo feature code to the cloud or the airport node,

[0020] If the cloud is uploaded, the cloud performs identity recognition authentication on the feature code with the cloud security identity library, obtains an authentication result, if the authentication result is inconsistent, the passenger cannot be security checked, if the authentication result is consistent, the passenger performs security check operation, completes security check, the cloud synchronously updates the passenger state to the security check state, and the cloud migrates the passenger identity information from the cloud security identity library to the cloud boarding identity library;

[0021] If the airport node is uploaded, the airport node performs identity recognition authentication on the feature code with the local security identity library, obtains an authentication result, if the authentication result is inconsistent, the passenger cannot be security checked, if the authentication result is consistent, the passenger performs security check operation, completes security check, the airport node returns the passenger security check state to the cloud, and migrates the passenger identity information from the local security identity library to the local boarding identity library; the cloud synchronously updates the passenger state to the security check state, and the cloud migrates the passenger identity information from the cloud security identity library to the cloud boarding identity library;

[0022] S8: The boarding service point collects the on-site face photo, extracts the face photo feature code, and then uploads the on-site face photo feature code to the cloud or the airport node,

[0023] If the cloud is uploaded, the cloud performs identity recognition authentication on the feature code with the cloud boarding identity library, obtains an authentication result, if the authentication result is inconsistent, the passenger cannot board, if the authentication result is consistent, the passenger performs boarding operation, completes boarding, the cloud synchronously updates the passenger state to the boarding state, and the cloud clears the passenger identity information from the cloud boarding identity library;

[0024] If the airport node is uploaded, the airport node will identify the feature code with the local boarding identity library, get the authentication result, if the authentication result is inconsistent, the passenger cannot board, if the authentication result is consistent, the passenger performs boarding operation, completes boarding, and the airport node returns the passenger boarding state to the cloud, clears the passenger identity information from the local boarding identity library, the cloud synchronously updates the passenger state to the boarding state, and the cloud clears the passenger identity information from the cloud boarding identity library;

[0025] S9: Collecting the on-site face photo at the arrival service point of the airport, extracting the face photo feature code, uploading the on-site face photo feature code to the cloud or the airport node by the arrival service point,

[0026] If uploaded to the cloud, the cloud will identify the feature code with the cloud arrival identity library, if the authentication is consistent, the passenger completes the arrival and completes the trip, the cloud synchronously updates the passenger state to the arrival state, and clears the passenger identity information from the cloud arrival identity library;

[0027] If uploaded to the airport node, the airport node will identify the feature code with the local arrival identity library, if the authentication is consistent, the passenger completes the arrival and completes the trip, the airport node returns the passenger arrival state to the cloud, clears the passenger identity information from the local arrival identity library, the cloud synchronously updates the passenger state to the arrival state, and the cloud clears the passenger identity information from the cloud arrival identity library.

[0028] As a further improvement, in step S8, if the flight information also includes transfer information / stopover information, after boarding, steps S8.1 and S8.2 are entered, otherwise step S9 is entered;

[0029] S8.1: The cloud extracts the corresponding passenger identity information from the passenger identity information library according to the transfer information / stopover information, the cloud forms a cloud transfer / stopover identity library according to the transfer airport / stopover airport, and the cloud sends the passenger boarding state to the transfer / stopover airport node, and the transfer / stopover airport node forms a local transfer / stopover identity library after receiving;

[0030] S8.2: The cloud updates the passenger boarding state to the standby transfer / stopover state; the transfer / stopover service point of the transfer / stopover airport node collects the on-site face photo, extracts the face photo feature code, and uploads the on-site face photo feature code to the cloud or the airport node by the transfer / stopover service point,

[0031] If uploaded to the cloud, the cloud will identify and authenticate the feature code with the cloud transfer / transit identity library to obtain an authentication result. If the authentication result is inconsistent, the passenger cannot transfer / transit. If the authentication result is consistent, the passenger completes the transfer / transit and performs the boarding operation again. Repeat step S8. The cloud synchronously updates the passenger state to the boarding state. The cloud clears the passenger identity information from the cloud transfer / transit identity library.

[0032] If uploaded to the airport node, the airport node will identify and authenticate the feature code with the local transfer / transit identity library to obtain an authentication result. If the authentication result is inconsistent, the passenger cannot transfer / transit. If the authentication result is consistent, the passenger completes the transfer / transit and performs the boarding operation again. Repeat step S8. The airport node returns the passenger boarding state to the cloud. The cloud synchronously updates the passenger state to the boarding state and clears the passenger identity information from the local transfer / transit identity library. The airport node clears the passenger identity information from the cloud transfer / transit identity library.

[0033] As a further improvement, in step S7, the security service point identifies the passenger who has passed the inspection and displays the electronic inspection stamp in the system of the security service point for the staff to view.

[0034] As a further improvement, in step S8, before the passenger boards, the boarding service point re-inspects the passenger and displays the electronic inspection stamp in the system of the boarding service point for the staff to view.

[0035] As a further improvement, in step S8, after the passenger boards, the passenger needs to pass through the corridor bridge. The corridor bridge service point re-inspects the passenger and displays the electronic inspection stamp in the system of the corridor bridge service point for the staff to view.

[0036] After adopting the above scheme, the gain effect of the present application is that:

[0037] 1. The present application establishes a full customs platform system. The platform system constructs the cloud and various independent airport nodes and forms the cloud identity library of each process service point in the cloud and the local identity library of each process service point in each airport node. When customs clearance, the passenger certificate information and the face biometric information are combined to form unique identity information data. Each service point collects the on-site face photo, extracts the face photo feature code, uploads the cloud for comparison, perfects the cloud identity library, or sends the airport node for comparison and perfects the local identity library, so as to sequentially clear the full process of check-in, security check, boarding, etc.

[0038] 2. Based on passenger traffic volume, airports are categorized into large, medium, and small airports. This invention is applicable to any large, medium, or small airport, and it not only covers a single airport but can also connect multiple airports. Large and medium-sized airports can purchase powerful servers to build local passenger identification service branch systems for passenger verification, while small airports, with limited capabilities, can choose to directly connect to the cloud for passenger verification. In this way, small airports do not need to spend a lot of money purchasing server equipment and building systems.

[0039] 3. The platform system of this invention is a unified service, unified standard, and unified operation and maintenance passenger identity information management center. It performs real-time scheduling and calculation to provide corresponding services to various service touchpoints in the airport. After registration, passengers can use facial recognition to identify themselves at any time and at any service touchpoint in any airport in China to complete the entire process of check-in, security check, boarding, and transfer / stopover. They can also quickly enjoy other services such as flight inquiry and VIP lounge by using facial recognition, realizing the application goal of "one-time registration, lifelong use, smooth journey, and worry-free travel". Attached Figure Description

[0040] Figure 1 This is a block diagram of information transmission in this invention;

[0041] Figure 2 This is a flowchart illustrating the information flow of each service point in this invention via cloud uploads.

[0042] Figure 3 This is a flowchart illustrating the information flow of each service point in this invention by uploading data to the local airport node. Detailed Implementation

[0043] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0044] This invention provides a method for airport passenger clearance throughout the entire process based on facial recognition technology, comprising the following steps:

[0045] S1: Establish an airport passenger end-to-end customs clearance platform system, such as Figure 1 As shown, the platform system includes a cloud and airport nodes, with each airport being an independent airport node;

[0046] S2: First, passengers register their identity information on the passenger client using their valid ID and a live facial photo, and authorize the client and platform system to query their itinerary and provide identity verification services using their ID information and facial photo. The client collects the passenger's ID information and live facial photo to the cloud, and the cloud extracts the facial live photo's feature code, combining the feature code with the ID information to form a passenger identity information database. Passengers can manage their own ID information and photo information according to their own circumstances to obtain better identity verification services.

[0047] S3: The cloud scheduling engine interfaces the booking system and the departure system (taking the CAAC booking and departure system as an example), compares the booking system and the departure system with the passenger identity information library, filters out the passenger flight information to be taken, and the flight information includes booking information, check-in information, security information, boarding information and other information, etc.;

[0048] S4: The cloud extracts the corresponding passenger identity information from the passenger identity information library according to the filtered flight information, and the cloud divides the extracted passenger identity information according to the departure airport and the arrival airport, and each of the departure airport and the arrival airport forms a cloud passenger travel big library, wherein the departure airport is further divided into a cloud check-in identity library, a cloud security identity library, and a cloud boarding identity library;

[0049] S5: The cloud distributes the cloud passenger travel big library, the cloud check-in identity library, the cloud security identity library, and the cloud boarding identity library to the corresponding departure airport node according to the flight information, each departure airport node stores itself to form a local passenger travel big library, a local check-in identity library, a local security identity library, and a local boarding identity library, and each check-in service point, security service point, and boarding service point of the departure airport is connected to the corresponding departure airport node; the cloud distributes the cloud arrival identity library to the corresponding arrival airport node according to the flight information to form a local arrival identity library, and each arrival airport arrival service point is connected to the corresponding arrival airport node; the cloud can provide passenger identity identification and authentication services for each service point of check-in, security, boarding, and arrival according to the subdivided passenger identity information library;

[0050] S6: The initial state of the passenger is the check-in state; the check-in service point deploys a binocular camera with face recognition technology to collect on-site face photos, extract face photo feature codes, and then upload the on-site face photo feature codes to the cloud or the airport node,

[0051] As shown in Figure 2 , if uploaded to the cloud, the cloud performs identity identification and authentication on the feature code and the cloud check-in identity library to obtain an authentication result, if the authentication result is inconsistent, the passenger cannot check in, if the authentication result is consistent, the passenger performs check-in operation, completes check-in, the cloud synchronously updates the passenger state to the check-in state, and the cloud migrates the passenger identity information from the cloud check-in identity library to the cloud security identity library;

[0052] As shown in Figure 3If the airport node is uploaded, the airport node performs identity recognition authentication on the feature code and the local check-in identity library, obtains an authentication result, if the authentication result is inconsistent, the passenger cannot check in, if the authentication result is consistent, the passenger performs check-in operation, completes check-in, the airport node returns the passenger check-in state to the cloud, and migrates the passenger identity information from the local check-in identity library to the local security identity library; the cloud synchronously updates the passenger state to the check-in state, and migrates the passenger identity information from the cloud check-in identity library to the cloud security identity library;

[0053] S7: Start security check, the security check service point deploys a binocular camera with human face recognition technology, collects an on-site face photo, extracts a face photo feature code, next, the security check service point uploads the on-site face photo feature code to the cloud or the airport node,

[0054] If the cloud is uploaded, the cloud performs identity recognition authentication on the feature code and the cloud security identity library, obtains an authentication result, if the authentication result is inconsistent, the passenger cannot pass security check, if the authentication result is consistent, the passenger performs security check operation, completes security check, the cloud synchronously updates the passenger state to the security check state, and the cloud migrates the passenger identity information from the cloud security identity library to the cloud boarding identity library;

[0055] If the airport node is uploaded, the airport node performs identity recognition authentication on the feature code and the local security identity library, obtains an authentication result, if the authentication result is inconsistent, the passenger cannot pass security check, if the authentication result is consistent, the passenger performs security check operation, completes security check, the airport node returns the passenger security check state to the cloud, and migrates the passenger identity information from the local security identity library to the local boarding identity library; the cloud synchronously updates the passenger state to the security check state, and migrates the passenger identity information from the cloud security identity library to the cloud boarding identity library;

[0056] S8: Start boarding, the boarding service point deploys a binocular camera with human face recognition technology, collects an on-site face photo, extracts a face photo feature code, next, the boarding service point uploads the on-site face photo feature code to the cloud or the airport node,

[0057] If the cloud is uploaded, the cloud performs identity recognition authentication on the feature code and the cloud boarding identity library, obtains an authentication result, if the authentication result is inconsistent, the passenger cannot board, if the authentication result is consistent, the passenger performs boarding operation, completes boarding, the cloud synchronously updates the passenger state to the boarding state, and the cloud clears the passenger identity information from the cloud boarding identity library;

[0058] If the airport node is uploaded, the airport node performs identity recognition authentication on the feature code with the local boarding identity library, obtains an authentication result, if the authentication result is inconsistent, the passenger cannot board, if the authentication result is consistent, the passenger performs boarding operation, completes boarding, the airport node returns the passenger boarding state to the cloud, and clears the passenger identity information from the local boarding identity library;

[0059] Some passengers are not direct, and the flight information also includes transfer information / stopover information, therefore, if the flight information also includes transfer information / stopover information, steps S8.1 and S8.2 are entered after boarding, otherwise, step S9 is entered;

[0060] S8.1: The cloud extracts corresponding passenger identity information from the passenger identity information library according to the transfer information / stopover information, the cloud forms a cloud transfer / stopover identity library according to the transfer airport / stopover airport, and the cloud sends the passenger boarding state to the transfer / stopover airport node, and the transfer / stopover airport node forms a local transfer / stopover identity library after receiving;

[0061] S8.2: The cloud updates the passenger boarding state to a waiting transfer / stopover state; start transfer / stopover, the transfer / stopover service point of the transfer / stopover airport node deploys a binocular camera with human face recognition technology, collects on-site human face photos, extracts human face photo feature codes, and the transfer / stopover service point uploads the on-site human face photo feature codes to the cloud or the airport node,

[0062] If the cloud is uploaded, the cloud performs identity recognition authentication on the feature code with the cloud transfer / stopover identity library, obtains an authentication result, if the authentication result is inconsistent, the passenger cannot transfer / stopover, if the authentication result is consistent, the passenger completes transfer / stopover, and performs boarding operation again, repeats step S8, the cloud synchronously updates the passenger state to a boarding state, and the cloud clears the passenger identity information from the cloud transfer / stopover identity library;

[0063] If the airport node is uploaded, the airport node performs identity recognition authentication on the feature code with the local transfer / stopover identity library, obtains an authentication result, if the authentication result is inconsistent, the passenger cannot transfer / stopover, if the authentication result is consistent, the passenger completes transfer / stopover, and performs boarding operation again, repeats step S8, the airport node returns the passenger boarding state to the cloud, and clears the passenger identity information from the local transfer / stopover identity library; the cloud synchronously updates the passenger state to a boarding state, and clears the passenger identity information from the cloud transfer / stopover identity library;

[0064] S9: The cloud sends the passenger boarding state to the arrival airport node, and the arrival airport forms a cloud arrival identity library;

[0065] After the plane lands, prepare to leave the station, and the arrival service point of the airport deploys binocular cameras with human face recognition technology, collects on-site face photos, extracts face photo feature codes, and uploads the on-site face photos to the cloud or the airport node,

[0066] If uploaded to the cloud, the cloud will identify and authenticate the feature code with the cloud arrival identity library, if the authentication is consistent, the passenger completes the arrival, completes the trip, and the cloud synchronously updates the passenger state to the arrival state; the passenger identity information is cleared from the cloud arrival identity library;

[0067] If uploaded to the airport node, the airport node will identify and authenticate the feature code with the local arrival identity library, if the authentication is consistent, the passenger completes the arrival, completes the trip, and the airport node returns the passenger arrival state to the cloud, and clears the passenger identity information from the local arrival identity library; the cloud synchronously updates the passenger state to the arrival state, and clears the passenger identity information from the cloud arrival identity library.

[0068] The present application also includes a security check review procedure after the security check procedure,

[0069] Firstly, in step S7, (1) the security check system checks the passenger according to the existing security check process, ensures the credibility and uniqueness of the passenger's identity, the accuracy of the passenger's flight information, and the accuracy of other security check states (the consistency of the three of the person, the certificate and the bill), realizes paperless customs clearance; (2) the security check service point identifies the passenger who has passed the check, and displays the electronic inspection stamp in the system of the security check service point for the staff to check; (3) the airport node migrates the passenger identity information from the security check identity library to the boarding identity library and the security check review identity library; (4) the airport node returns the passenger check state to the cloud.

[0070] Then, in step S8, the review procedure is set up for the passengers at the boarding gate and the corridor bridge / remote stand, respectively, (1) when the passenger boards at the boarding gate / corridor bridge / remote stand, the airport staff needs to confirm the check state of the passenger again; (2) the service point reviews the security check of the passenger, if the passenger has passed the check, the electronic inspection stamp is displayed in the system of the service point for the staff to check, and the verification state information is returned to the airport node, and the security check state review is completed.

[0071] The security check service point of the present application adds millimeter wave inspection information matching with passenger information, specifically:

[0072] (1) the security check service point system and the millimeter wave security check equipment establish a communication interface to realize real-time interaction of passenger scanning data;

[0073] (2) The millimeter wave device is provided with an integrated machine (including a face capturing camera), and when the passenger enters the millimeter wave device, the on-site face photo of the passenger is collected, the airport node is requested to perform identity recognition authentication, and the passenger information is located according to the recognition authentication result;

[0074] (3) If the airport node cannot perform identity recognition (for example, the passenger wears a mask, does not look at the camera, etc.), the passenger information can be located by scanning the boarding pass and valid flight certificate of the passenger;

[0075] (4) After the system locates the passenger information, the millimeter wave device pushes the scanning result, and the security system receives the result queue, the receiving time point, the passenger information locating time point and other conditions, and binds the passenger information and the millimeter wave result one by one, finally realizes the millimeter wave image and the person image binding, and presents the result to the on-site staff in real time.

[0076] Before the establishment of the platform system, the airport is divided into large and medium-sized airports and small airports according to the size of the passenger flow. The large and medium-sized airports can purchase servers with strong processing capacity to build a local passenger identity recognition service branch system for passenger identity verification, while the small airports have limited capacity and can choose to connect the cloud directly for passenger identity verification. In this way, the small airport does not need to spend a lot of cost to purchase server equipment to build a system.

[0077] Figure 1 After the passenger completes the identity information registration, the cloud forms a basic database of the cloud passenger identity information library to store passenger certificate information, passenger face pictures, passenger face picture feature codes and the like. At the same time, the cloud forms a rule library through business settings. The cloud receives the passenger flight information transmitted by the China Eastern Airlines booking and departure system through the interface platform, the scheduling engine connects the rule library, and performs scheduling operation through rule setting. After the scheduling operation is completed, the passenger identity information is extracted from the cloud and subdivided into cloud check-in identity library, security identity library, transfer / stopover identity library, boarding identity library and arrival identity library. At the same time, the cloud transmits / receives module (commonly known as: receiving Q cluster) transmits / receives the passenger identity information and flight state information to the airport interface platform of the large and medium-sized airport node. After the airport interface platform receives the passenger identity information and flight state information, it is analyzed and stored to form the passenger identity information library of the airport node. The cloud connects the airport service points of the small airport node through authorization authentication. The airport node service connects the airport service points through authorization authentication; after the airport node service points complete the service, the passenger flight state is returned to the cloud through the airport interface platform, and the cloud is updated synchronously after receiving.

[0078] The above only describes the preferred embodiment of the present application, and is not a limitation on the design of the present application. Any equivalent changes made according to the key design of the present application fall within the scope of protection of the present application.

Claims

1. An airport passenger whole-process customs clearance method based on face recognition technology, characterized in that: The method comprises the following steps: S1: establishing an airport passenger whole-process customs clearance platform system, the platform system comprising a cloud end and airport nodes, each airport being an independent airport node; S2: first, a passenger registers on a client, the client collecting passenger certificate information and a face live photo to the cloud end, the cloud end extracting a face live photo feature code, combining the feature code with the certificate information to form a passenger identity information library; S3: the cloud end interfaces with a booking system and a departure system, compares the booking system and the departure system with the passenger identity information library, screens and filters passenger flight information to be taken, the flight information comprising booking information, check-in information, security check information and boarding information; S4: the cloud end extracts corresponding passenger identity information from the passenger identity information library according to the screened and filtered flight information, the cloud end divides the passenger identity information according to a departure airport and an arrival airport, the departure airport and the arrival airport each forming a cloud end passenger travel big library, wherein the departure airport is further divided to form a cloud end check-in identity library, a cloud end security check identity library and a cloud end boarding identity library; the arrival airport forms a cloud end arrival identity library; S5: the cloud end distributes the cloud end passenger travel big library, the cloud end check-in identity library, the cloud end security check identity library and the cloud end boarding identity library to the departure airport nodes of corresponding flights according to the flight information, each departure airport node stores the corresponding information to form a local passenger travel big library, a local check-in identity library, a local security check identity library and a local boarding identity library, and each check-in service point, security check service point and boarding service point of the departure airport is connected to the corresponding departure airport node; the cloud end distributes the cloud end arrival identity library to the arrival airport nodes of corresponding flights according to the flight information to form a local arrival identity library, and each arrival service point of the arrival airport is connected to the corresponding arrival airport node; S6: the initial state of a passenger is a check-in state; a check-in service point collects a live face photo, extracts a face photo feature code, and then the check-in service point uploads the live face photo feature code to the cloud end or the airport node, if the cloud end, the cloud end performs identity recognition and authentication on the feature code and the cloud end check-in identity library to obtain an authentication result, if the authentication result is inconsistent, the passenger cannot check in, if the authentication result is consistent, the passenger performs check-in operation, completes check-in, the cloud end synchronously updates the passenger state to a check-in state, and the cloud end migrates the passenger identity information from the cloud end check-in identity library to the cloud end security check identity library; if the airport node, the airport node performs identity recognition and authentication on the feature code and the local check-in identity library to obtain an authentication result, if the authentication result is inconsistent, the passenger cannot check in, if the authentication result is consistent, the passenger performs check-in operation, completes check-in, the airport node returns the passenger check-in state to the cloud end, migrates the passenger identity information from the local check-in identity library to the local security check identity library, the cloud end synchronously updates the passenger state to a check-in state, and migrates the passenger identity information from the cloud end check-in identity library to the cloud end security check identity library; S7: a security check service point collects a live face photo, extracts a face photo feature code, and then the security check service point uploads the live face photo feature code to the cloud end or the airport node, If uploaded to the cloud, the cloud will identify and authenticate the feature code with the cloud security identity library to obtain an authentication result. If the authentication result is inconsistent, the passenger cannot pass security. If the authentication result is consistent, the passenger proceeds with security operations, completes security, and the cloud synchronously updates the passenger state to a security state. The cloud migrates the passenger identity information from the cloud security identity library to the cloud boarding identity library; If uploaded to the airport node, the airport node will identify and authenticate the feature code with the local security identity library to obtain an authentication result. If the authentication result is inconsistent, the passenger cannot pass security. If the authentication result is consistent, the passenger proceeds with security operations, completes security, and the airport node returns the passenger security state to the cloud. The airport node migrates the passenger identity information from the local security identity library to the local boarding identity library. The cloud synchronously updates the passenger state to a security state and migrates the passenger identity information from the cloud security identity library to the cloud boarding identity library; S8: The boarding service point collects the on-site face photo, extracts the face photo feature code, and then uploads the on-site face photo feature code to the cloud or the airport node, If uploaded to the cloud, the cloud will identify and authenticate the feature code with the cloud boarding identity library to obtain an authentication result. If the authentication result is inconsistent, the passenger cannot board. If the authentication result is consistent, the passenger proceeds with boarding operations, completes boarding, and the cloud synchronously updates the passenger state to a boarding state. The cloud clears the passenger identity information from the cloud boarding identity library; If uploaded to the airport node, the airport node will identify and authenticate the feature code with the local boarding identity library to obtain an authentication result. If the authentication result is inconsistent, the passenger cannot board. If the authentication result is consistent, the passenger proceeds with boarding operations, completes boarding, and the airport node returns the passenger boarding state to the cloud. The airport node clears the passenger identity information from the local boarding identity library. The cloud synchronously updates the passenger state to a boarding state and clears the passenger identity information from the cloud boarding identity library; S9: The arrival service point at the airport collects the on-site face photo, extracts the face photo feature code, and uploads the on-site face photo feature code to the cloud or the airport node, If uploaded to the cloud, the cloud will identify and authenticate the feature code with the cloud arrival identity library. If the authentication is consistent, the passenger completes arrival and completes the trip. The cloud synchronously updates the passenger state to an arrival state and clears the passenger identity information from the cloud arrival identity library; If uploaded to the airport node, the airport node will identify and authenticate the feature code with the local arrival identity library. If the authentication is consistent, the passenger completes arrival and completes the trip. The airport node returns the passenger arrival state to the cloud and clears the passenger identity information from the local arrival identity library. The cloud synchronously updates the passenger state to an arrival state and clears the passenger identity information from the cloud arrival identity library.

2. The airport passenger whole-process customs clearance method based on face recognition technology according to claim 1, characterized in that: In step S8, if the flight information also includes transfer information / stopover information, proceed to steps S8.1 and S8.2 after boarding, otherwise proceed to step S9; S8.1: The cloud extracts corresponding passenger identity information from the passenger identity information database according to the transfer information / stopover information. The cloud forms a cloud transfer / stopover identity database according to the transfer airport / stopover airport, and sends the passenger boarding state to the transfer / stopover airport node. After receiving, the transfer / stopover airport node forms a local transfer / stopover identity database; S8.2: The cloud updates the passenger boarding state to the transfer / stopover state. The transfer / stopover service point of the transfer / stopover airport node collects the on-site face photo, extracts the face photo feature code, and uploads the on-site face photo feature code to the cloud or the airport node, If uploaded to the cloud, the cloud performs identity recognition and authentication on the feature code and the cloud transfer / stopover identity database to obtain an authentication result. If the authentication result is inconsistent, the passenger cannot transfer / stopover. If the authentication result is consistent, the passenger completes the transfer / stopover and performs the boarding operation again. Repeat step S8. The cloud synchronously updates the passenger state to the boarding state, and the cloud clears the passenger identity information from the cloud transfer / stopover identity database; If uploaded to the airport node, the airport node performs identity recognition and authentication on the feature code and the local transfer / stopover identity database to obtain an authentication result. If the authentication result is inconsistent, the passenger cannot transfer / stopover. If the authentication result is consistent, the passenger completes the transfer / stopover and performs the boarding operation again. Repeat step S8. The airport node returns the passenger boarding state to the cloud, and the cloud synchronously updates the passenger state to the boarding state and clears the passenger identity information from the local transfer / stopover identity database. The airport node clears the passenger identity information from the cloud transfer / stopover identity database.

3. The airport passenger whole-process customs clearance method based on face recognition technology according to claim 1, characterized in that: In step S7, the security service point identifies the passenger who has passed the inspection, and displays the electronic inspection stamp in the system of the security service point for the staff to check.

4. The airport passenger whole-process customs clearance method based on face recognition technology according to claim 3, characterized in that: In step S8, before boarding, the boarding service point rechecks the passenger, and displays the electronic inspection stamp in the system of the boarding service point for the staff to check.

5. The airport passenger whole-process customs clearance method based on face recognition technology according to claim 3, characterized in that: In step S8, after boarding, the passenger still needs to pass through the corridor bridge. The corridor bridge service point rechecks the passenger, and displays the electronic inspection stamp in the system of the corridor bridge service point for the staff to check.

Images