Method for connecting face recognition payment device with merchant cash register and payment device

Abstract

The present application provides a method for connecting a face-scanning payment device to a merchant's cash register and a payment device, which are applied to the field of quick payment technology. The method includes: the payment device obtains merchant information and loads first connection data; after receiving the connection permission from the merchant's cash register, wirelessly connects with the merchant's cash register according to the first connection data; builds a local database, and the local database is used to store pictures and corresponding ID accounts; when the Internet connection of the payment device is unstable, activates the local database, obtains the customer's facial picture, and determines whether the facial picture is stored in the local database. If so, directly deducts money from the ID account corresponding to the facial picture; if not, collects the customer's facial picture through a preset four-dimensional model and enters it into the local database, and then obtains the customer's ID account entered by the merchant's cash register, and completes the deduction after the customer enters the password corresponding to the ID account for the first time.

Description

Technical Field

[0001] The present application relates to the field of quick payment technology, and in particular to a method for connecting a face-scanning payment device with a merchant's cash register and a payment device. Background Art

[0002] As for the connection method between payment devices and merchant cashier counters in the field of face-scanning payment, the existing technology commonly used in the industry is that the payment device is transferred to the merchant cashier counter through WeChat, Alipay or banks and other platforms to achieve connection. The specific performance method is: the payment device obtains the customer's face and the price of the product, determines the customer ID according to the customer's face through the database of WeChat, Alipay or banks and other platforms, and deducts the value consistent with the product price from the account corresponding to the customer ID. After WeChat, Alipay or banks and other platforms obtain a partial dividend of the product price, the remaining value is transferred to the merchant cashier counter;

[0003] This docking method currently used in the industry can effectively solve the storage problem of the facial database, so that there is no need to configure a docking database for the payment device, and the face can be stored through the platform; but it also has disadvantages: (1) The docking is unstable, because the payment device needs to be connected to the Internet to exchange data with the platform, and the merchant's cash register also needs to be connected to the Internet to exchange data with the platform. When the communication signal is unstable, the payment device and the merchant's cash register will face the problem of cash register crash; (2) After determining the account corresponding to the customer ID, the customer needs to enter the password before paying, because the accuracy of facial recognition is determined, but this goes against the original intention of quick payment. The payment speed is not as convenient as the fingerprint payment of the mobile phone, and the user experience is not high. Summary of the invention

[0004] The present application aims to solve the problem of unstable connection between payment devices and merchant cash registers and platforms in the prior art, as well as the technical problem that face recognition requires password input, resulting in slow payment speed and poor user experience, and to provide a method for connecting a face recognition payment device with a merchant cash register and a payment device.

[0005] This application adopts the following technical means to solve the technical problem:

[0006] This application proposes a method for connecting a face-scanning payment device with a merchant cash register, comprising:

[0007] The payment device obtains the merchant information and loads the first docking data;

[0008] After receiving the connection permission from the merchant's cashier counter, wirelessly connect with the merchant's cashier counter according to the first connection data;

[0009] Building a local database, the local database is used to store pictures and corresponding ID accounts;

[0010] When the Internet connection of the payment device is unstable, the local database is activated to obtain the customer's facial image, and it is determined whether the facial image is stored in the local database. If so, the payment is directly deducted from the ID account corresponding to the facial image; if not, the customer's facial image is collected through a preset four-dimensional model and entered into the local database, and then the customer's ID account entered at the merchant's cashier is obtained, and the deduction is completed after the customer enters the password corresponding to the ID account for the first time;

[0011] When the payment device has a stable Internet connection, the Internet database and the local database are activated to obtain the customer's facial picture, and determine whether the facial picture is stored in the local database. If so, the money is directly deducted from the ID account corresponding to the facial picture; if not, the customer's facial picture is collected through a preset four-dimensional model and entered into the local database, and then the ID account corresponding to the facial picture is retrieved from the Internet database. The deduction is completed after the customer enters the password corresponding to the ID account for the first time.

[0012] Further, when the Internet connection of the payment device is unstable, the steps of directly deducting money from the ID account corresponding to the facial image and / or completing the deduction after the customer enters the password corresponding to the ID account for the first time include:

[0013] Recording the ID account, the deduction value and confidential data, wherein the confidential data is the user's permission to acknowledge the deduction, which is created when the facial image is determined in the local data and / or when the customer enters the password corresponding to the ID account;

[0014] Monitor whether the current Internet speed reaches the minimum standard speed;

[0015] If so, the deduction will be made through the ID account, deduction amount and confidential data.

[0016] Furthermore, after the step of monitoring whether the current Internet speed reaches the minimum standard speed, the following steps are included:

[0017] If not, temporarily disconnect the first docking data;

[0018] Amplify the communication network signal and deduct money through ID account, deduction value and confidential data;

[0019] The increase is suspended, and the wireless connection with the merchant's cashier counter through the first connection data is restored, and the deduction data after the deduction is entered into the merchant's cashier counter.

[0020] Furthermore, when the Internet connection of the payment device is stable, the method of calibrating the four-dimensional model includes:

[0021] Retrieving facial images of customers from a local database and obtaining facial images of customers from an Internet database;

[0022] Compare the two facial images in the local database and the Internet database to obtain a face shape deviation tolerance interval;

[0023] The facial image verification accuracy of the customer is calibrated for the four-dimensional model through the facial shape deviation tolerance interval.

[0024] Furthermore, the method for collecting the facial image of the customer using the four-dimensional model comprises:

[0025] Collect a preset number of frames of customer facial images;

[0026] Construct the facial image corresponding to each frame in three-dimensional coordinates;

[0027] Compare the X, Y, and Z axis differences of the facial image of each frame's three-dimensional coordinates to obtain an accurate facial image.

[0028] Furthermore, the four-dimensional model is a pre-trained neural network model, and its training method includes:

[0029] Build and associate three-dimensional coordinates with the time axis to form the surface of the four-dimensional model;

[0030] Importing a facial video as training data into a four-dimensional model, disassembling the duration of the facial video and importing it into a timeline, and obtaining a facial image of each frame of the facial video after disassembly;

[0031] Loading each frame of the facial image into three-dimensional coordinates one by one, and performing image recognition on the facial image to determine a number of marking points in the facial image through image recognition, and then recording the positions of the marking points;

[0032] Performing time progression according to the time axis, mapping the facial images of the previous frame and the next frame, connecting the facial images of the previous frame and the next frame at the same marked point of the three-dimensional coordinates through the mapping, and after the time progression of the time axis is completed, establishing a 4D image of the portrait in the facial video;

[0033] A local database is established, an inner layer of a four-dimensional model is constructed, and the 4D image is stored in the local database.

[0034] The present application also provides a payment device, including:

[0035] The merchant unit is used for the payment device to obtain merchant information and load the first docking data;

[0036] A docking unit, configured to wirelessly dock with the merchant's cashier counter according to the first docking data after receiving the docking permission from the merchant's cashier counter;

[0037] A local storage unit, used to construct a local database, wherein the local database is used to store pictures and corresponding ID accounts;

[0038] The first payment channel unit is used to activate the local database when the Internet connection of the payment device is unstable, obtain the customer's facial image, and determine whether the facial image is stored in the local database. If so, directly deduct money from the ID account corresponding to the facial image; if not, collect the customer's facial image through a preset four-dimensional model and enter it into the local database, and then obtain the customer's ID account entered by the merchant's cashier, and complete the deduction after the customer enters the password corresponding to the ID account for the first time;

[0039] The second payment channel unit is used to activate the Internet database and the local database when the Internet connection of the payment device is stable, obtain the customer's facial picture, and determine whether the facial picture is stored in the local database. If so, deduct money directly from the ID account corresponding to the facial picture; if not, collect the customer's facial picture through a preset four-dimensional model and enter it into the local database, and then retrieve the ID account corresponding to the facial picture from the Internet database. The deduction is completed after the customer enters the password corresponding to the ID account for the first time.

[0040] Furthermore, the first payment channel unit includes:

[0041] a recording module for recording the ID account, the deduction value and confidential data, wherein the confidential data is the user's permission to acknowledge the deduction, which is created when the facial image is determined in the local data and / or when the customer enters the password corresponding to the ID account;

[0042] The network speed monitoring module is used to monitor whether the current Internet speed reaches the minimum standard speed;

[0043] The first deduction unit is used to deduct money through the ID account, deduction value and confidential data.

[0044] Furthermore, the first payment channel unit includes:

[0045] a disconnection module, for temporarily disconnecting the first docking data if no;

[0046] The second deduction module is used to amplify the communication network signal and deduct money through the ID account, deduction value and confidential data;

[0047] The recovery module is used to suspend the increase, resume the wireless connection with the merchant's cashier through the first connection data, and enter the deduction data after the deduction into the merchant's cashier.

[0048] The present application also proposes a face-scanning payment system, including a payment device, a merchant cashier counter, an Internet database and a management center, wherein the payment device includes a local database;

[0049] The payment device is used to obtain a facial image of the customer, determine an ID account corresponding to the facial image from a preset local database, and / or determine an ID account corresponding to the facial image from an Internet database, and the payment device is connected to a merchant cashier counter to transfer the deduction in the ID account to the merchant cashier counter;

[0050] The merchant cashier counter is used to connect with the payment device to collect payment;

[0051] The Internet database is used to store the customer's facial image, and is connected to the Internet database when the payment device network is stable;

[0052] The management center is used to record the registration, login and management of merchant cashiers through payment devices.

[0053] This application provides a method for connecting a face-scanning payment device with a merchant cash register and a payment device, which has the following beneficial effects:

[0054] The payment device obtains merchant information and loads first docking data, which is used for merchant registration and login; after receiving the docking permission from the merchant cashier, wireless docking is performed with the merchant cashier according to the first docking data to realize wireless docking between the payment device and the merchant cashier; a local database is constructed, and the local database is used to store pictures and corresponding ID accounts; when the Internet connection of the payment device is unstable, the local database is activated, the customer's facial picture is obtained, and it is determined whether the facial picture is stored in the local database. If so, the payment is directly deducted from the ID account corresponding to the facial picture; if not, the customer's facial picture is collected through a preset four-dimensional model and entered into the local database, and then the customer's ID account entered by the merchant cashier is obtained. The customer's first The deduction is completed after the password corresponding to the ID account is entered; when the Internet connection of the payment device is stable, the Internet database and the local database are activated, the customer's facial picture is obtained, and it is determined whether the facial picture is stored in the local database. If so, the deduction is made directly from the ID account corresponding to the facial picture; if not, the customer's facial picture is collected through a preset four-dimensional model and entered into the local database, and then the ID account corresponding to the facial picture is retrieved from the Internet database. The deduction is completed after the customer enters the password corresponding to the ID account for the first time, thereby solving the problem of unstable docking between payment devices and merchant cash registers and platforms in the prior art, as well as the process of face recognition requiring password entry, resulting in slow payment speed and poor user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a flow chart of the first embodiment of the method for connecting a face-scanning payment device with a merchant cash register in this application;

[0056] Figure 2This is a flow chart of a second embodiment of the method for connecting a face-scanning payment device with a merchant cash register in this application;

[0057] Figure 3 This is a block diagram of the payment device for this application;

[0058] Figure 4 This is the structural block diagram of the face-scanning payment system for this application.

[0059] The implementation, functional features and advantages of the present application will be further described in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0060] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0061] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0062] It should be noted that the terms "include", "comprises" and "have" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices. In the claims, specification and drawings of the present application, relational terms such as "first" and "second" are merely used to distinguish one entity / operation / object from another entity / operation / object, and do not necessarily require or imply any such actual relationship or order between these entities / operations / objects.

[0063] The present application preferably uses a face-scanning payment device as the execution subject, and executes a method for connecting the face-scanning payment device with a merchant's cash register through the face-scanning payment device (hereinafter referred to as the payment device).

[0064] Reference Figure 1 , is a flow chart of a method for connecting a face-scanning payment device with a merchant's cash register in the first embodiment of the present application;

[0065] A method for connecting a face-scanning payment device with a merchant cash register, comprising:

[0066] S100, the payment device obtains merchant information and loads first docking data;

[0067] Merchant information such as business license, store owner's identity information and telephone number, etc. In addition, after the payment device obtains the merchant information, it sends the merchant information to the management center. After approval by the management center, the face-scanning payment function of the above payment device is activated.

[0068] S200, after receiving the docking permission from the merchant cashier, wirelessly docking with the merchant cashier according to the first docking data;

[0069] It can be understood that the docking permission can be a protocol docking of wireless data signals such as Bluetooth connection. After the payment device receives the docking permission, the payment device and the merchant's cash register are wirelessly connected.

[0070] S300, building a local database, the local database is used to store pictures and corresponding ID accounts;

[0071] A local database is built in the payment device, and the local database is used to store the customer's facial images.

[0072] At this step, the intention of the present application is explained. A local database is set in the payment device. The local database can only store facial images of a certain memory capacity. The existing storage medium is generally 8G, 16G, 32G, 64G,..., 512G of memory. The storage medium is installed in the payment device to record the customer's facial image. In another embodiment, in order to clear the memory in time, when a facial image in the local database has not been verified for a preset time (verification is that the payment device obtains the customer's facial image and verifies the ID account corresponding to the facial image from the local database), the facial image is deleted. The above preset time is preferably one year.

[0073] In essence, in the process of merchants selling goods, merchants are regional, and the local database can store facial images of some frequent customers. During the sales process, the payment device will compare the acquired customer facial image with the corresponding facial image stored in the local database to determine the ID account. When they are consistent, the payment will be directly deducted without the customer having to enter a password.

[0074] S410, when the Internet connection of the payment device is unstable, activate the local database, obtain the customer's facial image, and determine whether the facial image is stored in the local database. If so, deduct the money directly from the ID account corresponding to the facial image; if not, collect the customer's facial image through a preset four-dimensional model and enter it into the local database, and then obtain the customer's ID account entered by the merchant's cashier, and complete the deduction after the customer enters the password corresponding to the ID account for the first time;

[0075] S420, when the Internet connection of the payment device is stable, activate the Internet database and the local database, obtain the customer's facial picture, and determine whether the facial picture is stored in the local database. If so, deduct the money directly from the ID account corresponding to the facial picture; if not, collect the customer's facial picture through a preset four-dimensional model and enter it into the local database, and then retrieve the ID account corresponding to the facial picture from the Internet database. The deduction is completed after the customer enters the password corresponding to the ID account for the first time.

[0076] Reference Figure 2 , is a flow chart of a method for connecting a face-scanning payment device with a merchant's cash register in the second embodiment of the present application;

[0077] S411, when the Internet connection of the payment device is unstable, activating the local database;

[0078] S412, obtaining a facial image of the customer and determining whether the facial image is stored in a local database;

[0079] S413, if yes, deduct the money directly from the ID account corresponding to the facial image;

[0080] S414, if not, collect the customer's facial image through the preset four-dimensional model and enter it into the local database, then obtain the customer's ID account entered by the merchant's cashier, and complete the deduction after the customer enters the password corresponding to the ID account for the first time;

[0081] Specifically, when the Internet connection of the payment device is unstable, the local database is activated. Through the local database, the user identity can be verified without relying on safe databases such as WeChat, Alipay and banks. From the above, it can be seen that because the stores where the payment devices are located are relatively regional, the number of facial images required to be recorded in the local database is not large, and generally 32G or 64G storage media can be used to handle it.

[0082] S421, when the Internet connection of the payment device is stable, activating the Internet database and the local database;

[0083] S422, obtaining a facial image of the customer and determining whether the facial image is stored in a local database;

[0084] S423, if yes, deduct the money directly from the ID account corresponding to the facial image;

[0085] S424, if not, collect the customer's facial picture through the preset four-dimensional model and enter it into the local database, and then retrieve the ID account corresponding to the facial picture from the Internet database. The customer completes the deduction after entering the password corresponding to the ID account for the first time.

[0086] The Internet database is used only when the Internet is stable and the local database does not record the customer's facial image. The ID account corresponding to the facial image is then retrieved from the Internet database. The deduction is completed after the customer enters the password corresponding to the ID account for the first time.

[0087] In one embodiment, when the Internet connection of the payment device is unstable, the steps of directly deducting money from the ID account corresponding to the facial image and / or completing the deduction after the customer enters the password corresponding to the ID account for the first time include:

[0088] Recording the ID account, the deduction value and confidential data, where the confidential data is the user's permission to acknowledge the deduction, which is created when the facial image is determined in the local data and / or when the customer enters the password corresponding to the ID account;

[0089] Monitor whether the current Internet speed reaches the minimum standard speed;

[0090] If so, the deduction will be made through the ID account, deduction amount and confidential data.

[0091] This step is used by the payment device to deduct the amount to be paid in the ID account, and transfer the amount to be paid data to the payment device after deducting the amount, so as to finally transfer the deducted amount to the merchant's cash register.

[0092] Furthermore, if the current Internet speed monitored above does not reach the minimum standard speed:

[0093] Then temporarily disconnect the first docking data;

[0094] Amplify the communication network signal and deduct money through ID account, deduction value and confidential data;

[0095] The increase is suspended, and the wireless connection with the merchant's cash register through the first connection data is restored, and the deduction data after the deduction is entered into the merchant's cash register.

[0096] Specifically, the strength of the current communication signal is related to the wave frequency, which is the same as the principle of upgrading 4G to 5G. A wave frequency generator is added to the existing communication signal, and the wave frequency generator adopts existing technology;

[0097] Because the delivery device and the merchant's cash register are connected wirelessly, it will occupy a certain communication frequency, so the first connection data is temporarily disconnected, and the communication network signal is amplified through the frequency generator to improve the communication strength of the payment device.

[0098] In one embodiment, when the payment device has a stable Internet connection, the method for calibrating the four-dimensional model includes:

[0099] Retrieving facial images of customers from a local database and obtaining facial images of customers from an Internet database;

[0100] Compare the two facial images in the local database and the Internet database to obtain the face shape deviation tolerance interval;

[0101] The facial image verification accuracy of the customer is calibrated for the four-dimensional model through the face shape deviation tolerance interval.

[0102] Specifically, the facial picture of the customer in the local database (hereinafter referred to as the first picture) is compared with the facial picture of the customer in the Internet database (hereinafter referred to as the second picture). It should be noted that the two pictures to be compared are of the same customer; the facial shape deviation tolerance interval is the tolerance interval based on the customer facial shape encoding in the first picture and the second picture. For example, the second picture is a facial picture stored in the Internet database one month ago, and the first picture is a facial picture stored in the local database obtained by the payment device on the same day. By comparing the first picture with the second picture, the changes in the customer's facial shape in the two pictures can be determined.

[0103] After determining the face shape deviation tolerance range, the face shape deviation tolerance range is used to calibrate the customer's facial image verification accuracy for the four-dimensional model.

[0104] In one embodiment, a method for collecting a customer's facial image using a four-dimensional model includes:

[0105] Collect a preset number of frames of customer facial images;

[0106] Construct the facial image corresponding to each frame in three-dimensional coordinates;

[0107] Compare the X, Y, and Z axis differences of the facial image of each frame's three-dimensional coordinates to obtain an accurate facial image.

[0108] In one embodiment, the four-dimensional model is a pre-trained neural network model, and its training method includes:

[0109] Build and associate three-dimensional coordinates with the time axis to form the surface of the four-dimensional model;

[0110] Import the facial video as training data into the four-dimensional model, decompose the duration of the facial video and import it into the timeline, and obtain the facial image of each frame of the facial video after decomposition;

[0111] Load each frame of the facial image into three-dimensional coordinates one by one, perform image recognition on the facial image, determine a number of marking points in the facial image through image recognition, and then record the positions of the marking points;

[0112] According to the time axis, the facial images of the previous frame and the next frame are mapped, and the facial images of the previous frame and the next frame are connected at the same marked point of the three-dimensional coordinates through mapping. After the time axis is progressively advanced, a 4D image of the portrait in the facial video is established;

[0113] Establish a local database, construct the inner layer of the four-dimensional model, and store the 4D images in the local database.

[0114] Reference Figure 3 , is a structural block diagram of a payment device proposed in this application, the payment device includes:

[0115] Merchant unit 1, used for the payment device to obtain merchant information and load first docking data;

[0116] A docking unit 2 is used for wirelessly docking with the merchant's cashier according to the first docking data after receiving the docking permission from the merchant's cashier;

[0117] A local storage unit 3 is used to construct a local database, and the local database is used to store pictures and corresponding ID accounts;

[0118] The first payment channel unit 41 is used to activate the local database when the Internet connection of the payment device is unstable, obtain the customer's facial image, and determine whether the facial image is stored in the local database. If so, directly deduct money from the ID account corresponding to the facial image; if not, collect the customer's facial image through a preset four-dimensional model and enter it into the local database, and then obtain the customer's ID account entered by the merchant's cashier, and complete the deduction after the customer enters the password corresponding to the ID account for the first time;

[0119] The second payment channel unit 42 is used to activate the Internet database and the local database when the Internet connection of the payment device is stable, obtain the customer's facial picture, and determine whether the facial picture is stored in the local database. If so, deduct money directly from the ID account corresponding to the facial picture; if not, collect the customer's facial picture through a preset four-dimensional model and enter it into the local database, and then retrieve the ID account corresponding to the facial picture from the Internet database. The deduction is completed after the customer enters the password corresponding to the ID account for the first time.

[0120] In one embodiment, the first payment channel unit 41 includes:

[0121] A recording module for recording the ID account, the deduction value and confidential data, wherein the confidential data is the user's permission to acknowledge the deduction, which is created when the facial image is determined in the local data and / or when the customer enters the password corresponding to the ID account;

[0122] The network speed monitoring module is used to monitor whether the current Internet speed reaches the minimum standard speed;

[0123] The first deduction unit is used to deduct money through the ID account, deduction value and confidential data.

[0124] In one embodiment, the first payment channel unit 41 includes:

[0125] a disconnection module, for temporarily disconnecting the first docking data if no;

[0126] The second deduction module is used to amplify the communication network signal and deduct money through the ID account, deduction value and confidential data;

[0127] The recovery module is used to suspend the increase, resume the wireless connection with the merchant's cashier through the first connection data, and enter the deduction data after the deduction into the merchant's cashier.

[0128] Reference Figure 4 , which is a structural block diagram of a face-scanning payment system proposed in this application;

[0129] The facial recognition payment system includes a payment device, a merchant cashier counter, an Internet database and a management center, wherein the payment device includes a local database;

[0130] The payment device is used to obtain the customer's facial image, determine the ID account corresponding to the facial image from a preset local database, and / or determine the ID account corresponding to the facial image from an Internet database, and the payment device is connected to the merchant's cash register to transfer the debit in the ID account to the merchant's cash register;

[0131] Merchant cashiers are used to connect with payment devices to collect payments;

[0132] The Internet database is used to store the customer's facial images and is connected to the Internet database when the payment device network is stable;

[0133] The management center is used to record the registration, login and management of merchant cash registers through payment devices.

[0134] To summarize, the payment device obtains merchant information and loads the first docking data, which is used for merchant registration and login; after receiving the docking permission from the merchant cashier, the payment device wirelessly docks with the merchant cashier according to the first docking data to realize the wireless docking between the payment device and the merchant cashier; a local database is constructed, and the local database is used to store pictures and corresponding ID accounts; when the Internet connection of the payment device is unstable, the local database is activated, the customer's facial picture is obtained, and it is determined whether the facial picture is stored in the local database. If so, the payment is directly deducted from the ID account corresponding to the facial picture; if not, the customer's facial picture is collected through a preset four-dimensional model and entered into the local database, and then the customer's ID account entered by the merchant cashier is obtained, and the customer The deduction is completed after the password corresponding to the ID account is entered for the first time; when the Internet connection of the payment device is stable, the Internet database and the local database are activated, the customer's facial picture is obtained, and it is determined whether the facial picture is stored in the local database. If so, the deduction is made directly from the ID account corresponding to the facial picture; if not, the customer's facial picture is collected through a preset four-dimensional model and entered into the local database, and then the ID account corresponding to the facial picture is retrieved from the Internet database. The deduction is completed after the customer enters the password corresponding to the ID account for the first time, thereby solving the problem of unstable docking between payment devices and merchant cash registers and platforms in the prior art, as well as the process of entering a password for face recognition, resulting in slow payment speed and poor user experience.

[0135] Although the embodiments of the present application have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present application, and that the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A method for connecting a face-scanning payment device with a merchant's cash register, characterized in that: include: The payment device obtains the merchant information and loads the first docking data; After receiving the connection permission from the merchant's cashier counter, wirelessly connect with the merchant's cashier counter according to the first connection data; Building a local database, the local database is used to store pictures and corresponding ID accounts; When the Internet connection of the payment device is unstable, the local database is activated to obtain the customer's facial image, and it is determined whether the facial image is stored in the local database. If so, the payment is directly deducted from the ID account corresponding to the facial image; if not, the customer's facial image is collected through a preset four-dimensional model and entered into the local database, and then the customer's ID account entered at the merchant's cashier is obtained, and the deduction is completed after the customer enters the password corresponding to the ID account for the first time; When the Internet connection of the payment device is stable, the Internet database and the local database are activated to obtain the customer's facial image, and it is determined whether the facial image is stored in the local database. If so, the payment is directly deducted from the ID account corresponding to the facial image; if not, the customer's facial image is collected through a preset four-dimensional model and entered into the local database, and then the ID account corresponding to the facial image is retrieved from the Internet database, and the deduction is completed after the customer enters the password corresponding to the ID account for the first time; When the Internet connection of the payment device is unstable, the steps of directly deducting the amount from the ID account corresponding to the facial image and / or completing the deduction after the customer enters the password corresponding to the ID account for the first time include: Recording the ID account, the deduction value and confidential data, wherein the confidential data is the user's permission to acknowledge the deduction, which is created when the facial image is determined in the local data and / or when the customer enters the password corresponding to the ID account; Monitor whether the current Internet speed reaches the minimum standard speed; If so, deductions are made using the ID account, deduction value, and confidential data; After the step of monitoring whether the current Internet speed reaches the minimum standard speed, the method further includes: If not, temporarily disconnect the first docking data; Amplify the communication network signal and deduct money through ID account, deduction value and confidential data; The increase is suspended, the wireless connection with the merchant cashier counter is resumed through the first connection data, and the deduction data after the deduction is entered into the merchant cashier counter; When the payment device has a stable Internet connection, the method to calibrate the 4D model includes: Retrieving facial images of customers from a local database and obtaining facial images of customers from an Internet database; Comparing the two facial images in the local database and the Internet database to obtain a face shape deviation tolerance interval; Calibrate the customer's facial image verification accuracy for the four-dimensional model using the face shape deviation tolerance interval; The method for collecting the facial image of the customer using the four-dimensional model comprises: Collect a preset number of frames of customer facial images; Construct the facial image corresponding to each frame in three-dimensional coordinates; Compare the X, Y, and Z axis differences of the facial image of each frame's three-dimensional coordinates to obtain an accurate facial image; The four-dimensional model is a pre-trained neural network model, and its training method includes: Build and associate three-dimensional coordinates with the time axis to form the surface of the four-dimensional model; Importing a facial video as training data into a four-dimensional model, disassembling the duration of the facial video and importing it into a timeline, and obtaining a facial image of each frame of the facial video after disassembly; Loading each frame of the facial image into three-dimensional coordinates one by one, and performing image recognition on the facial image to determine a number of marking points in the facial image through image recognition, and then recording the positions of the marking points; Performing time progression according to the time axis, mapping the facial images of the previous frame and the next frame, connecting the facial images of the previous frame and the next frame at the same marked point of the three-dimensional coordinates through the mapping, and after the time progression of the time axis is completed, establishing a 4D image of the portrait in the facial video; A local database is established, an inner layer of a four-dimensional model is constructed, and the 4D image is stored in the local database.

2. A payment device, characterized in that: The method for connecting a face-scanning payment device with a merchant cash register according to claim 1 is used, wherein the payment device comprises: The merchant unit is used for the payment device to obtain merchant information and load the first docking data; A docking unit, configured to wirelessly dock with the merchant's cashier according to the first docking data after receiving the docking permission from the merchant's cashier; A local storage unit, used to build a local database, wherein the local database is used to store pictures and corresponding ID accounts; The first payment channel unit is used to activate the local database when the Internet connection of the payment device is unstable, obtain the customer's facial image, and determine whether the facial image is stored in the local database. If so, directly deduct money from the ID account corresponding to the facial image; if not, collect the customer's facial image through a preset four-dimensional model and enter it into the local database, and then obtain the customer's ID account entered by the merchant's cashier, and complete the deduction after the customer enters the password corresponding to the ID account for the first time; The second payment channel unit is used to activate the Internet database and the local database when the Internet connection of the payment device is stable, obtain the customer's facial picture, and determine whether the facial picture is stored in the local database. If so, deduct money directly from the ID account corresponding to the facial picture; if not, collect the customer's facial picture through a preset four-dimensional model and enter it into the local database, and then retrieve the ID account corresponding to the facial picture from the Internet database. The deduction is completed after the customer enters the password corresponding to the ID account for the first time.

3. The payment device according to claim 2, characterized in that: The first payment channel unit includes: a recording module for recording the ID account, the deduction value and confidential data, wherein the confidential data is the user's permission to acknowledge the deduction, which is created when the facial image is determined in the local data and / or when the customer enters the password corresponding to the ID account; The network speed monitoring module is used to monitor whether the current Internet speed reaches the minimum standard speed; The first deduction unit is used to deduct money through the ID account, deduction value and confidential data.

4. The payment device according to claim 3, characterized in that: The first payment channel unit includes: a disconnection module, for temporarily disconnecting the first docking data if no; The second deduction module is used to amplify the communication network signal and deduct money through the ID account, deduction value and confidential data; The recovery module is used to suspend the increase, restore the wireless connection with the merchant's cashier through the first connection data, and enter the deduction data after the deduction into the merchant's cashier.

5. A face-scanning payment system, characterized in that: The method for connecting a face-scanning payment device with a merchant cashier counter according to claim 1 is used, wherein the face-scanning payment system comprises a payment device, a merchant cashier counter, an Internet database, and a management center, wherein the payment device comprises a local database; The payment device is used to obtain the customer's facial image, determine the ID account corresponding to the facial image from a preset local database, and / or determine the ID account corresponding to the facial image from an Internet database, and the payment device is connected to the merchant's cash register to transfer the deductions in the ID account to the merchant's cash register; the merchant's cash register is used to connect with the payment device to collect payments; the Internet database is used to store the customer's facial image and is connected to the Internet database when the payment device network is stable; the management center is used to record the registration, login and management of the merchant's cash register through the payment device.

Images