Scenic area safety write-off method and safety write-off system

Abstract

The present application discloses a scenic spot safety verification method and a safety verification system, the method comprising: obtaining a network IP address and verification time of a mobile terminal; determining the identity authenticity of a user; calculating a first current position of the mobile terminal according to the network IP address, and determining a target positioning navigation hardware device; obtaining a signal strength between the target positioning navigation hardware device and the mobile terminal, and calculating a first interval distance; obtaining a first scenic spot verification area and a scenic spot business hour interval, and determining time authenticity; receiving face recognition data, determining the on-site authenticity of the user, and calculating the similarity between the face recognition data and the image recognition data; determining a reservation verification rate of the scenic spot; determining a verification safety characterization value of the scenic spot according to identity authenticity, the first interval distance, time authenticity, on-site authenticity, similarity and reservation verification rate; and starting a preset verification process when the safety characterization value is greater than or equal to a preset safety standard value.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of scenic spot write-off management, and in particular to a scenic spot safety write-off method and a safety write-off system. Background Art

[0002] In recent years, with the vigorous development of domestic tourism, the number of tourists in major scenic spots has continued to increase. In order to protect the ecological environment, ensure the safety of tourists and enhance the tour experience, many popular scenic spots have begun to implement a ticket reservation system. This system aims to reasonably allocate tourism resources, avoid overcrowding in scenic spots, and provide tourists with better planning and experience.

[0003] At present, the ticket reservation and verification process of most scenic spots is as follows: tourists make ticket reservations through the official website of the scenic spot, official APP or third-party platform, and need to fill in personal information, including name, identity information, etc. when making reservations. The reservation system will conduct a preliminary verification of the filled-in identity information to ensure the validity of the information, and then automatically allocate tickets according to the order of reservations and the capacity of the scenic spot on the day. Tourists can choose electronic tickets or go to the ticket office of the scenic spot to pick up paper tickets. After arriving at the scenic spot, tourists need to verify at the entrance. Verification methods usually include: scanning the QR code of the electronic ticket; checking identity information; manually checking the reservation information. After verification is completed, tourists can enter the park for sightseeing.

[0004] However, in actual operations, some lawless elements will collect a large amount of real identity information in advance, make reservations and cancellations through the above-mentioned reservation and cancellation processes, and resell the tickets, making the real ticket holder inconsistent with the reservation information. In addition, some scenic spots use cancellation machines in violation of regulations to cancel tickets before the tourists actually arrive, which seriously disrupts the normal operation of the scenic spot reservation system, reduces the effectiveness of the scenic spot reservation system, and brings huge challenges to scenic spot management. Summary of the invention

[0005] The embodiments of the present application provide a scenic spot safety verification method and a safety verification system, which are used to solve the problem of interference with the normal operation of the existing scenic spot reservation system, effectively improve the effectiveness of the scenic spot reservation system, and facilitate scenic spot management.

[0006] To achieve the above objectives, the embodiments of the present application adopt the following technical solutions:

[0007] In a first aspect, a scenic spot safety verification method is provided, which is applied to a scenic spot safety verification system, wherein the scenic spot safety verification system includes a control terminal, a mobile terminal, and at least one positioning and navigation hardware device, and the method includes:

[0008] The control terminal acquires the network IP address and verification time of the mobile terminal in response to receiving the user's identity information and image recognition data sent by the mobile terminal, wherein the identity information and the image recognition data are obtained by performing image recognition on the user's unique identity identification credential image;

[0009] The control terminal determines the authenticity of the user's identity based on the identity information and the image recognition data;

[0010] The control terminal sends a positioning request to the mobile terminal, and if no return data from the mobile terminal is received within a preset time period, calculates a first current position of the mobile terminal according to the network IP address;

[0011] The control terminal determines a target positioning and navigation hardware device in at least one of the positioning and navigation hardware devices through the first current position positioning;

[0012] The control terminal obtains the signal strength between the target positioning and navigation hardware device and the mobile terminal, and calculates a first interval distance between the mobile terminal and the target positioning and navigation hardware device according to the signal strength;

[0013] The control terminal obtains the first scenic spot verification area where the target positioning navigation hardware device is located and the scenic spot business time interval corresponding to the first scenic spot verification area, and determines the time authenticity according to the verification time and the scenic spot business time interval;

[0014] The control terminal, in response to receiving the face recognition data sent by the mobile terminal, determines the user's on-site authenticity according to the face recognition data, and calculates the similarity between the face recognition data and the image recognition data;

[0015] The control terminal obtains the current write-off data of the scenic spot;

[0016] The control terminal determines the reservation write-off rate of the scenic spot according to the historical write-off data;

[0017] The control terminal determines the cancellation safety representation value of the scenic spot according to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the reservation cancellation rate;

[0018] The control terminal starts a preset write-off process when the security characterization value is greater than or equal to a preset security standard value;

[0019] When the security characterization value is less than a preset security standard value, the control terminal interrupts the verification process.

[0020] In a possible implementation manner of the first aspect, after the control terminal sends the positioning request to the mobile terminal, the method further includes:

[0021] If the return data from the mobile terminal is received within a preset time period, the control terminal calculates the second current position of the mobile terminal according to the return data;

[0022] The control terminal determines a second scenic spot cancellation area according to the second current position, and obtains the scenic spot location of the second scenic spot cancellation area by querying in a preset scenic spot cancellation location database;

[0023] The control terminal calculates a second interval distance between the mobile terminal and the second scenic spot verification area according to the second current position positioning and the scenic spot positioning;

[0024] The control terminal obtains the first scenic spot verification area where the target positioning navigation hardware device is located and the scenic spot business time interval corresponding to the first scenic spot verification area, and determines the time authenticity according to the verification time and the scenic spot business time interval;

[0025] The control terminal determines the user's on-site authenticity in response to receiving the face recognition data sent by the mobile terminal, and calculates the similarity between the face recognition data and the image recognition data;

[0026] The control terminal obtains the current write-off data of the scenic spot;

[0027] The control terminal determines the reservation write-off rate of the scenic spot according to the historical write-off data;

[0028] The control terminal determines a verification safety characterization value of the scenic spot according to the identity authenticity, the second interval distance, the time authenticity, the scene authenticity, the similarity and the reservation verification rate.

[0029] In another possible implementation manner of the first aspect, the unique identity identification credential image includes characters and a portrait, and before the control terminal receives the user's identity information and image recognition data sent by the mobile terminal, the process includes:

[0030] The control terminal converts the user's unique identity identification credential image into a grayscale image, and performs threshold segmentation on the grayscale image to convert the grayscale image into a binary image containing only black and white pixels;

[0031] The control terminal rotates and corrects the binary image to ensure that the characters in the binary image are arranged horizontally;

[0032] The control terminal locates and segments the character area where the identity information is located and the portrait area where the image recognition data is located in the binary image;

[0033] The control terminal uses a preset OCR algorithm to recognize characters in the character area to obtain identity information;

[0034] The control terminal uses a pre-trained deep learning model to extract features from the portrait area to obtain a facial feature vector, wherein the facial feature vector is used to represent the image recognition data;

[0035] The control terminal determines the authenticity of the user's identity based on the identity information and the image recognition data, including:

[0036] Determining target identity information matching the facial feature vector in a preset database;

[0037] Calculating the similarity between the target identity information and the identity information of the user;

[0038] When the similarity is greater than a preset similarity threshold, sending a verification request so that the user enters a verification password through the verification request;

[0039] In response to receiving the number of times the user inputs the verification password sent by the mobile terminal, calculating the security level using a preset security level calculation formula according to the number of inputs;

[0040] The identity authenticity is calculated using a preset identity authenticity calculation formula based on the similarity and the security level.

[0041] In another possible implementation manner of the first aspect, calculating the first current position of the mobile terminal according to the network IP address includes:

[0042] Segmenting the network IP address to obtain multiple IP address segments;

[0043] In the preset geographic location database, query and summarize the geographic location corresponding to each IP address segment;

[0044] In the case where the summarized geographical locations are different, querying a preset ISP database for ISP information corresponding to the network IP address, the ISP information including the ISP location;

[0045] Determine at least one network terminal test point according to the summarized geographical locations, wherein each network terminal test point corresponds to one geographical location;

[0046] Sending a test data packet to a mobile terminal so that the mobile terminal sends the test data packet to each of the network terminal test points;

[0047] In response to receiving the time of returning the data packet from each network terminal test point sent by the mobile terminal, calculating the round-trip delay between the mobile terminal and each network terminal test point, and taking the shortest round-trip delay as the target delay;

[0048] Using a preset distance calculation formula, the distance from the mobile terminal to the network terminal test point corresponding to the target delay is calculated according to the target delay;

[0049] A first current position location is determined based on the distance and the ISP position.

[0050] In another possible implementation of the first aspect, the scenic spot business hour interval includes the scenic spot business start time and the scenic spot business end time, and determining the time authenticity according to the verification time and the scenic spot business hour interval includes:

[0051] If the verification time is within the business hours of the scenic spot, the time authenticity is determined to be 1;

[0052] If the verification time is not within the business hours of the scenic spot, and the verification time is before the business start time of the scenic spot, a first time period between the verification time and the business start time of the scenic spot is calculated, and a first coefficient corresponding to the first time period is determined in a preset authenticity database, and a value of 1 minus the first coefficient is used as the time authenticity;

[0053] If the verification time is not within the business hours of the scenic spot, and the verification time is after the closing time of the scenic spot, a second time period between the verification time and the closing time of the scenic spot is calculated, and a second coefficient corresponding to the second time period is determined in the authenticity database, and the value of 1 minus the second coefficient is used as the time authenticity.

[0054] In another possible implementation manner of the first aspect, determining the on-site authenticity of the user according to the face recognition data, and calculating the similarity between the face recognition data and the image recognition data, includes:

[0055] Performing continuity detection on the face recognition data to determine whether the face movement trajectory is normal;

[0056] When the face movement trajectory is normal, a preset liveness algorithm is used to calculate the scene authenticity according to the face recognition data;

[0057] Determining a key facial feature vector in the face recognition data, and determining a key facial feature vector in the face feature vector, wherein the key facial feature vector corresponds to the key facial feature vector in a one-to-one manner;

[0058] Calculating the cosine similarity between each of the key facial feature vectors and the corresponding key facial feature vector;

[0059] An average value of all the cosine similarities is used as the similarity between the face recognition data and the image recognition data.

[0060] In another possible implementation manner of the first aspect, the control terminal determines the cancellation safety characterization value of the scenic spot according to the identity authenticity, the first interval distance, the time authenticity, the on-site authenticity, the similarity, and the reservation cancellation rate, including:

[0061] According to the identity information of the user, obtaining the user information of the user, the user information including the number of verifications and the verification success rate;

[0062] Determining weights of the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the reservation verification rate respectively according to the number of verifications and the verification success rate;

[0063] According to the weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the on-site authenticity, the similarity and the reservation cancellation rate, a weighted average of the identity authenticity, the first interval distance, the time authenticity, the on-site authenticity, the similarity and the reservation cancellation rate is calculated, and the weighted average is used as a value representing the cancellation safety of the scenic spot.

[0064] In another possible implementation manner of the first aspect, determining the weights of the identity authenticity, the first interval distance, the time authenticity, the on-site authenticity, the similarity, and the appointment verification rate respectively according to the number of verifications and the verification success rate includes:

[0065] When the number of verifications is greater than a preset number threshold, and the verification success rate is less than a preset success rate threshold, a first weight allocation strategy is adopted to respectively allocate first weights to the identity authenticity, the first interval distance, the time authenticity, the on-site authenticity, the similarity, and the appointment verification rate;

[0066] When the number of verifications is not greater than the number threshold and the verification success rate is less than the success rate threshold, a second weight allocation strategy is adopted to respectively allocate second weights to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the appointment verification rate;

[0067] When the number of verifications is not greater than the number threshold and the verification success rate is greater than the success rate threshold, a third weight allocation strategy is adopted to respectively allocate third weights to the identity authenticity, the first interval distance, the time authenticity, the on-site authenticity, the similarity and the appointment verification rate;

[0068] When the number of verifications is greater than the number threshold and the verification success rate is greater than the success rate threshold, the fourth weight allocation strategy is adopted to respectively allocate fourth weights to the identity authenticity, the first interval distance, the time authenticity, the on-site authenticity, the similarity and the appointment verification rate.

[0069] In another possible implementation manner of the first aspect, the first weight allocation strategy includes:

[0070] Obtaining initial weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity, and the reservation cancellation rate, respectively;

[0071] According to the successful verification rate and the initial weight, a first weight allocation formula is used to calculate the first weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the appointment verification rate, respectively, wherein the first weights corresponding to the identity authenticity and the scene authenticity are both greater than the corresponding initial weights, the first weight of the similarity is consistent with the corresponding initial weight, and the first weights corresponding to the time authenticity, the first interval distance and the appointment verification rate are both less than the corresponding initial weights;

[0072] The second weight allocation strategy includes:

[0073] Obtaining initial weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity, and the reservation cancellation rate, respectively;

[0074] According to the successful verification rate and the initial weight, a second weight allocation formula is used to calculate the second weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the appointment verification rate, respectively, wherein the second weights corresponding to the identity authenticity, the scene authenticity and the similarity are all greater than the corresponding initial weights, the second weights corresponding to the time authenticity and the first interval distance are both consistent with the corresponding initial weights, and the second weight corresponding to the appointment verification rate is less than the corresponding initial weight;

[0075] The third weight allocation strategy includes:

[0076] Obtaining initial weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity, and the reservation cancellation rate, respectively;

[0077] According to the successful verification rate and the initial weight, a third weight allocation formula is used to calculate the third weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the appointment verification rate, respectively, wherein the third weights corresponding to the identity authenticity and the scene authenticity are both smaller than the corresponding initial weights, the third weight of the similarity is consistent with the corresponding initial weight, and the third weights corresponding to the time authenticity, the first interval distance and the appointment verification rate are both larger than the corresponding initial weights;

[0078] The fourth weight distribution strategy includes:

[0079] Obtaining initial weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity, and the reservation cancellation rate, respectively;

[0080] According to the successful verification rate and the initial weight, the fourth weight allocation formula is adopted to calculate the fourth weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the on-site authenticity, the similarity and the appointment verification rate, respectively, wherein the fourth weights corresponding to the identity authenticity and the on-site authenticity are both smaller than the corresponding initial weights, the fourth weight of the similarity is consistent with the corresponding initial weight, the fourth weights corresponding to the time authenticity, the first interval distance and the time authenticity are both greater than the corresponding initial weights, and the fourth weight of the appointment verification rate is greater than the fourth weights corresponding to the first interval distance and the time authenticity.

[0081] In a second aspect, the present application provides a scenic spot safety verification system, the scenic spot safety verification system comprising a control terminal, a mobile terminal and at least one positioning and navigation hardware device, the control terminal comprising:

[0082] a memory configured to store instructions; and

[0083] The processor is configured to call the instructions from the memory and implement the above-mentioned scenic spot safety verification method when executing the instructions.

[0084] Through the above technical scheme, firstly, by verifying the user's identity information and image recognition data, it is possible to effectively prevent criminals from using the collected real identity information to make reservations and write-offs for scenic spots. In addition, the authenticity of the user's identity can be determined based on the identity information and image recognition data, thereby filtering out reservations using false or stolen identity information. Secondly, through positioning technology and signal strength analysis, it is possible to accurately determine whether the user has actually arrived at the scenic spot, effectively preventing the scenic spot from illegally using the write-off machine to write off without the tourist actually arriving. The control terminal can obtain the network IP address of the mobile terminal, calculate its current location, and further verify the accuracy of the user's location through the positioning navigation hardware device to ensure that the user is indeed in the scenic spot during the write-off process. Furthermore, by comparing the write-off time with the business hours of the scenic spot, the authenticity of the time can be determined, avoiding the write-off behavior during abnormal business hours. In addition, face recognition technology is introduced to further verify the user's on-site authenticity and identity consistency by comparing the face recognition data collected on site with the image recognition data at the time of reservation, effectively preventing the situation where the real ticket holder does not match the reservation information after ticket resale. In addition, the overall operation status of the scenic spot is also taken into account. By analyzing the current write-off data and historical write-off data, the reservation write-off rate is calculated, and the rationality of the write-off behavior is evaluated from a macro perspective. Finally, all the above factors are taken into consideration to calculate a write-off security representation value. Only when the write-off security representation value reaches or exceeds the preset security standard will the write-off process be started, otherwise the write-off will be interrupted. The above multi-level and multi-angle verification mechanism can greatly improve the security and accuracy of the write-off process, and effectively curb the fraudulent behavior of lawbreakers and the illegal operation of scenic spots. It not only effectively solves the problems of identity information abuse, false write-off, ticket reselling, etc., but also improves the credibility and effectiveness of the entire reservation system, which can effectively ensure that only real reservation users can complete the write-off when they actually arrive at the scenic spot, so as to better achieve the goals of scenic spot capacity management, improving tourist experience, and protecting the ecological environment. At the same time, the write-off mechanism can also effectively deter potential violations and reduce the speculative motives of lawbreakers. In summary, the scenic spot safety verification method can effectively reduce the difficulty and cost of scenic spot management, improve the execution efficiency of the reservation system, solve the problem of interference with the normal operation of the existing scenic spot reservation system, effectively improve the effectiveness of the scenic spot reservation system, and facilitate scenic spot management.

[0085] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] Figure 1 A schematic diagram of the structure of a control terminal provided in an embodiment of the present application;

[0087] Figure 2 A schematic diagram of a flow chart of a scenic spot safety verification method provided in an embodiment of the present application;

[0088] Figure 3 A schematic diagram of the structure of a scenic area safety verification system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0089] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the specific implementation methods described herein are only used to illustrate and explain the embodiments of the present application, and are not used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0090] It should be noted that if the embodiments of the present application involve directional indications (such as up, down, left, right, front, back...), such directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0091] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0092] Figure 1 A schematic structural diagram of the control terminal 10 is shown.

[0093] The control terminal 10 may include a processor 1, an external memory interface 6, an internal memory 5, a universal serial bus (USB) interface 7, an antenna a, an antenna b, a mobile communication module 2, a wireless communication module 3, a display screen 4, and the like.

[0094] It is to be understood that the structure illustrated in the embodiment of the present invention does not constitute a specific limitation on the control terminal 10. In other embodiments of the present application, the control terminal 10 may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

[0095] The processor 1 may include one or more processing units, for example, the processor 1 may include an application processor (AP), a modem processor, a graphics processor (GPU), an image signal processor (ISP), a controller, a memory, a video codec, a digital signal processor (DSP), a baseband processor, and / or a neural-network processing unit (NPU), etc. Different processing units may be independent devices or integrated into one or more processors.

[0096] The controller may be the nerve center and command center of the control terminal 10. The controller may generate an operation control signal according to the instruction operation code and the timing signal to complete the control of fetching and executing instructions.

[0097] A memory may also be provided in the processor 1 for storing instructions and data. In some embodiments, the memory in the processor 1 is a cache memory. The memory may store instructions or data that have just been used or are used cyclically by the processor 1. If the processor 1 needs to use the instruction or data again, it may be directly called from the memory. This avoids repeated access, reduces the waiting time of the processor 1, and thus improves the efficiency of the system.

[0098] In some embodiments, the processor 1 may include one or more interfaces. The interface may include an inter-integrated circuit (I2C) interface, an inter-integrated circuit sound (I2S) interface, a pulse code modulation (PCM) interface, a universal asynchronous receiver / transmitter (UART) interface, a mobile industry processor interface (MIPI), a general-purpose input / output (GPIO) interface, and / or a universal serial bus (USB) interface, etc.

[0099] The I2C interface is a bidirectional synchronous serial bus, including a serial data line (SDA) and a serial clock line (SCL). In some embodiments, the processor 1 may include multiple groups of I2C buses. The processor 1 may be coupled to the touch sensor 180K, the charger, the flash, the camera 8, etc. through different I2C bus interfaces. For example, the processor 1 may be coupled to the touch sensor 180K through the I2C interface, so that the processor 1 communicates with the touch sensor 180K through the I2C bus interface to realize the touch function of the control terminal 10.

[0100] The I2S interface can be used for audio communication. In some embodiments, the processor 1 can include multiple I2S buses. The processor 1 can be coupled to the audio module via the I2S bus to achieve communication between the processor 1 and the audio module.

[0101] The PCM interface can also be used for audio communication, sampling, quantizing and encoding analog signals. In some embodiments, the audio module and the wireless communication module 3 can be coupled via a PCM bus interface.

[0102] The UART interface is a universal serial data bus for asynchronous communication. The bus can be a bidirectional communication bus. It converts the data to be transmitted between serial communication and parallel communication. In some embodiments, the UART interface is generally used to connect the processor 1 and the wireless communication module 3. For example, the processor 1 communicates with the Bluetooth module in the wireless communication module 3 through the UART interface to implement the Bluetooth function.

[0103] The MIPI interface can be used to connect the processor 1 with peripheral devices such as the display screen 4 and the camera 8. The MIPI interface includes a camera serial interface (CSI), a display serial interface (DSI), etc. In some embodiments, the processor 1 and the camera 8 communicate via the CSI interface to realize the shooting function of the control terminal 10. The processor 1 and the display screen 4 communicate via the DSI interface to realize the display function of the control terminal 10.

[0104] The GPIO interface can be configured by software. The GPIO interface can be configured as a control signal or as a data signal. In some embodiments, the GPIO interface can be used to connect the processor 1 with the camera 8, the display 4, the wireless communication module 3, etc. The GPIO interface can also be configured as an I2C interface, an I2S interface, a UART interface, a MIPI interface, etc.

[0105] The USB interface 7 is an interface that complies with USB standard specifications, and specifically can be a Mini USB interface, a Micro USB interface, a USB Type C interface, etc. The USB interface 7 can be used to connect a charger to charge the control terminal 10, and can also be used to transmit data between the control terminal 10 and peripheral devices.

[0106] It is understandable that the interface connection relationship between the modules illustrated in the embodiment of the present invention is only a schematic illustration and does not constitute a structural limitation on the control terminal 10. In other embodiments of the present application, the control terminal 10 may also adopt different interface connection methods in the above embodiments, or a combination of multiple interface connection methods.

[0107] The wireless communication function of the control terminal 10 can be realized by antenna a, antenna b, mobile communication module 2, wireless communication module 3, modem processor and baseband processor.

[0108] Antenna a and antenna b are used to transmit and receive electromagnetic wave signals. Each antenna in the control terminal 10 can be used to cover a single or multiple communication frequency bands. Different antennas can also be reused to improve the utilization rate of the antennas. For example, antenna a can be reused as a diversity antenna for a wireless local area network. In some other embodiments, the antenna can be used in combination with a tuning switch.

[0109] The mobile communication module 2 can provide solutions for wireless communications including 2G / 3G / 4G / 5G applied to the control terminal 10. The mobile communication module 2 may include at least one filter, a switch, a power amplifier, a low noise amplifier (LNA), etc. The mobile communication module 2 can receive electromagnetic waves through the antenna a, and filter, amplify, and process the received electromagnetic waves, and transmit them to the modulation and demodulation processor for demodulation. The mobile communication module 2 can also amplify the signal modulated by the modulation and demodulation processor, and convert it into electromagnetic waves for radiation through the antenna a. In some embodiments, at least some of the functional modules of the mobile communication module 2 can be set in the processor 1. In some embodiments, at least some of the functional modules of the mobile communication module 2 can be set in the same device as at least some of the modules of the processor 1.

[0110] The modem processor may include a modulator and a demodulator. The modulator is used to modulate the low-frequency baseband signal to be sent into a medium-high frequency signal. The demodulator is used to demodulate the received electromagnetic wave signal into a low-frequency baseband signal. The demodulator then transmits the demodulated low-frequency baseband signal to the baseband processor for processing. After the low-frequency baseband signal is processed by the baseband processor, it is passed to the application processor. The application processor outputs a sound signal through an audio device, or displays an image or video through a display screen 4. In some embodiments, the modem processor may be an independent device. In other embodiments, the modem processor may be independent of the processor 1 and be set in the same device as the mobile communication module 2 or other functional modules.

[0111] The wireless communication module 3 can provide wireless communication solutions including wireless local area networks (WLAN) (such as wireless fidelity (Wi-Fi) networks), bluetooth (BT), global navigation satellite system (GNSS), frequency modulation (FM), near field communication (NFC), infrared (IR), etc., which are applied to the control terminal 10. The wireless communication module 3 can be one or more devices integrating at least one communication processing module. The wireless communication module 3 receives electromagnetic waves via antenna b, modulates and filters the electromagnetic wave signals, and sends the processed signals to the processor 1. The wireless communication module 3 can also receive the signal to be sent from the processor 1, modulate the frequency, amplify it, and convert it into electromagnetic waves for radiation through the antenna b.

[0112] In some embodiments, the antenna a of the control terminal 10 is coupled to the mobile communication module 2, and the antenna b is coupled to the wireless communication module 3, so that the control terminal 10 can communicate with the network and other devices through wireless communication technology. The wireless communication technology may include global system for mobile communications (GSM), general packet radio service (GPRS), code division multiple access (CDMA), wideband code division multiple access (WCDMA), time-division code division multiple access (TD-SCDMA), long term evolution (LTE), BT, GNSS, WLAN, NFC, FM, and / or IR technology. GNSS may include the global positioning system (GPS), the global navigation satellite system (GLONASS), the Beidou navigation satellite system (BDS), the quasi-zenith satellite system (QZSS) and / or the satellite based augmentation system (SBAS).

[0113] The control terminal 10 realizes the display function through a GPU, a display screen 4, and an application processor. The GPU is a microprocessor for image processing, which connects the display screen 4 and the application processor. The GPU is used to perform mathematical and geometric calculations for graphics rendering. The processor 1 may include one or more GPUs, which execute program instructions to generate or change display information.

[0114] The display screen 4 is used to display images, videos, etc. The display screen 4 includes a display panel. The display panel can be a liquid crystal display (LCD), an organic light-emitting diode (OLED), an active matrix organic light-emitting diode or an active matrix organic light-emitting diode, a flexible light-emitting diode (FLED), Miniled, MicroLed, Micro-oLed, a quantum dot light emitting diode (QLED), etc. In some embodiments, the control terminal 10 may include 1 or N display screens 4, where N is a positive integer greater than 1.

[0115] NPU is a neural network (NN) computing processor. By drawing on the structure of biological neural networks, such as the transmission mode between neurons in the human brain, it can quickly process input information and can also continuously self-learn. Through NPU, applications such as intelligent cognition of the control terminal 10 can be realized, such as image recognition, face recognition, voice recognition, text understanding, etc.

[0116] The external memory interface 6 can be used to connect an external memory card, such as a Micro SD card, to expand the storage capacity of the control terminal 10. The external memory card communicates with the processor 1 through the external memory interface 6 to implement a data storage function. For example, files such as music and videos are stored in the external memory card.

[0117] The internal memory 5 can be used to store computer executable program codes, and the executable program codes include instructions. The processor 1 executes various functional applications and data processing of the control terminal 10 by running the instructions stored in the internal memory 5. The internal memory 5 may include a program storage area and a data storage area. Among them, the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc. The data storage area may store data created during the use of the control terminal 10 (such as audio data, a phone book, etc.), etc. In addition, the internal memory 5 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, a universal flash storage (UFS), etc.

[0118] Based on the above control terminal, Figure 2 The following is a schematic diagram showing a flow chart of a scenic spot safety verification method according to an embodiment of the present application. Figure 2As shown, an embodiment of the present application provides a scenic spot safety verification method, which is applied to a scenic spot safety verification system. The scenic spot safety verification system includes a control terminal, a mobile terminal and at least one positioning and navigation hardware device. The method may include the following steps.

[0119] S101, the control terminal obtains the network IP address and verification time of the mobile terminal in response to receiving the user's identity information and image recognition data sent by the mobile terminal, wherein the identity information and image recognition data are obtained by performing image recognition on the user's unique identity identification credential image;

[0120] S102, the control terminal determines the authenticity of the user's identity based on the identity information and image recognition data;

[0121] S103, the control terminal sends a positioning request to the mobile terminal, and if no return data from the mobile terminal is received within a preset time period, a first current position of the mobile terminal is calculated according to the network IP address;

[0122] S104, controlling the terminal to determine a target positioning and navigation hardware device in at least one positioning and navigation hardware device through the first current position positioning;

[0123] S105, the control terminal obtains the signal strength between the target positioning and navigation hardware device and the mobile terminal, and calculates a first interval distance between the mobile terminal and the target positioning and navigation hardware device according to the signal strength;

[0124] S106, the control terminal obtains the first scenic spot verification area where the target positioning navigation hardware device is located and the scenic spot business hour interval corresponding to the first scenic spot verification area, and determines the time authenticity according to the verification time and the scenic spot business hour interval;

[0125] S107, in response to receiving the face recognition data sent by the mobile terminal, the control terminal determines the user's on-site authenticity according to the face recognition data, and calculates the similarity between the face recognition data and the image recognition data;

[0126] S108, the control terminal obtains the current write-off data of the scenic spot;

[0127] S109, the control terminal determines the reservation write-off rate of the scenic spot according to the historical write-off data;

[0128] S110, the control terminal determines the cancellation safety representation value of the scenic spot according to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the reservation cancellation rate;

[0129] S111. When the security characterization value is greater than or equal to the preset security standard value, the control terminal starts a preset write-off process;

[0130] S112. When the security characterization value is less than the preset security standard value, the control terminal interrupts the verification process.

[0131] The control terminal first receives the user identity information and image recognition data sent by the mobile terminal. The identity information and image recognition data are obtained by performing image recognition on the unique identity identification certificate provided by the user. The control terminal then obtains the network IP address and verification time of the mobile terminal. In specific implementation, the user needs to upload a photo of the ID card on the mobile terminal. The mobile terminal recognizes the ID card information through the OCR algorithm and extracts key information such as name and personal ID. At the same time, the mobile terminal records the current time as the verification time and sends its own IP address to the control terminal. After receiving the above information, the control terminal temporarily stores it in the memory or database for subsequent processing.

[0132] The control terminal determines the authenticity of the user's identity based on the received identity information and image recognition data. In specific implementation, the control terminal can compare the received identity information with the reservation database to confirm whether there is a matching reservation record. Secondly, the image recognition algorithm can be used to analyze the authenticity of the ID card photo to check whether there are PS traces or other anomalies. In addition, real-time identity verification can be performed by calling the preset interface. The control terminal can assign different weights to each verification method and finally calculate an identity authenticity score. For example, a score range of 0 to 100 can be set, where 90 points or more means that the identity is very likely to be authentic, 70-90 points means that it may be authentic but needs further verification, and below 70 points means that the identity is questionable, so that the behavior of using false identity information to make reservations can be effectively filtered out, and the security and reliability of the verification process can be improved.

[0133] In order to determine the user's location, the control terminal sends a positioning request to the mobile terminal. If no return data is received from the mobile terminal within a preset time period (such as 30 seconds), the control terminal calculates the first current location of the mobile terminal based on the previously acquired network IP address. In specific implementation, the control terminal can first try to obtain the precise location through the GPS or base station positioning of the mobile terminal. If this method fails, the IP address positioning method will be used instead. IP address positioning is usually achieved by querying IP address databases, which contain the correspondence between IP addresses and geographic locations. The control terminal can use multiple IP geographic location databases for cross-validation to improve positioning accuracy, thereby ensuring that the user's approximate location information can be obtained even if the user device cannot be positioned normally.

[0134] After determining the first current position positioning information, the control terminal uses the first current position positioning information to determine the target positioning navigation hardware device closest to the user's current position among the multiple positioning navigation hardware devices in the scenic area. In specific implementation, the control terminal can first obtain the location information of all positioning navigation hardware devices, and the location information is pre-stored in the local database. Then, the control terminal calculates the distance between the first current position positioning and each positioning navigation hardware device. The calculation method can use the Haversing formula to process the great circle distance between two points on the earth's surface. After obtaining all distances, the control terminal selects the positioning navigation hardware device with the shortest distance as the target device. If there are multiple devices with similar distances, the device with the strongest signal or the most recently installed device can be selected as the target to determine the hardware device that is most suitable for further interaction with the user device, and prepare for subsequent precise positioning and distance calculation, thereby improving the accuracy and efficiency of the verification process.

[0135] In order to calculate the distance between the target positioning and navigation hardware device and the mobile terminal, the control terminal can obtain the signal strength between the target positioning and navigation hardware device and the mobile terminal, and calculate the first interval distance between the two based on the signal strength. During the implementation process, the target positioning and navigation hardware device will continue to broadcast the signal. After the mobile terminal receives the signal, it measures the signal strength (the signal strength is in dBm) and sends this data to the control terminal. The control terminal can then use the signal attenuation model to estimate the distance. The signal attenuation model can be a logarithmic distance path loss model, as shown below:

[0136] ;

[0137] Among them, the transmission power and constant are known parameters, and the path loss index depends on the environment and has a value between 2 and 4. The control terminal can dynamically adjust the parameters to improve the accuracy according to the specific environment of the scenic spot (such as whether there are obstructions), thereby providing more accurate short-distance measurements than GPS, effectively verifying whether the user is actually in the scenic spot verification area, thereby preventing remote false verification.

[0138] In order to effectively prevent the cancellation behavior during non-normal business hours, the control terminal first obtains the cancellation area information of the first scenic spot where the target positioning navigation hardware device is located, and the scenic spot business hours interval corresponding to the area. Then, the control terminal compares the cancellation time with the scenic spot business hours interval to determine the time authenticity. In the specific implementation, the detailed information of each cancellation area can be pre-stored in the database, including its corresponding positioning navigation hardware device ID, geographical range, business hours, etc. The control terminal obtains the cancellation area information associated with the target positioning navigation hardware device by querying the database. The calculation of time authenticity can adopt a preset scoring system. If the cancellation time falls completely within the business hours interval, 100 points can be obtained. If it is slightly beyond the business hours (for example, within 15 minutes ahead or later), 80-90 points can be obtained. If it is seriously beyond the business hours, the score may be quickly reduced to 0, thereby effectively preventing the cancellation behavior during non-normal business hours and improving the time accuracy and compliance of the cancellation process.

[0139] In order to further confirm whether the user is the one who is verifying the ticket, in this embodiment, the control terminal receives the face recognition data sent by the mobile terminal, and uses the face recognition data to determine the user's on-site authenticity, and calculates the similarity between the face recognition data and the previous image recognition data. During the implementation process, the user needs to take real-time face photos on the mobile terminal. The mobile terminal uses the front camera to capture the user's facial image and generates a facial feature vector through a facial feature extraction algorithm (such as principal component analysis PCA or linear discriminant analysis LDA). The above data is sent to the control terminal. The control terminal first performs a liveness detection to ensure that the captured face is a real face rather than a photo. Then, the control terminal compares the facial feature vector with the pre-stored image recognition data. The comparison can use methods such as cosine similarity or Euclidean distance. The similarity score ranges from 0 to 1, where 1 indicates a complete match and 0 indicates a complete difference. A threshold value (such as 0.8) can be set, and if it exceeds this threshold, it is determined to be verified. The on-site authenticity can also take into account other factors, such as the lighting conditions of the shooting environment, whether the user cooperates, etc., thereby greatly improving the security of the verification process and effectively preventing identity fraud and ticket reselling.

[0140] The control terminal then obtains the current write-off data of the scenic spot. The current write-off data may include the number of tickets written off on the day, the write-off distribution in each time period, the write-off status of different types of tickets, etc. In specific implementation, the control terminal can be connected to the write-off database of the scenic spot in real time, and obtain the latest write-off statistics through SQL query or API call. In addition, a regular update mechanism can be set up, such as updating the data every 5 minutes, to ensure the real-time nature of the data. The acquired data may include the total write-off number, the write-off number of each entrance, the write-off number of different ticket types, the write-off trend per hour, etc. The control terminal can perform preliminary processing on the above data and calculate key indicators, such as the current write-off rate (the number of written-offs / the total number of reservations), the write-off speed during peak hours, etc., so that the control terminal can grasp the passenger flow of the scenic spot in real time, which helps to prevent excessive write-offs or abnormal write-off behaviors.

[0141] After that, the control terminal can determine the reservation write-off rate of the scenic spot based on the historical write-off data. The control terminal can access a data warehouse containing write-off data for a period of time in the past (such as the past 30 days or the past year). The historical data may include information such as the number of daily reservations, the actual number of write-offs, and the time distribution of write-offs. The control terminal can use time series analysis methods, such as exponential smoothing, to process historical data to obtain an average reservation write-off rate. At the same time, seasonal factors, holiday effects, etc. can also be considered, and the prediction results can be optimized using a preset prediction model. The calculation of the reservation write-off rate can distinguish different types of dates (such as weekdays, weekends, and holidays) to improve the accuracy of the prediction. In addition, the confidence interval of the reservation write-off rate can be calculated to take into account possible fluctuations, thereby establishing an expected normal write-off behavior benchmark, providing an important reference for identifying potential abnormal write-off activities, and thus improving the overall safety and effectiveness of the write-off process.

[0142] After obtaining the above six control factors (including identity authenticity, first interval distance, time authenticity, scene authenticity, face recognition similarity and appointment cancellation rate), the control terminal can use the weighted average algorithm to integrate the above six control factors to obtain the security characterization value. Among them, each factor is assigned a different weight to reflect its importance in the security assessment. For example, identity authenticity and scene authenticity may have a higher weight, while the weight of the appointment cancellation rate may be relatively low. In the specific implementation, each factor is standardized to a range of 0-1, and then the weight calculation is applied. In addition, the security characterization value can be further optimized by preset rules. For example, if a factor is below a specific threshold, it is directly judged as unsafe. The final security characterization value can be a score from 0 to 100, where 100 represents the highest security. In the specific implementation process, the weights and rules can be continuously optimized based on actual operation data and security event feedback to achieve a comprehensive evaluation of the cancellation process.

[0143] When the security characterization value is greater than or equal to the preset security standard value, the control terminal starts the preset verification process. In specific implementation, a security standard value can be preset, such as 80 points (assuming that the full score of the security characterization value is 100 points). When the calculated security characterization value reaches or exceeds this standard, the control terminal triggers the verification process. The start of the verification process means that the system believes that the identity and behavior of the current user meet the security standards and can verify and use the ticket or reservation. The specific verification process may include the following steps: First, the control terminal generates a unique verification code, which can be a QR code containing a timestamp and encryption information. Then, the verification code is sent to the user's mobile terminal. The user can then scan the verification code on a dedicated device at the entrance of the scenic spot. After the scan is completed, the control terminal will update the database and mark the ticket as "cancelled". The control terminal can record a detailed verification log, including verification time, location, operator and other information. If the scenic spot uses a physical ticket checking device, the control terminal will send a signal to the physical ticket checking device to allow entry. Finally, the control terminal sends a successful verification notification to the user, and updates the real-time passenger flow statistics of the scenic spot. If the scenic spot has capacity restrictions, the control terminal will also automatically update the current number of visitors. The whole process is fast and efficient, which not only ensures safety, but also improves user experience and scenic spot operation efficiency, effectively ensuring that only users who have undergone comprehensive security verification can complete the verification process, improving the safety and reliability of scenic spot management.

[0144] If the security characterization value is less than the preset security standard value, the control terminal interrupts the verification process. Specifically, the control terminal immediately stops all operations related to the verification and marks the verification attempt as "failed". The specific implementation of the interruption process includes the following steps: First, the control terminal immediately freezes the verification request of the user and temporarily blocks any further verification operations. At the same time, a warning message is sent to the user's mobile device to inform the reason for the verification failure. In addition, the user can be prompted to verify personal information or try again. The control terminal will also automatically mark the failed verification attempt and record the relevant information in the user's behavior log for subsequent analysis and risk assessment. If the same user fails to verify multiple times in a short period of time, the control terminal will automatically increase the risk level of the user and increase the difficulty of verification in future verification. At the same time, the control terminal sends a real-time alert to the scenic area manager to remind him to pay attention to potential abnormal situations. Managers can view detailed failure reasons and user information through a dedicated management interface, and can manually intervene and verify when necessary. This mechanism can not only effectively prevent fraud, but also help identify possible misjudgments of the control terminal, thereby continuously optimizing the verification algorithm and security strategy. By promptly interrupting the risky write-off process, we not only protect the interests of the scenic spot, but also ensure that the rights and interests of legitimate users are not affected.

[0145] First, by verifying the user's identity information and image recognition data, this embodiment can effectively prevent criminals from using the collected real identity information to make reservations and write-offs for scenic spots. In addition, the authenticity of the user's identity can be determined based on the identity information and image recognition data, thereby filtering out reservations using false or stolen identity information. Secondly, through positioning technology and signal strength analysis, it can be accurately determined whether the user has actually arrived at the scenic spot, effectively preventing the scenic spot from illegally using the write-off machine to write off without the tourist actually arriving. The control terminal can obtain the network IP address of the mobile terminal, calculate its current location, and further verify the accuracy of the user's location through the positioning navigation hardware device to ensure that the user is indeed in the scenic spot during the write-off process. Furthermore, by comparing the write-off time with the business hours of the scenic spot, the authenticity of the time can be determined, avoiding the write-off behavior during abnormal business hours. In addition, face recognition technology is introduced, and by comparing the face recognition data collected on site with the image recognition data at the time of reservation, the user's on-site authenticity and identity consistency are further verified, effectively preventing the situation where the real ticket holder does not match the reservation information after ticket resale. In addition, the overall operation status of the scenic spot is also taken into account. By analyzing the current write-off data and historical write-off data, the reservation write-off rate is calculated, and the rationality of the write-off behavior is evaluated from a macro perspective. Finally, all the above factors are taken into consideration to calculate a write-off security representation value. Only when the write-off security representation value reaches or exceeds the preset security standard will the write-off process be started, otherwise the write-off will be interrupted. The above multi-level and multi-angle verification mechanism can greatly improve the security and accuracy of the write-off process, and effectively curb the fraudulent behavior of lawbreakers and the illegal operation of scenic spots. It not only effectively solves the problems of identity information abuse, false write-off, ticket reselling, etc., but also improves the credibility and effectiveness of the entire reservation system, which can effectively ensure that only real reservation users can complete the write-off when they actually arrive at the scenic spot, so as to better achieve the goals of scenic spot capacity management, improving tourist experience, and protecting the ecological environment. At the same time, the write-off mechanism can also effectively deter potential violations and reduce the speculative motives of lawbreakers. In summary, the scenic spot safety verification method can effectively reduce the difficulty and cost of scenic spot management, improve the execution efficiency of the reservation system, solve the problem of interference with the normal operation of the existing scenic spot reservation system, effectively improve the effectiveness of the scenic spot reservation system, and facilitate scenic spot management.

[0146] In one implementation manner of this embodiment, after the control terminal sends the positioning request to the mobile terminal, the following steps are also included:

[0147] S201, if the return data of the mobile terminal is received within a preset time period, the control terminal calculates the second current position of the mobile terminal according to the return data;

[0148] S202, the control terminal determines the second scenic spot cancellation area according to the second current position, and obtains the scenic spot location of the second scenic spot cancellation area by querying in a preset scenic spot cancellation location database;

[0149] S203, the control terminal calculates a second interval distance between the mobile terminal and the second scenic spot verification area according to the second current position positioning and the scenic spot positioning;

[0150] S204, the control terminal obtains the first scenic spot verification area where the target positioning navigation hardware device is located and the scenic spot business hour interval corresponding to the first scenic spot verification area, and determines the time authenticity according to the verification time and the scenic spot business hour interval;

[0151] S205, the control terminal determines the user's on-site authenticity in response to receiving the face recognition data sent by the mobile terminal, and calculates the similarity between the face recognition data and the image recognition data;

[0152] S206, the control terminal obtains the current write-off data of the scenic spot;

[0153] S207, the control terminal determines the reservation write-off rate of the scenic spot according to the historical write-off data;

[0154] S208. The control terminal determines a verification safety characterization value of the scenic spot according to the identity authenticity, the second interval distance, the time authenticity, the scene authenticity, the similarity and the reservation verification rate.

[0155] After the control terminal sends the positioning request, it waits for the return data from the mobile terminal within a preset time period. This preset time period can be set to range from a few seconds to tens of seconds, depending on network conditions and requirements. When the return data is received, the control terminal immediately parses the return data to calculate the second current position of the mobile terminal. The return data contains information such as longitude and latitude coordinates, altitude, and positioning accuracy. For example, if the return data shows that the longitude is 120.1234, the latitude is 30.5678, and the accuracy is 10 meters, the control terminal can combine the historical positioning data and map information, and optimize the above position through algorithms such as Kalman filtering, and finally obtain a more accurate second current position positioning.

[0156] The control terminal determines the corresponding second scenic area cancellation area according to the calculated second current position. First, the control terminal compares the coordinates of the second current position with the predefined scenic area geographic boundary. The scenic area geographic boundary information is stored in the preset scenic area cancellation positioning database, and each cancellation area is represented by a polygon or other complex geometric shape. For example, if the second current position is (120.1234, 30.5678), it is checked whether the second current position falls within any predefined cancellation area. Assuming that the coordinates fall into the cancellation area of ​​the "East Gate Entrance Area", the control terminal determines the "East Gate Entrance Area" as the second scenic area cancellation area. Subsequently, the control terminal queries the scenic area cancellation positioning database for detailed information on this area, including geographic coordinates, area area, allowed cancellation equipment types, etc. For example, the scenic area cancellation positioning database returns the center coordinates of the "East Gate Entrance Area" as (120.1230, 30.5680), with an area of ​​500 square meters, and supports mobile device and fixed gate cancellation, so as to ensure that it can accurately determine whether the user is in a suitable location for cancellation operations.

[0157] After determining the second scenic area cancellation area and its corresponding scenic area location, the control terminal calculates the second interval distance between the mobile terminal and the cancellation area. The distance calculation algorithm in the geographic information system (GIS) can be used. Specifically, the Haversine formula can be used, which takes into account the curvature of the earth and can accurately calculate the spherical distance between two geographic coordinate points. In specific implementation, the control terminal substitutes the second current position positioning coordinates of the mobile terminal and the scenic area positioning coordinates (usually the center point or the nearest boundary point of the cancellation area) into the Haversine formula. For example, assuming that the second current position is (120.1234, 30.5678) and the scenic area is (120.1230, 30.5680), the second interval distance obtained after substitution into the formula is 50 meters. It should be noted that the calculation of the distance value not only takes into account the difference in longitude and latitude, but also the difference in altitude, thereby providing a more accurate three-dimensional space distance. In addition, considering the positioning error, an error range, such as ±5 meters, is added to the calculation result to determine whether the user is indeed in a legal cancellation location, thereby effectively preventing improper behaviors such as remote deception.

[0158] In order to determine that the user is redeemed at the appropriate time, first, the control terminal obtains the redemption area information of the first scenic spot where the target positioning navigation hardware device is located. Specifically, it can be implemented through the device management database, which stores the location information of each redemption device. Next, the scenic spot business hour interval corresponding to the redemption area is obtained. The scenic spot business hour interval is stored in the database of the scenic spot management system. For example, assuming that the system determines that the target device is located in the "East Gate Entrance Area", the corresponding business hours are 09:00-17:00 every day. Then, the control terminal compares the user's redemption time with the business hour interval. Assuming that the user's redemption time is 14:30, the relative position of the time point within the business hours is calculated, which can be represented by a standardized value from 0 to 1. In the above example, 14:30 can correspond to a standardized value of 0.6875 (assuming that it is in hours, (14.5-9) / (17-9)). Other factors can also be considered, such as whether it is a peak period, whether it is close to closing time, etc. Finally, a time authenticity score, such as 0.95 (out of a maximum score of 1), is given to reflect the likelihood that the user will redeem the ticket at the appropriate time, which helps prevent time-related fraud, such as using expired tickets or attempting to enter the park during non-business hours.

[0159] After receiving the face recognition data sent by the mobile terminal, the control terminal immediately starts the scene authenticity confirmation and similarity calculation process. First, for the determination of scene authenticity, the quality and characteristics of the received face recognition data are analyzed. This includes checking the image clarity, lighting conditions, face angle, etc. Specifically, deep learning algorithms such as convolutional neural networks (CNNs) can be used to extract facial features and compare them with pre-stored face templates. For example, if it is detected that the image contains a complete frontal face and the lighting is good, a scene authenticity score of 0.9 (out of 1) can be given.

[0160] Next, calculate the similarity between the face recognition data and the previously collected image recognition data. Specifically, a deep learning model, such as a Siamese network, can be used to learn the similarity between two images. Assuming the similarity is 0.85 (the full score is 1), it means that the currently captured face is highly similar to the pre-stored image. In summary, through the comparison of similarity, the user's identity can be effectively verified to prevent deception using static images such as photos. By comprehensively considering the authenticity and similarity of the scene, it can be more reliably confirmed whether the user is actually on the scene and whether the identity matches, thereby significantly improving the security of the verification process.

[0161] In order to calculate the reservation cancellation rate, the control terminal needs to first obtain the current cancellation data of the scenic spot. Specifically, the current cancellation data includes the number of cancelled tickets, the current number of visitors, the usage of various ticket types, etc. The current cancellation data is stored in the central database and is kept updated through a real-time synchronization mechanism. For example, it shows that 500 people have completed cancellation in the past hour, including 300 ordinary tickets, 150 student tickets, and 50 VIP tickets. The control terminal will analyze the above data to calculate some key indicators, such as the cancellation rate (the ratio of the number of cancelled tickets to the number of pre-sold tickets), the peak cancellation speed, etc. Assuming that the total number of pre-sold tickets on the day is 5,000, and 3,000 tickets have been cancelled, the current cancellation rate is 60%. In addition, the control terminal can also compare the above data with historical data for the same period to identify anomalies. For example, if it is found that the current cancellation speed is significantly higher than usual, a system alarm is triggered to alert managers to possible abnormalities.

[0162] Afterwards, the control terminal determines the reservation write-off rate of the scenic spot based on the historical write-off data. First, the control terminal extracts all reservation and write-off records within a certain period of time (such as the past 3 months) from the database. Specifically, the historical write-off data includes the number of daily reservations, the actual number of write-offs, the distribution of write-off times, etc. For example, by analyzing the data of the past 90 days, it was found that the total number of reservations was 100,000 and the actual number of write-offs was 95,000. The overall reservation write-off rate was calculated to be 95% (95,000 / 100,000). In specific implementation, the write-off rates can be calculated separately for different types of tickets (such as ordinary tickets, group tickets, and VIP tickets) for security assessment and to improve overall operational efficiency.

[0163] After obtaining the identity authenticity, the second interval distance, the time authenticity, the on-site authenticity, the similarity and the appointment cancellation rate, the cancellation security characterization value can be calculated by weighted summation. For example, the weights are: identity authenticity (0.2), the second interval distance (0.15), the time authenticity (0.15), the on-site authenticity (0.2), the similarity (0.2), and the appointment cancellation rate (0.1). If the values ​​of the indicators are: identity authenticity (0.95), the second interval distance (converted to a 0-1 score, assuming 0.9), the time authenticity (0.95), the on-site authenticity (0.9), the similarity (0.85), and the appointment cancellation rate (0.95), then the calculated cancellation security characterization value is: 0.95×0.2+0.9×0.15+0.95×0.15+0.9×0.2+0.85×0.2+0.95×0.1=0.9125. The value ranges from 0 to 1, and the closer it is to 1, the higher the security of the write-off process.

[0164] In this implementation, through real-time positioning and distance calculation, it can be ensured that users must be within the specified area to cancel the ticket, preventing remote deception. Face recognition and similarity comparison effectively verify the authenticity of the user's identity. The calculation of time authenticity ensures that the cancellation operation is carried out within the legal time period. The final security representation value combines all key factors and provides a reliable basis for cancellation decisions. It not only effectively prevents ticket fraud and illegal entry, but also greatly improves the security and efficiency of scenic spot ticket management.

[0165] In one implementation of this embodiment, the unique identity identification credential image includes characters and portraits. Before the control terminal receives the user's identity information and image recognition data sent by the mobile terminal, the following steps are included:

[0166] S301, controlling the terminal to convert the user's unique identity identification credential image into a grayscale image, and performing threshold segmentation on the grayscale image to convert the grayscale image into a binary image containing only black and white pixels;

[0167] S302, controlling the terminal to rotate and correct the binary image to ensure that the text in the binary image is arranged horizontally;

[0168] S303, the control terminal locates and segments the character area where the identity information is located and the portrait area where the image recognition data is located in the binary image;

[0169] S304, controlling the terminal to use a preset OCR algorithm to recognize characters in the character area to obtain identity information;

[0170] S305, the control terminal uses a pre-trained deep learning model to extract features from the portrait area to obtain a facial feature vector, wherein the facial feature vector is used to represent image recognition data;

[0171] The control terminal determines the authenticity of the user's identity based on the identity information and image recognition data, including:

[0172] S306, determining target identity information matching the facial feature vector in a preset database;

[0173] S307, calculating the similarity between the target identity information and the user's identity information;

[0174] S308: When the similarity is greater than a preset similarity threshold, a verification request is sent to enable the user to enter a verification password through the verification request;

[0175] S309, in response to receiving the number of times the user inputs the verification password sent by the mobile terminal, calculating the security level using a preset security level calculation formula according to the number of inputs;

[0176] S310: Using a preset identity authenticity calculation formula, the identity authenticity is calculated based on the similarity and security.

[0177] In the image recognition process, the user's unique identity credential image is first converted into a grayscale image. Specifically, the RGB value of each pixel in the color image can be converted into a single grayscale value. The conversion formula is: Grayscale value = 0.299×R+0.587×G+0.114×B, where R, G, and B represent the color intensity of the red, green, and blue channels, respectively. Converting to a grayscale image can reduce the complexity of the image and facilitate subsequent processing.

[0178] Next, threshold segmentation is performed on the grayscale image to convert it into a binary image. Threshold segmentation is an image segmentation method. You can first select a threshold value T, set pixels with grayscale values ​​greater than T to white (255), and set pixels less than or equal to T to black (0). The adaptive threshold method can be used to determine the optimal threshold. For example, suppose there is an 8×8 grayscale image with a grayscale value range of 0-255, and the adaptive threshold method is used to determine the threshold value to be 128. Then, pixel values ​​greater than 128 will be set to 255 (white), and pixel values ​​less than or equal to 128 will be set to 0 (black), and a binary image containing only black and white pixels can be obtained, thereby simplifying the image, highlighting the main features in the image (such as text and contours), and removing background noise. At the same time, binary images require less storage space and are faster to process than grayscale images and color images.

[0179] Since the OCR algorithm is good at processing horizontally arranged text, after obtaining the binary image, the binary image is rotated and corrected to ensure that the text in the image is arranged horizontally. The steps of rotation correction can be: use an edge detection algorithm (such as Canny edge detection) to detect the edges in the image. Then, apply Hough transform to detect the straight lines in these edges. In document images such as identity unique identification certificate images, the text lines usually form long horizontal lines, which form obvious peaks in Hough space. By analyzing these peaks, the main direction of the image can be determined. For example, if the most significant straight line detected forms a 5-degree angle with the horizontal direction, then the entire image needs to be rotated 5 degrees counterclockwise for correction.

[0180] In the rotation operation, for each pixel position after rotation, the weighted average of its four nearest neighboring pixels in the original image is calculated as the new pixel value to ensure that the text is arranged horizontally in the image, improve the overall image layout, and facilitate subsequent region positioning and segmentation.

[0181] Afterwards, the character area where the identity information is located and the portrait area where the image recognition data is located are located and segmented in the corrected binary image. For standardized documents such as identity unique identification credential images, different areas can be located using their fixed layout. In this embodiment, the image is first projected horizontally and vertically, and the number of black pixels in each row and column is calculated. In the horizontal projection, the text lines will form obvious peaks, while in the vertical projection, different information blocks (such as name, ID number, etc.) will form different peak intervals. For example, suppose there is a simplified identity unique identification credential image, and its horizontal projection is shown in the following table:

[0182]

[0183] Referring to the above table, it can be seen that lines 2-3 and lines 6-7 may correspond to two different text lines.

[0184] For the location of the portrait area, a pre-trained neural network model can be used to detect the location of the face in the image. After determining each area, the image cropping algorithm can be used to segment each area. For example, if the bounding box coordinates of the detected face are (x, y, w, h), then these coordinates can be used to extract the face image from the original image.

[0185] For the character area, in the binary image, adjacent black pixels are regarded as belonging to the same character. In this way, each character can be separated.

[0186] After segmenting the character area and the portrait area, the preset OCR (Optical Character Recognition) algorithm is used to recognize the characters in the character area to obtain identity information. OCR is a technology that converts text in an image into machine-editable text.

[0187] In this embodiment, the steps of the OCR algorithm may include: 1. Further image enhancement of the segmented character area, such as denoising, binarization, and thinning. 2. Extracting visual features of the character image using a CNN model. 3. Modeling the visual feature sequence extracted by the CNN model using an RNN model to capture the contextual relationship between characters. 4. Converting the output of the RNN model into a final text sequence using a CTC (Connectionist Temporal Classification) algorithm.

[0188] For example, assume that the image contains the two characters "Zhang San". The OCR algorithm first converts the image into a series of feature vectors, and then uses an RNN model to process these feature vectors and output a sequence of probability distributions. Finally, the decoding algorithm (CTC algorithm) can find the most likely text sequence from this sequence of probability distributions, that is, "Zhang San". In summary, the text information in the image can be converted into a machine-processable text format.

[0189] A pre-trained deep learning model is used to extract features from the portrait area to obtain a face feature vector. The face feature vector is used to represent the image recognition data. Specifically, in this embodiment, the deep learning model is a convolutional neural network (CNN), which can be trained with a large number of face images to extract the key features of the face. Since the method of using a deep learning model to extract face features is relatively common, this application will not elaborate on it.

[0190] After the face feature vector is extracted, the target identity information that matches the extracted face feature vector is determined in a preset database. The preset database is a database of known identity face feature vectors. In the database, each vector is associated with corresponding identity information (such as name, ID, etc.). The search process can include: 1. Calculate the similarity. 2. Sort all the vectors in the database according to the similarity. 3. Only retain the results with a similarity higher than this threshold.

[0191] For example, assume that the face feature vector of user A is [0.1, 0.2, 0.3, …, 0.9], and a certain feature vector stored in the database is [0.11, 0.21, 0.31, …, 0.91]. The result calculated using the cosine similarity is 0.98. If the preset threshold is 0.95, it is considered that the matching target identity information is found, so that the target identity information most similar to the user's face features can be quickly and accurately found from a large amount of data.

[0192] After that, the similarity between the target identity information obtained from the database and the identity information provided by the user is compared. The identity information includes multiple dimensions such as name, age, and gender. For the information in each dimension, the cosine similarity can be used to calculate the similarity. Specifically, in implementation, assume that the target identity information is {name: "Zhang San", age: 30, gender: "male"}, and the identity information provided by the user is {name: "Zhang San", age: 31, gender: "male"}. The calculated name similarity is 1.0, the age similarity is 0.95, and the gender similarity is 1.0. If the weights of each dimension are 0.4, 0.4, and 0.2 respectively, the final similarity is 0.4×1.0 + 0.4×0.95 + 0.2×1.0 = 0.98.

[0193] When the similarity is greater than the preset similarity threshold, a verification request is sent so that the user enters the verification password through the verification request. Specifically, when the calculated similarity exceeds this threshold, the preliminary verification is determined to be passed, and then a verification request is sent to the user. This verification request can be sent in a variety of ways, such as SMS, email, or App push notification. The verification request contains clear instructions requiring the user to enter a preset verification password. In specific implementation, assume that the preset similarity threshold is 0.9, and the similarity calculated in the previous step is 0.95, which exceeds the threshold. At this time, a verification request is generated, the content of which can be: "Dear user, to ensure the security of your account, please enter your verification password below.", to add an extra layer of security verification, effectively reducing the risk of identity fraud.

[0194] In response to the number of times the user inputs the verification password sent by the receiving mobile terminal, the security is calculated using a preset security calculation formula based on the number of inputs. First, the number of times the user inputs the verification password needs to be recorded. Each time the user enters the password, whether correct or not, the number of inputs will be counted. Then, the security is calculated based on the number of inputs using a preset security calculation formula. The security calculation formula is used to characterize the inverse relationship between the number of inputs and the security, that is, the more times the inputs are made, the lower the security. The security calculation formula is:

[0195] Safety = 1 / (1+e (输入次数-3) ).

[0196] In the security calculation formula, when the number of inputs is 0, the security is close to 1; when the number of inputs increases, the security gradually decreases and approaches 0. In specific implementation, assuming that the user enters an incorrect password for the first time, the number of inputs is 1, and the security calculated by substituting it into the formula is about 0.95. If the user enters an incorrect password 5 times in a row, the security will be reduced to about 0.27, in order to dynamically evaluate the security of the user's input behavior. Multiple incorrect inputs will lead to a decrease in security, thereby triggering more stringent security measures to effectively prevent attacks such as brute force cracking.

[0197] Finally, the preset identity authenticity calculation formula is used to calculate the identity authenticity based on the similarity and security. The identity authenticity calculation formula is:

[0198] Identity authenticity = α × similarity + (1-α) × security;

[0199] Among them, α is a weight coefficient used to adjust the relative importance of similarity and security. In specific implementation, assuming α=0.7, the similarity calculated above is 0.95, and the security is 0.95 (the user enters the password correctly for the first time), then the identity authenticity = 0.7×0.95+0.3×0.95=0.95. If the user enters the wrong password multiple times and the security is reduced to 0.27, the identity authenticity will be reduced to 0.7×0.95+0.3×0.27≈0.75.

[0200] This implementation first uses facial recognition technology to perform a preliminary identity match, then compares detailed identity information, and then adds an extra layer of security by verifying the password. At the same time, the system also takes into account the user's input behavior and dynamically adjusts the security assessment. Finally, all factors are combined to calculate an identity authenticity score. This multi-factor, multi-level verification method greatly improves the accuracy and security of identity authentication, and can effectively prevent identity fraud and unauthorized access.

[0201] In one implementation of this embodiment, calculating the first current location of the mobile terminal according to the network IP address includes the following steps:

[0202] S401, dividing the network IP address to obtain multiple IP address segments;

[0203] S402, in a preset geographic location database, querying and summarizing the geographic location corresponding to each IP address segment;

[0204] S403, when the summarized geographical locations are different, querying the ISP information corresponding to the network IP address in a preset ISP database, the ISP information including the ISP name and the ISP location;

[0205] S404, determining at least one network terminal test point according to the summarized geographical locations, wherein each network terminal test point corresponds to a geographical location;

[0206] S405, sending a test data packet to the mobile terminal, so that the mobile terminal sends the test data packet to each network terminal test point;

[0207] S406, in response to receiving the time of returning the data packet from each network terminal test point sent by the mobile terminal, calculating the round-trip delay between the mobile terminal and each network terminal test point, and taking the shortest round-trip delay as the target delay;

[0208] S407, using a preset distance calculation formula to calculate the distance from the mobile terminal to the network terminal test point corresponding to the target delay according to the target delay;

[0209] S408: Determine the first current location according to the distance and the ISP location.

[0210] A standard IPv4 address consists of four 8-bit binary numbers, usually expressed in dotted decimal form, such as 192.168.1.1. The purpose of splitting an IP address is to split it into smaller units for more precise geographic location queries. The splitting method can be adopted in a step-by-step interception from left to right to obtain multiple IP address segments of different lengths. For example, for the IP address 192.168.1.1, the following IP address segments can be obtained: 192, 192.168, 192.168.1, and 192.168.1.1. In specific implementation, string processing functions can be used to gradually build IP address segments of different lengths with dots as separators, so that queries can be gradually narrowed from the widest network range (only one byte) to the most specific single IP address, increasing the flexibility and accuracy of subsequent geographic location queries.

[0211] In the preset geographic location database, the geographic location corresponding to each IP address segment is queried and summarized. Specifically, the geographic location database contains the IP address range and the corresponding geographic location information. The database contains the starting and ending ranges of the IP address segment, as well as the corresponding geographic information such as country, region, and city. During the query process, each IP address segment obtained is compared with the IP range in the database to find the matching geographic location information. For example, for the IP address segment 192.168, the matching item "192.168.0.0-192.168.255.255" may be found in the database, and the corresponding geographic location is "country A, province B". After the query is completed, all query results are summarized to obtain multiple geographic location information of different granularities. In specific implementation, a binary search algorithm can be used to speed up the query process. The summary results can be: national level (country A), provincial level (province B), city level (city C) and other information.

[0212] In the case where the geographical locations after aggregation are different, the ISP information corresponding to the network IP address is queried in the preset ISP database. The ISP information includes the ISP name and ISP location. In this embodiment, it is first necessary to determine whether there are differences in the geographical location information aggregated in the previous step. If there are differences, it means that the geographical location information of the IP address is not clear enough, and it is necessary to further query the ISP (Internet Service Provider) information to assist in positioning. The ISP database contains the ISP information to which each IP address segment belongs, usually including the ISP name, the geographical location of the ISP, etc. During the query process, the complete IP address is used for precise matching in the ISP database. For example, for the IP address 192.168.1.1, the ISP information queried can be: ISP name "A Telecom", ISP location "C City".

[0213] According to the summarized geographical location, at least one network terminal test point is determined, wherein each network terminal test point corresponds to a geographical location. Specifically, the network terminal test point can be selected according to the summarized geographical location information. The network terminal test point is a server or network device pre-deployed at each geographical location for network delay testing. The principle of selecting the test point is to cover all possible geographical locations as much as possible while ensuring the network quality and reliability of the test point. For example, if the summarized geographical locations include "City D" and "City E", then one or more test points with good network quality can be selected in each of these two cities. In specific implementation, a test point database can be maintained, containing information such as the IP address, geographical location, and network status of each test point. The selection process can take into account factors such as the matching degree of the geographical location and the current load of the test point, so as to prepare suitable test points for subsequent network delay testing, and the test point is geographically close to the possible user location, which helps to improve the accuracy of location positioning.

[0214] Send a test data packet to the mobile terminal so that the mobile terminal sends a test data packet to each network terminal test point. First, construct a test instruction packet containing information of all selected test points. The instruction includes parameters such as the IP address list of the test points, the size of the test data packet, and the number of tests. Then, send the test instruction packet to the mobile terminal. After receiving the instruction, the mobile terminal will generate a test data packet according to the instruction content and send these data packets to each specified test point. For example, assuming that three test points are selected, the test instruction may include the IP addresses of the test servers in these three cities, and specify to send 5 ICMP packets of 64 bytes in size. In specific implementation, the UDP or TCP protocol can be used to send the test instruction to the mobile terminal. The client software of the mobile terminal can parse this instruction and perform the corresponding network test, so that the network delay data with multiple possible locations can be collected through the network test actively initiated by the mobile terminal, providing key network feature information for subsequent location positioning.

[0215] In response to the time of receiving the return data packet sent by each network terminal test point from the mobile terminal, the round-trip delay between the mobile terminal and each network terminal test point is calculated, and the shortest round-trip delay is used as the target delay. First, the mobile terminal records the sending time of each test data packet and the time when the corresponding reply is received. Round-trip delay refers to the total time from the sending of a data packet to the receiving of a reply. For each test point, multiple test packets are usually sent and the average is taken to reduce the impact of network fluctuations. The calculation formula is: RTT=(receiving time-sending time) / 2. After repeating this process for all test points, the shortest average RTT is selected as the target delay.

[0216] Afterwards, the preset distance calculation formula is used to calculate the distance from the mobile terminal to the network terminal test point corresponding to the target delay according to the target delay. The distance calculation formula can be: distance = (target delay × speed of light × 0.66) / 2. Among them, 0.66 is an empirical coefficient used to compensate for the delay introduced by the network equipment and the factor that the speed of light in the optical fiber is lower than the speed of light in vacuum.

[0217] Finally, the first current location can be determined based on the distance and ISP location. First, draw a circle with the ISP location as the center and the calculated distance as the radius. The range of the circle is the possible location area. Then, combined with the Geographic Information System (GIS) data, find the most likely location point within the circle. For example, if the ISP location is "City C" and the calculated distance is 50 kilometers, then the possible location point may fall in the suburbs of the city or the surrounding provincial area. The final first current location can be a specific coordinate point or a more likely area.

[0218] This implementation method first obtains preliminary location information through IP address segmentation and geographic location database query, and then uses ISP information to supplement and verify. Next, by selecting appropriate network test points and performing real-time network delay tests, more dynamic and accurate network feature data is obtained. Finally, by converting network delay into physical distance and combining it with ISP location information, the final location estimate is obtained. This method does not rely on traditional positioning technologies such as GPS, but makes full use of network infrastructure and features, and can provide effective location positioning indoors or in environments with weak GPS signals. At the same time, due to the use of multi-source data fusion and real-time network testing, it has high accuracy and adaptability, and can provide more accurate locations.

[0219] In one implementation of this embodiment, the scenic spot business hours include the scenic spot business start time and the scenic spot business end time, and determining the time authenticity according to the verification time and the scenic spot business hours includes the following steps:

[0220] S501, if the verification time is within the scenic spot business hours, determine the time authenticity as 1;

[0221] S502: If the verification time is not within the business hours of the scenic spot, and the verification time is before the business start time of the scenic spot, a first time period between the verification time and the business start time of the scenic spot is calculated, and a first coefficient corresponding to the first time period is determined in a preset authenticity database, and a value of 1 minus the first coefficient is used as the time authenticity;

[0222] S503. If the verification time is not within the business hours of the scenic spot, and the verification time is after the closing time of the scenic spot, then calculate the second time period between the verification time and the closing time of the scenic spot, and determine the second coefficient corresponding to the second time period in the authenticity database, and use the value of 1 minus the second coefficient as the time authenticity.

[0223] In this embodiment, the scenic spot business hours interval consists of the scenic spot start business time and end business time. For example, the business hours of a scenic spot are from 9:00 am to 17:00 pm. The verification time refers to the actual time when tourists use tickets or services. In specific implementation, the time can be converted into a unified format, such as converting all times into minutes (calculated from 0:00 on the same day). For example, if the verification time is 14:30, it can be converted to 870 minutes (14×60+30). Then, this value is compared with the business start time (9:00, converted to 540 minutes) and the end time (17:00, converted to 1020 minutes). If the verification time is greater than or equal to the start time and less than or equal to the end time, it is determined that the verification time is within the business hours interval. In this case, the time authenticity is set to 1, indicating complete credibility. This value of 1 means that the verification behavior occurs within the expected normal time, and there is no abnormal or suspicious situation, so as to quickly identify normal verification behavior.

[0224] If the verification time is not within the business hours of the scenic spot, and the verification time is before the business start time of the scenic spot, the first time period between the verification time and the business start time of the scenic spot is calculated, and the first coefficient corresponding to the first time period is determined in the preset authenticity database, and the value of 1 minus the first coefficient is used as the time authenticity. First, it is necessary to calculate the time difference between the verification time and the business start time of the scenic spot, which is the first time period. The calculation method can be to convert the two time points into minutes and then subtract them. For example, if the verification time is 8:30 (510 minutes) and the business start time of the scenic spot is 9:00 (540 minutes), the first time period is 30 minutes. Next, use the preset authenticity database. The database stores coefficient values ​​corresponding to different time periods, and the coefficient values ​​reflect different degrees of time anomalies. For example, the database may contain the following mapping: 0-15 minutes corresponds to a coefficient of 0.1, 15-30 minutes corresponds to a coefficient of 0.2, 30-60 minutes corresponds to a coefficient of 0.4, and so on. According to the calculated first time period, the corresponding first coefficient is searched in the database. In this example, the first coefficient corresponding to 30 minutes is 0.2. Finally, the time authenticity is calculated using the formula 1-first coefficient. In the above example, the time authenticity is 1-0.2=0.8. This value reflects that although the cancellation time is earlier than the business hours, it still has a certain degree of credibility, so that the cancellation behavior at unconventional times can be quantitatively evaluated, taking into account possible special circumstances (such as early admission) and retaining a certain amount of doubt.

[0225] If the verification time is not within the business hours of the scenic spot, and the verification time is after the closing time of the scenic spot, the second time period between the verification time and the closing time of the scenic spot is calculated, and the second coefficient corresponding to the second time period is determined in the authenticity database, and the value of 1 minus the second coefficient is used as the time authenticity. First, it is necessary to calculate the time difference between the verification time and the closing time of the scenic spot, which is the second time period. The calculation method is similar to the previous step, converting the two time points into minutes and then subtracting them. For example, if the verification time is 18:30 (1110 minutes) and the closing time of the scenic spot is 17:00 (1020 minutes), the second time period is 90 minutes. Next, the preset authenticity database is also used. The database has different coefficient settings for situations beyond the end time, because it is less common to be later than the end time than to be earlier than the start time. For example, the database may contain the following mapping: 0-30 minutes corresponds to a coefficient of 0.2, 30-60 minutes corresponds to a coefficient of 0.4, 60-120 minutes corresponds to a coefficient of 0.6, and so on. According to the calculated second time period, the corresponding second coefficient is searched in the database. In this example, the second coefficient corresponding to 90 minutes is 0.6. Finally, the time authenticity is calculated using the formula 1-second coefficient. In this example, the time authenticity is 1-0.6=0.4. This lower value reflects that the write-off time is significantly later than the business hours, and the credibility is lower, indicating that the write-off behavior that exceeds the normal business hours is given a lower credibility score, which helps to identify possible abnormal or fraudulent behavior.

[0226] This implementation method, through detailed division and quantitative evaluation of the relationship between the write-off time and the business hours of the scenic spot, can not only accurately identify the write-off behavior during normal business hours, but also conduct reasonable credibility evaluation of the write-off during non-routine times, so as to facilitate the safety management of the scenic spot.

[0227] In one implementation of this embodiment, determining the user's on-site authenticity based on face recognition data and calculating the similarity between the face recognition data and the image recognition data includes the following steps:

[0228] S601, performing continuity detection on face recognition data to determine whether the face movement trajectory is normal;

[0229] S602: When the face movement trajectory is normal, a preset liveness algorithm is used to calculate the scene authenticity according to the face recognition data;

[0230] S603, determining a key facial feature vector in the face recognition data, and determining a key facial feature vector in the face feature vector, wherein the key facial feature vector corresponds to the key facial feature vector one-to-one;

[0231] S604, calculating the cosine similarity between each face key part feature vector and the corresponding key facial feature vector;

[0232] S605: Taking the average value of all cosine similarities as the similarity between the face recognition data and the image recognition data.

[0233] In order to determine the user's on-site authenticity, it is first necessary to conduct continuity detection on the face recognition data to determine whether the face motion trajectory is normal, in order to verify whether the recognized object is a real face, rather than a deception method such as a photo or video playback. In specific implementation, it is first necessary to obtain a series of continuous face image frames. These image frames are usually captured by a camera at a fixed frame rate (such as 30 frames per second). Then, these continuous image frames are analyzed to extract key information such as the position, angle and expression of the face in each frame. Next, the above key information is connected in chronological order to form a face motion trajectory. A normal face motion trajectory should show natural and smooth changes. For example, the rotation of the head should be continuous without abrupt jumps. The specific judgment method may include calculating the change speed and acceleration of the face position between adjacent frames to check whether there are abnormal mutations. In addition, the changes in facial expressions can also be analyzed. Under normal circumstances, the changes in expressions should be gradual rather than instantaneous. For example, if a face suddenly changes from a completely frontal view to a side view, or an expression instantly changes from a smile to a frown between two adjacent frames, it is judged as an abnormal trajectory, which can identify deception methods such as using static photos or pre-recorded videos, effectively improving the security and reliability of face recognition.

[0234] After confirming that the face movement trajectory is normal, further verify whether the identified object is a real living body at the scene. The preset liveness algorithm can use a deep learning model to analyze the texture details of the face image. Real faces usually have richer texture information, while printed photos or images on the display lack this detail. Secondly, it can detect subtle movements such as micro-expressions and blinking, which occur naturally on real faces and are difficult to simulate in static images or simple videos. For example, the number of blinks in a certain period of time can be counted. Normal people blink 15-20 times per minute. If no blinking is detected for a long time, it may be a non-live object. In addition, the illumination distribution of the face image can be analyzed to determine whether it is a three-dimensional object. The real face will show natural light and shadow changes at different angles. In specific implementation, the user can be asked to make a series of actions according to the prompts, such as turning the head, blinking or opening the mouth, and then the preset algorithm can be used to detect whether these actions are completed as required, and analyze the naturalness of the actions. Finally, the various indicators are scored comprehensively to obtain a scene authenticity score between 0 and 1, where 1 indicates a high probability of being a real living body and 0 indicates a high probability of deception. For example, if the blinking frequency is normal, the micro-expressions are natural, the lighting distribution conforms to the characteristics of a three-dimensional object, and all required actions are completed accurately, a high score of more than 0.95 can be obtained, which greatly improves the security and reliability of the face recognition system.

[0235] In order to accurately compare the real-time captured face data with the pre-stored face feature data. First, it is necessary to locate the key parts in the face recognition data, which usually include eyes, nose, mouth, chin contour, etc. The positioning method can use a feature point detection algorithm. After locating the key parts, the feature vector of each part can be extracted. A feature vector is a set of values ​​that describes the unique characteristics of the part, including information such as shape, texture, and color. For example, for the eyes, the feature vector may contain information such as the size, shape, and angle of the eye corners; for the nose, it may contain the height of the nose bridge, the width of the nose wing, etc. The above features can be extracted from the image using a deep learning model (such as a convolutional neural network). At the same time, in the pre-stored face feature vector, the corresponding key facial feature vector also needs to be determined. The key facial feature vector has the same dimension and meaning as the real-time captured face key part feature vector for direct comparison. For example, if the real-time captured eye feature vector is a vector containing 100 values, then the stored eye feature vector should also be 100-dimensional, and each dimension corresponds to the same feature. This one-to-one correspondence ensures the accuracy and effectiveness of the comparison.

[0236] In order to quantitatively evaluate the similarity between the real-time captured facial features and the pre-stored facial features, cosine similarity can be used to determine. Cosine similarity is an indicator to measure the similarity of the directions of two vectors, and its value range is between -1 and 1, 1 means the directions are exactly the same, -1 means the directions are completely opposite, and 0 means orthogonal. In face recognition, it is generally believed that the closer the cosine similarity is to 1, the more similar the two feature vectors are. Since the method of using cosine similarity to determine the degree of similarity is relatively common, this application will not go into details. Cosine similarity is calculated for each key part (such as eyes, nose, mouth, etc.), and a series of cosine similarity values ​​can be obtained, which can accurately quantify the similarity of different facial parts.

[0237] The average value of all cosine similarities is used as the similarity between face recognition data and image recognition data to synthesize the similarities of each key part and obtain an overall similarity score. In specific implementation, assume that there are n key parts, and the cosine similarities of each part are s1, s2, ..., sn. Then, calculate the arithmetic mean of these values: S=(s1+s2+...+sn) / n. The average value S is the final face similarity score. After that, a threshold value (such as 0.9) can be set. When the final similarity exceeds the threshold value, it is determined to be the same person.

[0238] This implementation can achieve a secure face recognition process by accurately comparing complex facial features and ultimately giving a comprehensive similarity score.

[0239] In one implementation of this embodiment, the control terminal determines the cancellation safety characterization value of the scenic spot according to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the reservation cancellation rate, including the following steps:

[0240] S701. Obtain user information of the user according to the user's identity information, where the user information includes the number of times the verification is conducted and the verification success rate;

[0241] S702, according to the number of verifications and the verification success rate, respectively determine the weights of identity authenticity, first interval distance, time authenticity, scene authenticity, similarity and appointment verification rate;

[0242] S703. Calculate the weighted average of identity authenticity, first interval distance, time authenticity, on-site authenticity, similarity and reservation cancellation rate according to the corresponding weights of identity authenticity, first interval distance, time authenticity, on-site authenticity, similarity and reservation cancellation rate, and use the weighted average as the cancellation safety representation value of the scenic spot.

[0243] According to the user's identity information, the user's user information is obtained, and the user information includes the number of times the verification has been made and the success rate of the verification, so as to establish the user's historical behavior file based on the user information, so as to evaluate its credibility later. In specific implementation, when the user provides identity information, the corresponding user record can be found in the preset user database. The user record contains the user's verification history data, where the number of verifications refers to the total number of times the user has used the appointment in the past, and the verification success rate is the ratio of the number of successful uses of the appointment and actual visits to the total number of appointments. For example, if a user has made a total of 10 appointments, 8 of which were successfully visited and completed the verification, then the number of verifications is 10 and the verification success rate is 80%. By analyzing the user's historical behavior, the credibility of the user's current behavior can be more accurately predicted, thereby improving the security and efficiency of the entire verification system.

[0244] According to the number of verifications and the verification success rate, the weights of identity authenticity, first interval distance, time authenticity, on-site authenticity, similarity and appointment verification rate are determined respectively. In specific implementation, the weights of various indicators can be adjusted according to the number of verifications and the verification success rate of the user through a weight allocation algorithm. For example, for users with a large number of verifications and a high success rate, the weights of identity authenticity and on-site authenticity can be reduced; on the contrary, for users with a small number of verifications or a low success rate, the weights of identity authenticity and on-site authenticity can be increased.

[0245] According to the corresponding weights of identity authenticity, first interval distance, time authenticity, on-site authenticity, similarity and reservation cancellation rate, the weighted average of identity authenticity, first interval distance, time authenticity, on-site authenticity, similarity and reservation cancellation rate is calculated, and the weighted average is used as the cancellation safety characterization value of the scenic spot. In specific implementation, first ensure that all indicators are standardized to the same range, usually between 0 and 1. Then, use the weighted average formula to calculate the final safety characterization value:

[0246] S=(w1×s1+w2×s2+…+wi×si) / (w1+w2+…+wi);

[0247] Where wi is the weight of the ith indicator, and si is the score of the ith indicator. For example, suppose the following data: identity authenticity s1=0.9, weight w1=0.8; first interval distance s2=0.85, weight w2=0.6; time authenticity s3=0.95, weight w3=0.7; on-site authenticity s4=0.88, weight w4=0.9; similarity s5=0.92, weight w5=0.85; appointment cancellation rate s6=0.8, weight w6=0.5. Substitute into the formula to calculate: S=(0.8×0.9+0.6×0.85+0.7×0.95+0.9×0.88+0.85×0.92+0.5×0.8) / (0.8+0.6+0.7+0.9+0.85+0.5)≈0.892. 0.892 is the write-off safety characterization value.

[0248] This implementation method can flexibly adjust the strictness of the review according to the user's credit status through dynamic weight allocation, effectively improving the efficiency and accuracy of the write-off system.

[0249] In one implementation of this embodiment, according to the number of verifications and the verification success rate, weights of identity authenticity, first interval distance, time authenticity, scene authenticity, similarity and appointment verification rate are determined respectively, including the following steps:

[0250] S801. When the number of times of cancellation is greater than a preset number threshold and the cancellation success rate is less than a preset success rate threshold, a first weight allocation strategy is adopted to respectively allocate first weights to identity authenticity, first interval distance, time authenticity, on-site authenticity, similarity and appointment cancellation rate;

[0251] S802: When the number of verifications is not greater than the number threshold and the verification success rate is less than the success rate threshold, a second weight allocation strategy is adopted to respectively allocate second weights to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the appointment verification rate;

[0252] S803. When the number of cancellations is not greater than the number threshold and the cancellation success rate is greater than the success rate threshold, a third weight allocation strategy is adopted to respectively allocate third weights to identity authenticity, first interval distance, time authenticity, on-site authenticity, similarity, and appointment cancellation rate;

[0253] S804. When the number of verification times is greater than the number threshold and the verification success rate is greater than the success rate threshold, the fourth weight allocation strategy is adopted to allocate fourth weights to identity authenticity, first interval distance, time authenticity, on-site authenticity, similarity and appointment verification rate respectively.

[0254] When the number of times of verification is greater than the preset number threshold, and the verification success rate is less than the preset success rate threshold, the first weight allocation strategy is adopted to allocate the first weights of identity authenticity, first interval distance, time authenticity, on-site authenticity, similarity and appointment verification rate respectively. In this case, although the user has more verification experience, the success rate is low, and there may be potential risks. Therefore, the first weight allocation strategy gives higher weights to identity authenticity and on-site authenticity to strengthen the verification of user identity and on-site conditions. In specific implementation, a preset weight allocation table can be used, for example: identity authenticity weight 0.25, first interval distance weight 0.15, time authenticity weight 0.15, on-site authenticity weight 0.20, similarity weight 0.15, and appointment verification rate weight 0.10. In this way, users with high frequency but low success rate can be more strictly audited, effectively reducing potential fraud risks and improving the security of the verification system.

[0255] When the number of verifications is not greater than the number threshold, and the verification success rate is less than the success rate threshold, the second weight allocation strategy is adopted to allocate the second weights of identity authenticity, first interval distance, time authenticity, on-site authenticity, similarity and appointment verification rate respectively. In this case, the user has less verification experience and a low success rate, and belongs to a high-risk group. Therefore, the second weight allocation strategy will be more stringent, giving all security indicators a higher weight to comprehensively evaluate the user's credibility. In specific implementation, the following weight allocation can be adopted: identity authenticity weight 0.22, first interval distance weight 0.18, time authenticity weight 0.18, on-site authenticity weight 0.20, similarity weight 0.12, and appointment verification rate weight 0.10. By increasing the weights of the first interval distance and time authenticity, a comprehensive and strict review of new users or low-credit users can be carried out to effectively prevent possible fraud and improve the security and reliability of the verification system.

[0256] When the number of verifications is not greater than the number threshold, and the verification success rate is greater than the success rate threshold, the third weight allocation strategy is adopted to allocate the third weights of identity authenticity, first interval distance, time authenticity, on-site authenticity, similarity and appointment verification rate respectively. In this case, although the user has not verified many times, the success rate is high and he is a high-quality user. Therefore, the third weight allocation strategy gives users a better experience while ensuring basic security. In specific implementation, the following weight allocation can be adopted: identity authenticity weight 0.20, first interval distance weight 0.15, time authenticity weight 0.15, on-site authenticity weight 0.18, similarity weight 0.12, appointment verification rate weight 0.20. By reducing the weight of security indicators and increasing the weight of appointment verification rate, a more convenient verification experience is provided for honest and trustworthy new users.

[0257] When the number of verifications is greater than the number threshold, and the verification success rate is greater than the success rate threshold, the fourth weight allocation strategy is adopted to allocate the fourth weights of identity authenticity, first interval distance, time authenticity, on-site authenticity, similarity and appointment verification rate respectively. In this case, the user has both rich verification experience and a high success rate, and is a high-credit user. Therefore, the fourth weight allocation strategy is the most relaxed, aiming to provide the best verification experience for such users. In specific implementation, the following weight allocation can be adopted: identity authenticity weight 0.15, first interval distance weight 0.12, time authenticity weight 0.12, on-site authenticity weight 0.15, similarity weight 0.16, and appointment verification rate weight 0.30. This allocation method greatly reduces the weights of various security indicators, while significantly increasing the weight of the appointment verification rate, so as to simplify the verification process of high-credit users to the greatest extent, thereby providing the fastest verification service for long-term honest users, effectively improving user satisfaction and verification efficiency, and ensuring the basic security of verification through appropriate security checks, achieving the best balance between security and user experience.

[0258] The layered weight allocation strategy of this implementation method can dynamically adjust the strictness of security audits based on the user's historical behavior, which not only ensures strict control of high-risk users, but also provides a convenient experience for honest users, achieving a refined balance between security and user experience, and greatly improving the intelligence and adaptability of the verification system.

[0259] In one implementation of this embodiment, the first weight allocation strategy includes the following steps:

[0260] S901, obtaining initial weights corresponding to identity authenticity, first interval distance, time authenticity, scene authenticity, similarity, and reservation cancellation rate respectively;

[0261] S902. Calculate the first weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity, and the appointment cancellation rate, respectively, using the first weight allocation formula according to the successful cancellation rate and the initial weight, wherein the first weights corresponding to the identity authenticity and the scene authenticity are both greater than the corresponding initial weights, the first weight of the similarity is consistent with the corresponding initial weight, and the first weights corresponding to the time authenticity, the first interval distance, and the appointment cancellation rate are both less than the corresponding initial weights;

[0262] The second weight distribution strategy includes:

[0263] S903, obtaining initial weights corresponding to identity authenticity, first interval distance, time authenticity, scene authenticity, similarity, and reservation cancellation rate respectively;

[0264] S904. According to the successful verification rate and the initial weight, a second weight allocation formula is used to calculate the second weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the appointment verification rate, respectively. The second weights corresponding to the identity authenticity, the scene authenticity and the similarity are all greater than the corresponding initial weights, the second weights corresponding to the time authenticity and the first interval distance are consistent with the corresponding initial weights, and the second weight corresponding to the appointment verification rate is less than the corresponding initial weight.

[0265] The third weight distribution strategy includes:

[0266] S905, obtaining initial weights corresponding to identity authenticity, first interval distance, time authenticity, scene authenticity, similarity, and reservation cancellation rate respectively;

[0267] S906. According to the successful verification rate and the initial weight, a third weight allocation formula is used to calculate the third weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the appointment verification rate, respectively. The third weights corresponding to the identity authenticity and the scene authenticity are both smaller than the corresponding initial weights, the third weight of the similarity is consistent with the corresponding initial weight, and the third weights corresponding to the time authenticity, the first interval distance and the appointment verification rate are both larger than the corresponding initial weights.

[0268] The fourth weight distribution strategy includes:

[0269] S907, obtaining initial weights corresponding to identity authenticity, first interval distance, time authenticity, scene authenticity, similarity, and reservation cancellation rate respectively;

[0270] S908. According to the successful verification rate and the initial weight, the fourth weight allocation formula is used to calculate the fourth weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the site authenticity, the similarity and the appointment verification rate, respectively. Among them, the fourth weights corresponding to the identity authenticity and the site authenticity are both smaller than the corresponding initial weights, the fourth weight of the similarity is consistent with the corresponding initial weight, the fourth weights corresponding to the time authenticity, the first interval distance and the time authenticity are both larger than the corresponding initial weights, and the fourth weight of the appointment verification rate is larger than the fourth weights corresponding to the first interval distance and the time authenticity.

[0271] For the first weight allocation strategy, first obtain the initial weights corresponding to identity authenticity, first interval distance, time authenticity, on-site authenticity, similarity, and appointment cancellation rate. The initial weight is a pre-set benchmark value based on historical data analysis, reflecting the degree of influence of each indicator on the cancellation result under normal circumstances. For example, the initial weights can be set as: identity authenticity 0.20, first interval distance 0.15, time authenticity 0.15, on-site authenticity 0.20, similarity 0.15, appointment cancellation rate 0.15. The sum of the initial weights is 1.

[0272] Afterwards, the first weight of each indicator is calculated using the first weight allocation formula based on the successful verification rate and the initial weight. The first weight allocation formula is designed for situations where the number of verifications is high but the success rate is low, and is intended to improve security. In specific implementation, the following formula can be used: new weight = initial weight + (1-successful verification rate) × adjustment coefficient. Among them, the weight is increased when the adjustment coefficient is positive, and the weight is reduced when it is negative. For example, assuming that the successful verification rate is 0.6, the adjustment coefficients are: identity authenticity 0.1, on-site authenticity 0.1, similarity 0, time authenticity -0.05, first interval distance -0.05, appointment verification rate -0.1. Then, the new weight of identity authenticity is 0.20 + (1-0.6) × 0.1 = 0.24. In this way, the weights of identity authenticity and on-site authenticity are increased, similarity remains unchanged, and the weights of other indicators are reduced. In this way, identity and on-site verification can be strengthened for high-risk users, effectively improving the system's anti-fraud ability.

[0273] For the second weight allocation strategy, step S903 is similar to step S901, and the initial weights corresponding to the indicators are first obtained. The initial weights can be set the same as in step S901, that is, identity authenticity 0.20, first interval distance 0.15, time authenticity 0.15, site authenticity 0.20, similarity 0.15, and reservation cancellation rate 0.15.

[0274] Afterwards, according to the successful verification rate and the initial weight, the second weight allocation formula is used to calculate the second weight of each indicator. The second weight allocation strategy is aimed at users with low verification times and low success rate, who belong to the high-risk group. The second weight allocation formula is used to comprehensively improve the weight of security indicators. In specific implementation, the following formula can be used: new weight = initial weight + (1-successful verification rate) × adjustment coefficient. Assuming that the successful verification rate is 0.4, the adjustment coefficients are: identity authenticity 0.15, on-site authenticity 0.15, similarity 0.1, time authenticity 0, first interval distance 0, appointment verification rate -0.4. Then, the new weight of identity authenticity is 0.20 + (1-0.4) × 0.15 = 0.29. In this way, the weights of identity authenticity, on-site authenticity and similarity are significantly increased, the time authenticity and the first interval distance remain unchanged, and the weight of the appointment verification rate is greatly reduced, so that the demand for more stringent security measures can be taken for high-risk new users, effectively improving the system's ability to prevent potential fraud.

[0275] For the third weight allocation strategy, step S905 is similar to S901 and S903, and firstly, the initial weight corresponding to each indicator is obtained. The initial weight setting can be consistent, that is, identity authenticity 0.20, first interval distance 0.15, time authenticity 0.15, site authenticity 0.20, similarity 0.15, and reservation cancellation rate 0.15.

[0276] Then, according to the successful verification rate and the initial weight, the third weight allocation formula is used to calculate the third weight of each indicator. The third weight allocation strategy is aimed at users with low verification times but high success rates. The third weight allocation formula is used to appropriately reduce the weights of some security indicators while increasing the weights of convenience-related indicators. In specific implementation, the following formula can be used: new weight = initial weight + successful verification rate × adjustment coefficient. Assuming that the successful verification rate is 0.8, the adjustment coefficients are: identity authenticity -0.1, on-site authenticity -0.1, similarity 0, time authenticity 0.05, first interval distance 0.05, and appointment verification rate 0.1. Then, the new weight of identity authenticity is 0.20+0.8×(-0.1)=0.12. Through the third weight allocation formula, the weights of identity authenticity and on-site authenticity are reduced, similarity remains unchanged, and the weights of time authenticity, first interval distance and appointment verification rate are increased, so that more convenient services can be provided to potential high-quality users, while maintaining moderate security, effectively balancing user experience and system security.

[0277] For the fourth weight allocation strategy, step S907 is similar to S901, S903 and S905. First, the initial weights corresponding to each indicator are obtained. The initial weight settings can remain consistent. Afterwards, the fourth weight of each indicator is calculated using the fourth weight allocation formula based on the successful verification rate and the initial weight. The fourth weight allocation strategy is aimed at users with a high number of verifications and a high success rate. The fourth weight allocation formula is used to significantly reduce the weight of safety indicators, while significantly increasing the weight of convenience-related indicators, especially the appointment verification rate.

[0278] In specific implementation, the following formula can be used: New weight = initial weight + successful verification rate × adjustment coefficient. Assuming the successful verification rate is 0.9, the adjustment coefficients are: identity authenticity -0.15, on-site authenticity -0.15, similarity 0, time authenticity 0.05, first interval distance 0.05, and appointment verification rate 0.2. Then, the new weight of identity authenticity is 0.20+0.9×(-0.15)=0.065, and the new weight of appointment verification rate is 0.15+0.9×0.2=0.33. In this way, the weights of identity authenticity and on-site authenticity are greatly reduced, similarity remains unchanged, time authenticity and first interval distance are slightly increased, and the weight of appointment verification rate is significantly increased, thereby providing maximum convenience to high-credit users while maintaining minimum security checks and effectively improving user satisfaction.

[0279] This implementation method uses a multi-level weight allocation strategy to dynamically adjust the importance of each indicator based on the user's verification history, achieving a refined balance between security and convenience. For high-risk users, identity authentication and on-site verification are strengthened; for high-quality users, the verification process is simplified and efficiency is improved. This not only improves verification security, but also optimizes user experience, effectively improving the intelligence and adaptability of the verification system.

[0280] Reference Figure 3 The embodiment of the present application also provides a scenic spot safety verification system, which includes a control terminal, a mobile terminal and at least one positioning and navigation hardware device, and the control terminal includes:

[0281] a memory configured to store instructions; and

[0282] The processor is configured to call instructions from the memory and implement the above-mentioned scenic spot safety verification method when executing the instructions.

[0283] In this embodiment, the control terminal and the mobile terminal can be connected via a wireless network (such as Wi-Fi or 4G / 5G network) so that scenic spot staff can use mobile terminals (such as smartphones or tablets) to perform verification operations at any location in the scenic spot without being restricted by fixed locations. An application is installed in the mobile terminal for performing verification operations, collecting on-site information, verifying the identity of tourists, and other tasks.

[0284] The positioning and navigation hardware device is connected to both the control terminal and the mobile terminal. The positioning and navigation hardware device may include a GPS receiver, a Bluetooth beacon or other indoor positioning systems to provide accurate location information to help determine the specific location of tourists and staff.

[0285] The control terminal is connected to the positioning and navigation hardware device through the network (wired LAN or wireless network) to receive real-time location data. At the same time, the mobile terminal can also communicate directly with the positioning and navigation hardware device through the network (wired LAN or wireless network) to obtain more accurate local location information.

[0286] An embodiment of the present application also provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to enable a machine to execute the above-mentioned scenic spot safety verification method.

[0287] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0288] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0289] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0290] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0291] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0292] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0293] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0294] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0295] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A scenic spot safety verification method, characterized in that: Applied to a scenic spot safety verification system, the scenic spot safety verification system includes a control terminal, a mobile terminal and at least one positioning and navigation hardware device, and the method includes: The control terminal acquires the network IP address and verification time of the mobile terminal in response to receiving the user's identity information and image recognition data sent by the mobile terminal, wherein the identity information and the image recognition data are obtained by performing image recognition on the user's unique identity identification credential image; The control terminal determines the authenticity of the user's identity based on the identity information and the image recognition data; The control terminal sends a positioning request to the mobile terminal, and if no return data from the mobile terminal is received within a preset time period, calculates a first current position of the mobile terminal according to the network IP address; The control terminal determines a target positioning and navigation hardware device in at least one of the positioning and navigation hardware devices through the first current position positioning; The control terminal obtains the signal strength between the target positioning and navigation hardware device and the mobile terminal, and calculates a first interval distance between the mobile terminal and the target positioning and navigation hardware device according to the signal strength; The control terminal obtains the first scenic spot verification area where the target positioning navigation hardware device is located and the scenic spot business time interval corresponding to the first scenic spot verification area, and determines the time authenticity according to the verification time and the scenic spot business time interval; The control terminal, in response to receiving the face recognition data sent by the mobile terminal, determines the user's on-site authenticity according to the face recognition data, and calculates the similarity between the face recognition data and the image recognition data; The control terminal obtains the current write-off data of the scenic spot; The control terminal determines the reservation write-off rate of the scenic spot according to the historical write-off data; The control terminal determines the cancellation safety representation value of the scenic spot according to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the reservation cancellation rate; The control terminal starts a preset write-off process when the security characterization value is greater than or equal to a preset security standard value; When the security characterization value is less than a preset security standard value, the control terminal interrupts the verification process.

2. The method according to claim 1, characterized in that After the control terminal sends the positioning request to the mobile terminal, the method further includes: If the return data from the mobile terminal is received within a preset time period, the control terminal calculates the second current position of the mobile terminal according to the return data; The control terminal determines a second scenic spot cancellation area according to the second current position, and obtains the scenic spot location of the second scenic spot cancellation area by querying in a preset scenic spot cancellation location database; The control terminal calculates a second interval distance between the mobile terminal and the second scenic spot verification area according to the second current position positioning and the scenic spot positioning; The control terminal obtains the first scenic spot verification area where the target positioning navigation hardware device is located and the scenic spot business time interval corresponding to the first scenic spot verification area, and determines the time authenticity according to the verification time and the scenic spot business time interval; The control terminal determines the user's on-site authenticity in response to receiving the face recognition data sent by the mobile terminal, and calculates the similarity between the face recognition data and the image recognition data; The control terminal obtains the current write-off data of the scenic spot; The control terminal determines the reservation write-off rate of the scenic spot according to the historical write-off data; The control terminal determines a verification safety characterization value of the scenic spot according to the identity authenticity, the second interval distance, the time authenticity, the scene authenticity, the similarity and the reservation verification rate.

3. The method according to claim 1, characterized in that The unique identity identification credential image includes characters and portraits, and before the control terminal receives the user's identity information and image recognition data sent by the mobile terminal, it includes: The control terminal converts the user's unique identity identification credential image into a grayscale image, and performs threshold segmentation on the grayscale image to convert the grayscale image into a binary image containing only black and white pixels; The control terminal rotates and corrects the binary image to ensure that the characters in the binary image are arranged horizontally; The control terminal locates and segments the character area where the identity information is located and the portrait area where the image recognition data is located in the binary image; The control terminal uses a preset OCR algorithm to recognize characters in the character area to obtain identity information; The control terminal uses a pre-trained deep learning model to extract features from the portrait area to obtain a facial feature vector, wherein the facial feature vector is used to represent the image recognition data; The control terminal determines the authenticity of the user's identity based on the identity information and the image recognition data, including: Determining target identity information matching the facial feature vector in a preset database; Calculating the similarity between the target identity information and the identity information of the user; When the similarity is greater than a preset similarity threshold, sending a verification request so that the user enters a verification password through the verification request; In response to receiving the number of times the user inputs the verification password sent by the mobile terminal, calculating the security level using a preset security level calculation formula according to the number of inputs; The identity authenticity is calculated using a preset identity authenticity calculation formula based on the similarity and the security level.

4. The method according to claim 1, characterized in that: The calculating the first current position of the mobile terminal according to the network IP address includes: Segmenting the network IP address to obtain multiple IP address segments; In the preset geographic location database, query and summarize the geographic location corresponding to each IP address segment; In the case where the summarized geographical locations are different, querying a preset ISP database for ISP information corresponding to the network IP address, the ISP information including the ISP location; Determine at least one network terminal test point according to the summarized geographical locations, wherein each network terminal test point corresponds to one geographical location; Sending a test data packet to a mobile terminal so that the mobile terminal sends the test data packet to each of the network terminal test points; In response to receiving the time of returning the data packet from each network terminal test point sent by the mobile terminal, calculating the round-trip delay between the mobile terminal and each network terminal test point, and taking the shortest round-trip delay as the target delay; Using a preset distance calculation formula, the distance from the mobile terminal to the network terminal test point corresponding to the target delay is calculated according to the target delay; A first current position location is determined based on the distance and the ISP position.

5. The method according to claim 1, characterized in that The scenic spot business hours include the scenic spot business start time and the scenic spot business end time, and determining the time authenticity according to the verification time and the scenic spot business hours includes: If the verification time is within the business hours of the scenic spot, the time authenticity is determined to be 1; If the verification time is not within the business hours of the scenic spot, and the verification time is before the business start time of the scenic spot, a first time period between the verification time and the business start time of the scenic spot is calculated, and a first coefficient corresponding to the first time period is determined in a preset authenticity database, and a value of 1 minus the first coefficient is used as the time authenticity; If the verification time is not within the business hours of the scenic spot, and the verification time is after the closing time of the scenic spot, a second time period between the verification time and the closing time of the scenic spot is calculated, and a second coefficient corresponding to the second time period is determined in the authenticity database, and the value of 1 minus the second coefficient is used as the time authenticity.

6. The method according to claim 3, characterized in that The determining the user's on-site authenticity according to the face recognition data and calculating the similarity between the face recognition data and the image recognition data includes: Performing continuity detection on the face recognition data to determine whether the face movement trajectory is normal; When the face movement trajectory is normal, a preset liveness algorithm is used to calculate the scene authenticity according to the face recognition data; Determining a key facial feature vector in the face recognition data, and determining a key facial feature vector in the face feature vector, wherein the key facial feature vector corresponds to the key facial feature vector in a one-to-one manner; Calculating the cosine similarity between each of the key facial feature vectors and the corresponding key facial feature vector; An average value of all the cosine similarities is used as the similarity between the face recognition data and the image recognition data.

7. The method according to claim 1, characterized in that The control terminal determines the cancellation safety characterization value of the scenic spot according to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the reservation cancellation rate, including: According to the identity information of the user, obtaining the user information of the user, the user information including the number of verifications and the verification success rate; Determining weights of the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the reservation verification rate respectively according to the number of verifications and the verification success rate; According to the weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the on-site authenticity, the similarity and the reservation cancellation rate, a weighted average of the identity authenticity, the first interval distance, the time authenticity, the on-site authenticity, the similarity and the reservation cancellation rate is calculated, and the weighted average is used as a value representing the cancellation safety of the scenic spot.

8. The method according to claim 7, characterized in that The step of determining the weights of the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the reservation verification rate respectively according to the verification times and the verification success rate includes: When the number of verifications is greater than a preset number threshold, and the verification success rate is less than a preset success rate threshold, a first weight allocation strategy is adopted to respectively allocate first weights to the identity authenticity, the first interval distance, the time authenticity, the on-site authenticity, the similarity, and the appointment verification rate; When the number of verifications is not greater than the number threshold and the verification success rate is less than the success rate threshold, a second weight allocation strategy is adopted to respectively allocate second weights to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the appointment verification rate; When the number of verifications is not greater than the number threshold and the verification success rate is greater than the success rate threshold, a third weight allocation strategy is adopted to respectively allocate third weights to the identity authenticity, the first interval distance, the time authenticity, the on-site authenticity, the similarity and the appointment verification rate; When the number of verifications is greater than the number threshold and the verification success rate is greater than the success rate threshold, the fourth weight allocation strategy is adopted to respectively allocate fourth weights to the identity authenticity, the first interval distance, the time authenticity, the on-site authenticity, the similarity and the appointment verification rate.

9. The method according to claim 8, characterized in that The first weight allocation strategy includes: Obtaining initial weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity, and the reservation cancellation rate, respectively; According to the successful verification rate and the initial weight, a first weight allocation formula is used to calculate the first weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the appointment verification rate, respectively, wherein the first weights corresponding to the identity authenticity and the scene authenticity are both greater than the corresponding initial weights, the first weight of the similarity is consistent with the corresponding initial weight, and the first weights corresponding to the time authenticity, the first interval distance and the appointment verification rate are both less than the corresponding initial weights; The second weight allocation strategy includes: Obtaining initial weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity, and the reservation cancellation rate, respectively; According to the successful verification rate and the initial weight, a second weight allocation formula is used to calculate the second weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the appointment verification rate, respectively, wherein the second weights corresponding to the identity authenticity, the scene authenticity and the similarity are all greater than the corresponding initial weights, the second weights corresponding to the time authenticity and the first interval distance are both consistent with the corresponding initial weights, and the second weight corresponding to the appointment verification rate is less than the corresponding initial weight; The third weight allocation strategy includes: Obtaining initial weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity, and the reservation cancellation rate, respectively; According to the successful verification rate and the initial weight, a third weight allocation formula is used to calculate the third weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity and the appointment verification rate, respectively, wherein the third weights corresponding to the identity authenticity and the scene authenticity are both smaller than the corresponding initial weights, the third weight of the similarity is consistent with the corresponding initial weight, and the third weights corresponding to the time authenticity, the first interval distance and the appointment verification rate are both larger than the corresponding initial weights; The fourth weight distribution strategy includes: Obtaining initial weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the scene authenticity, the similarity, and the reservation cancellation rate, respectively; According to the successful verification rate and the initial weight, the fourth weight allocation formula is adopted to calculate the fourth weights corresponding to the identity authenticity, the first interval distance, the time authenticity, the on-site authenticity, the similarity and the appointment verification rate, respectively, wherein the fourth weights corresponding to the identity authenticity and the on-site authenticity are both smaller than the corresponding initial weights, the fourth weight of the similarity is consistent with the corresponding initial weight, the fourth weights corresponding to the time authenticity, the first interval distance and the time authenticity are both greater than the corresponding initial weights, and the fourth weight of the appointment verification rate is greater than the fourth weights corresponding to the first interval distance and the time authenticity.

10. A scenic spot safety verification system, characterized in that: The scenic area safety verification system includes a control terminal, a mobile terminal and at least one positioning navigation hardware device, and the control terminal includes: a memory configured to store instructions; and A processor is configured to call the instructions from the memory and to implement the scenic spot safety verification method according to any one of claims 1 to 9 when executing the instructions.

Images