A park identity authentication management method

Abstract

This invention relates to a big data processing method, disclosing a park identity authentication management method. This method includes determining whether the license plate data matches a pre-stored registered license plate. If they match, it determines whether the decrypted vehicle body data matches the pre-stored vehicle body data corresponding to the license plate data. If they do not match, it determines whether the facial image matches either the pre-stored facial data corresponding to the decrypted vehicle body data or the pre-stored facial data corresponding to the license plate data. If they match, the barrier is raised when the vehicle enters the park. This invention integrates license plate data, vehicle body data, and facial image authentication through the above-mentioned park authentication method. First, it overcomes the inconvenience of repeated registration and updates required when park owners change vehicles or license plates. Second, to prevent the problem of external vehicles using counterfeit license plates, it uses vehicle posture to eliminate vehicles using counterfeit plates, thereby increasing park security. Third, through asymmetric encryption, it prevents the park management system from being attacked and unable to recognize vehicles properly.

Description

Technical Field

[0001] This invention relates to a big data processing method, and more particularly to a campus identity authentication management method. Background Technology

[0002] Although our traditional security screening system includes facial recognition and license plate recognition systems, these are all independent entities that cannot be integrated. This requires more security personnel and results in a waste of security screening costs.

[0003] Furthermore, although some registered vehicles are not using counterfeit license plates, the large number of registrants makes it easy for them to switch vehicles or license plates. Each data update cycle is long, the review process is complex, and the costs are high.

[0004] Furthermore, in addition to examining the vehicle's appearance, driving habits can also reflect the user's identity to some extent. Conversely, without considering driving habits, it is difficult to inspect stolen vehicles or vehicles with identical license plates, leaving certain security risks.

[0005] In terms of security, more security checks mean greater security, but also less convenience; while fewer security checks mean greater convenience, but also less security.

[0006] Therefore, we need a campus identity authentication management method that simultaneously considers security, convenience, and combines multiple identity authentication methods. Summary of the Invention

[0007] The technical problem to be solved by the present invention is to provide a campus identity authentication management method that simultaneously takes into account security, convenience and multiple identity authentication.

[0008] This invention provides a campus identity authentication management method, including...

[0009] S100: Obtain vehicle license plate data, vehicle body data, and entry time;

[0010] S200: Determine whether the license plate data matches the pre-stored registered license plate. If it matches, proceed to S300; if it does not match, proceed to S420.

[0011] S300: Determine whether the decrypted vehicle body data matches the pre-stored vehicle body data corresponding to the pre-stored license plate data. If they match, proceed to S500; otherwise, proceed to S410.

[0012] S410. Determine whether the face image matches one of the pre-stored face data corresponding to the decrypted vehicle body data or the pre-stored face data corresponding to the license plate data. If they match, proceed to S500.

[0013] S420. Determine whether the face data corresponding to the decrypted vehicle body data matches the face image. If they match, proceed to S500.

[0014] S500: Raise the barrier when the vehicle enters the park; raise the barrier after obtaining the payment information corresponding to the payment QR code when the vehicle leaves the park.

[0015] This invention provides a method for park identity authentication management, wherein determining whether the decrypted vehicle body data matches the pre-stored vehicle body data corresponding to the pre-stored license plate data includes:

[0016] Determine whether the color and shape of the decrypted vehicle body data match the color and shape of the pre-stored vehicle body data corresponding to the pre-stored license plate data. If they do not match, issue an alarm signal.

[0017] If a match is found, it is determined whether the time interval between the identification of the license plate data and the last identification of another license plate data is less than a first preset threshold. If it is less, the process jumps to S500. If it is not less, a trajectory point is configured for the license plate in the vehicle body data. The position where the trajectory point appears this time is configured as the first position, and the position where the trajectory point is stationary for a preset time is configured as the second position. A first vector is constructed from the first position to the second position. It is determined whether the first vector is similar to the second vector of the pre-stored vehicle body data corresponding to the pre-stored license plate data. If they are similar, the process jumps to S500. If they are not similar, an alarm signal is issued.

[0018] This invention provides a campus identity authentication management method, wherein determining whether the decrypted vehicle data matches a facial image, and if they match, includes:

[0019] Determine whether the color and shape of the decrypted vehicle body data match the color and shape of the pre-stored vehicle body data corresponding to the pre-stored license plate data. If they do not match, issue an alarm signal.

[0020] If a match is found, it is determined whether the time interval between the identification of the license plate data and the last identification of another license plate data is less than a first preset threshold (because the trajectory points collected in a short time due to traffic jams and queuing are not representative). If it is less than the threshold, the process jumps to S500. If it is not less than the threshold, a trajectory point is configured for the license plate in the vehicle body data. The position where the trajectory point appears this time is configured as the first position, and the position where the trajectory point is stationary for a preset time is configured as the second position. A first vector is constructed from the first position to the second position. It is determined whether the first vector is similar to the second vector of the pre-stored vehicle body data corresponding to the pre-stored face data that matches the face image. If they are similar, the process jumps to S500. If they are not similar, an alarm signal is issued.

[0021] This invention provides a park identity authentication management method, wherein the update steps of S410 ("if a match is found, proceed to S500 and update the pre-stored face data corresponding to the pre-stored decrypted vehicle body data and the pre-stored face data corresponding to the pre-stored license plate data according to the face image") and S420 ("if a match is found, proceed to S500 and update the license plate data according to the decrypted vehicle body data and the face image") include the following steps:

[0022] S431. Determine whether the total number of updates to the license plate data, vehicle body data, and face data of the user corresponding to the face image within a first preset time exceeds a first threshold. If not, execute the update step; if so, then: the step of "determining whether the first vector is similar to the second vector of the pre-stored vehicle body data corresponding to the pre-stored license plate data; if similar, jump to S500; if not similar, issue an alarm signal." includes:

[0023] Using the viewfinder of the video that acquires the vehicle body data of the first vector as the first map base, the height and radius of the cone are generated by multiplying the first vector by a length weighting coefficient, and the two cones are respectively placed at the starting point and the ending point of the first vector, wherein the center of the base of the cone coincides with the starting point and the ending point of the first vector, and a first three-dimensional map of the first vector is generated on the first map base.

[0024] Using the viewfinder of the video with the pre-stored second vector vehicle body data as the second map base, the height and radius of the cone are generated by multiplying the second vector by the length weight coefficient, and the two cones are respectively placed at the starting point and the ending point of the second vector, wherein the center of the base of the cone coincides with the starting point and the ending point of the second vector, and a second three-dimensional map of the second vector is generated on the map base.

[0025] Determine whether there exists a second percentage in at least one of the second three-dimensional maps that does not overlap with the first three-dimensional map and has a volume smaller than the total volume of the first three-dimensional map. The second percentage is calculated as the first percentage * the first threshold / the total number of updates. If such a percentage exists, the map is considered similar. If not, the map is considered dissimilar. Then, delete the pre-stored license plate data, vehicle body data, and face data of the user corresponding to the face image.

[0026] This invention provides a park identity authentication management method, wherein the update steps of S410 ("if a match is found, proceed to S500 and update the pre-stored face data corresponding to the pre-stored decrypted vehicle body data and the pre-stored face data corresponding to the pre-stored license plate data according to the face image") and S420 ("if a match is found, proceed to S500 and update the license plate data according to the decrypted vehicle body data and the face image") include the following steps:

[0027] S431. Determine whether the total number of updates to the license plate data, vehicle body data, and facial data of the user corresponding to the facial image within a first preset time exceeds a first threshold. If not, execute the update step; if so, then: in the step "S300, the vehicle body data and park entry time are encrypted in a first asymmetric form based on the hash value of the license plate data", the following steps are included:

[0028] S301. Use the SHA256 algorithm to obtain the first hash value of the license plate data;

[0029] S302, Take the last n bits of the first hash value 2 The number of bits, where n = n + 1, and the n 2 No more than 256;

[0030] S303, vehicle body data, and / or entry time are configured into an n*n grid, and the sequence number of each cell is configured sequentially to obtain the first n... 2 Grid;

[0031] S304. Compare the last n bits of the first hash value from back to front. 2 The first hash value is sorted by the size of the first bit and the last n bits of the hash value in descending order. 2 The first sequence is generated by bits, where, in the case of the same numerical value, the later one in the first hash value is considered larger;

[0032] S305, the first n 2 The grid numbers are adjusted according to the first sequence to obtain the second n. 2 The grid is considered as completing the first asymmetric encryption.

[0033] This invention provides a park identity authentication management method, wherein the update steps of S410 ("if a match is found, proceed to S500 and update the pre-stored face data corresponding to the pre-stored decrypted vehicle body data and the pre-stored face data corresponding to the pre-stored license plate data according to the face image") and S420 ("if a match is found, proceed to S500 and update the license plate data according to the decrypted vehicle body data and the face image") include the following steps:

[0034] S431. Determine whether the total number of updates of the license plate data, vehicle body data, and face data of the user corresponding to the face image within a first preset time exceeds a first threshold. If so, delete the pre-stored license plate data, vehicle body data, and face data of the user corresponding to the face image.

[0035] The present invention provides a method for park identity authentication management, wherein the park is an industrial park or an office park.

[0036] Secondly, the present invention provides a park identity authentication management system, including...

[0037] The camera system is used to acquire vehicle license plate data, vehicle body data, and entry time.

[0038] A processor, which performs the following steps:

[0039] S200: Determine whether the license plate data matches the pre-stored registered license plate. If it matches, proceed to S300; if it does not match, proceed to S420.

[0040] S300: Determine whether the decrypted vehicle body data matches the pre-stored vehicle body data corresponding to the pre-stored license plate data. If they match, proceed to S500; otherwise, proceed to S410.

[0041] S410. Determine whether the face image matches one of the pre-stored face data corresponding to the decrypted vehicle body data or the pre-stored face data corresponding to the license plate data. If they match, proceed to S500.

[0042] S420. Determine whether the face data corresponding to the decrypted vehicle body data matches the face image. If they match, proceed to S500.

[0043] S500: Raise the barrier when the vehicle enters the park; raise the barrier after obtaining the payment information corresponding to the payment QR code when the vehicle leaves the park.

[0044] A lever raising system, which is used to raise the lever upon receiving a lever raising command.

[0045] Thirdly, a computing device.

[0046] Including processor and memory;

[0047] The memory is used to store computer instructions;

[0048] The processor is used to execute computer instructions stored in the memory, causing the computing device to perform the steps of a campus identity authentication management method.

[0049] Fourthly, a computer program product, the computer program product including computer instructions, the computer instructions instructing a computing device to perform steps implementing a campus identity authentication management method.

[0050] The present invention provides a park identity authentication management method that differs from existing technologies in that it integrates license plate data, vehicle body data, and facial image composite authentication into the aforementioned park authentication method. First, it overcomes the inconvenience of repeated registration and updates required when park owners change vehicles or license plates. Second, to prevent the problem of vehicles outside the park using cloned license plates, it uses vehicle body posture to eliminate vehicles using cloned plates, thereby increasing park security. Third, it uses asymmetric encryption to prevent the park management system from being attacked and unable to recognize correctly.

[0051] The following description, in conjunction with the accompanying drawings, further illustrates a campus identity authentication management method of the present invention. Attached Figure Description

[0052] Figure 1 This is a flowchart of a campus identity authentication management method. Detailed Implementation

[0053] like Figure 1 As shown, a campus identity authentication management method includes...

[0054] S100: Obtain vehicle license plate data, vehicle body data, and entry time;

[0055] S200: Determine whether the license plate data matches the pre-stored registered license plate. If it matches, proceed to S300; if it does not match, proceed to S420.

[0056] S300: The vehicle body data and entry time are encrypted using the first asymmetric encryption method based on the hash value of the license plate data. The server decrypts the vehicle body data using the first asymmetric encryption method based on the hash value of the license plate data to obtain the decrypted vehicle body data. The server then determines whether the decrypted vehicle body data matches the pre-stored vehicle body data corresponding to the pre-stored license plate data. If they match, the server jumps to S500. If they do not match, the server jumps to S410.

[0057] S410: Obtain the facial image corresponding to the license plate data, and determine whether the facial image matches one of the pre-stored facial data corresponding to the decrypted vehicle body data or the pre-stored facial data corresponding to the license plate data. If they match, proceed to S500 and update the pre-stored facial data corresponding to the decrypted vehicle body data and the pre-stored facial data corresponding to the license plate data according to the facial image. If they do not match, issue an alarm signal.

[0058] S420: Obtain the facial image corresponding to the license plate data. The vehicle body data and entry time are encrypted using a first asymmetric encryption method based on the hash value of the license plate data. The server then performs a first asymmetric decryption using the hash value of the license plate data to obtain the decrypted vehicle body data. Determine whether the facial data corresponding to the decrypted vehicle body data matches the facial image. If they match, proceed to S500 and update the license plate data based on the decrypted vehicle body data and the facial image. If they do not match, issue an alarm signal.

[0059] S500: Raise the barrier when a vehicle enters the park and record the entry time; when the vehicle leaves the park, output a payment QR code based on the entry time, exit time, and pre-stored parking unit price, and raise the barrier after obtaining the payment information corresponding to the payment QR code.

[0060] This invention integrates license plate data, vehicle body data, and facial image for composite authentication through the aforementioned park authentication method. First, it overcomes the inconvenience of repeated registration and updates required when park owners change their vehicles or license plates. Second, to prevent the problem of vehicles outside the park using cloned license plates, it uses vehicle posture to eliminate vehicles using cloned license plates, thereby increasing park security. Third, it uses asymmetric encryption to prevent the park management system from being attacked and unable to recognize the system properly.

[0061] Specifically, in S200, this invention first determines whether the license plate is a pre-stored registered license plate, that is, distinguishing between temporary vehicles and long-term rental vehicles. Long-term rental vehicles can further jump to S300, while temporary vehicles directly jump to S420. In S300, we encrypt the data transmission through a first asymmetric encryption and a first asymmetric decryption method to prevent the entire system from being compromised by hackers and thus unable to operate normally, and also to prevent others from tampering with the data. Furthermore, in S300, we identify whether the vehicle body matches the vehicle body in the pre-stored license plate data, which can identify whether it is a cloned vehicle based on the vehicle body data, thereby increasing the accuracy of vehicle identification. If the vehicle body matches, we jump to S500; otherwise, we jump to S410. In S410, we add the concept of matching facial images with the aforementioned license plate and vehicle body. Using facial images as user identification tags, we match the already collected license plates and vehicle bodies, so that when a user changes vehicles or license plates, the matching vehicle body and license plate can be updated in a timely manner, so that the user can enter smoothly next time without having to collect facial images again. In S420, even temporary vehicles can only be driven by personnel within the park. Therefore, it's necessary to verify if the vehicle and facial recognition match. If they match, the system redirects to S500 and updates the license plate to indicate it's a non-temporary vehicle; otherwise, an alarm is triggered. S500 is similar to a traditional parking lot, where a QR code is assigned based on entry and exit time, and the barrier is raised after payment is received. It's worth noting that for some long-term rental or prepaid vehicles, even if the user doesn't scan the QR code, the system can still raise the barrier directly based on the prepaid amount or long-term rental verification.

[0062] In simple terms, the control method of this invention is as follows:

[0063] If the license plate matches, S300 checks the vehicle body; if the body matches, it passes; if the body doesn't match, they check the driver's face. If the vehicle passes S300, it fails S420.

[0064] For S300, check if the vehicle body and license plate are the same. If they are the same, it is S500; otherwise, it is S410.

[0065] If the license plate does not match, S420 checks the vehicle body and the person's face; if it passes, S500 will check and an alarm will sound.

[0066] The phrase "obtaining vehicle license plate data, vehicle body data, and entry time" can be specifically explained as follows: The objectives of this invention are: 1. To automatically update the license plates and vehicles of park owners or individuals as security targets; 2. To prevent cloned vehicles from entering and posing a threat to park security to unsafe external vehicles and personnel; 3. To prevent hackers from using asymmetric encryption to transmit data. Therefore, when a user enters the park, the system needs to obtain license plate data, vehicle body data representing the vehicle model and driving habits, and entry time. Specifically, the vehicle body data can be a video recording from the last license plate recognition to the recognition of this license plate data, and then to the point where the license plate data can no longer be recognized. For example, if the previous license plate was A, and the license plate data being studied is B, then after license plate A completely disappears, the video recording begins. This video will definitely show license plate B, and the video ends when license plate B can no longer be recognized. Using this video as vehicle body data helps in studying the vehicle's color, shape, and driving habits, thereby preventing cloned license plates and other security problems. Among these, vehicle color and shape are mandatory items in the vehicle body data. However, driving behavior habits are not considered when queuing because vehicles may obstruct the view of the vehicle's movement trajectory, whether the vehicle is queuing or not.

[0067] The phrase "determining whether the license plate data matches the pre-stored registered license plates" can be understood as follows: All owners or individuals within the park are considered secure nodes; that is, the park should only allow entry to owners or individuals registered within the park, or individuals accompanied by registered owners or individuals. Therefore, as a preliminary judgment point, we should still, like a conventional parking fee system, identify whether the license plate is a registered license plate. That is, compare the license plate data identified by the camera with the pre-stored registered license plates in the database to see if the vehicle is registered. If it is a registered vehicle, proceed to S300, which only checks whether the vehicle is using a counterfeit license plate. If not, the barrier is raised; if it is, the user's identity is further verified to identify illegal use of a counterfeit license plate. If it is not a registered vehicle, proceed to S420, which checks whether the vehicle body and user identity match, ruling out the possibility that the user changed their license plate to circumvent license plate restrictions.

[0068] In other words, only registered vehicles and registered personnel are allowed to enter this park. This is a prerequisite for ensuring the safety of this park.

[0069] The phrase "S300, vehicle body data, and entry time are encrypted using a first asymmetric encryption method based on the hash value of the license plate data" can be understood as follows: vehicle body data, for example, a short video, approximately 5-60 seconds long; and entry time, for example, January 1, 2025, 12:00:00. These two data points are crucial for subsequent entry and exit from the park; without them, the gate cannot be raised. Therefore, if this critical data is tampered with by a hacker and sent to the server, the gate could be controlled by unauthorized individuals. To prevent such attacks, we can employ the following asymmetric encryption method:

[0070] S301. Use the SHA256 algorithm to obtain the first hash value of the license plate data;

[0071] S302, Take the last n bits of the first hash value 2 bit, where n 2 No more than 256;

[0072] S303, vehicle body data, and / or entry time are configured into an n*n grid, and the sequence number of each cell is configured sequentially to obtain the first n... 2 Grid;

[0073] S304. Compare the last n bits of the first hash value from back to front. 2 The first hash value is sorted by the size of the first bit and the last n bits of the hash value in descending order. 2 The first sequence is generated by bits, where, in the case of the same numerical value, the later one in the first hash value is considered larger;

[0074] S305, the first n 2 The grid numbers are adjusted according to the first sequence to obtain the second n. 2The grid is considered as completing the first asymmetric encryption.

[0075] This invention configures n... 2 The grid is divided into the first n. 2 The grid, then using the last n bits of the first hash value of the license plate data. 2 Use bits to configure this first n 2 The grid order forms a first asymmetric encryption. The second n-th result received later after the first asymmetric encryption... 2 The grid can be decrypted directly using this method to study the decrypted vehicle body data, or the amount can be calculated based on the decrypted entry time and the barrier can be raised after payment is received.

[0076] For example, if n is 2, then the video data of the vehicle body and the image data of the park entry time are divided into a four-grid format, with the first n being... 2 The grid; the original sequence numbers in the first row from left to right are 1, 2, and the original sequence numbers in the second row from left to right are 3, 4. Since the first hash value is a hexadecimal number generated by the SHA256 algorithm, its last four digits, for example, 33BA, arranged from back to front in descending order, generate the first sequence: 3421. That is, the original sequence numbers 1234 are transformed into a sorting of 3421. 2 The grid becomes the second n 2 The grid. And it can be decrypted in this way. Specifically, in hexadecimal, B is equivalent to 12, A is equivalent to 11, so B is greater than A, which is greater than 3. The second 3 is later in the sequence, so the first sequence is: 3421.

[0077] The value of n can be adjusted under different conditions; a larger n results in greater safety, while a smaller n results in greater speed. The value of n can be adjusted according to specific circumstances.

[0078] The phrase "the server obtains decrypted vehicle body data by performing a first asymmetric decryption based on the hash value of the license plate data" can be understood as follows: Typically, the vehicle body data and entry time, encrypted using the first asymmetric encryption method, are sent to the server along with the license plate data. The server then decrypts the vehicle body data and entry time using the aforementioned encryption method based on the license plate data. Specifically, this can be achieved as follows:

[0079] S311. Use the SHA256 algorithm to obtain the first hash value of the license plate data;

[0080] S312, Take the last n bits of the first hash value 2 bit, where n 2 No more than 256;

[0081] S313, vehicle body data and / or entry time, are configured into an n*n grid, and the sequence number of each cell is configured sequentially to obtain the second n. 2 Grid;

[0082] S314. Compare the last n bits of the first hash value from back to front. 2 The first hash value is sorted by the size of the first bit and the last n bits of the hash value in descending order. 2 The original sequence is generated based on the first sequence, wherein, in the case of the same numerical value, the later one in the first hash value is considered larger;

[0083] S315, the second n 2 The grid numbers are adjusted according to the original sequence to obtain the first n. 2 The grid is considered as completing the first asymmetric decryption.

[0084] The present invention uses the above method to use license plate data, which is embedded in the original parking fee system, as the password for decrypting the first asymmetric encryption, thereby preventing hackers from cracking it and ensuring relative security and convenience.

[0085] The phrase "S410, Obtain the facial image corresponding to the license plate data" can be understood as follows: the facial image obtained at this time is not pre-stored in the database. Rather, due to the aforementioned vehicle mismatch issue, it is necessary to collect the facial image of the user who gets out of the vehicle with this license plate data on-site for identity verification. Therefore, the facial image at this time is newly captured by the camera, and it is either a facial image of someone who has just gotten out of the car or a person actively photographed by the parking lot attendant.

[0086] The phrase "determining whether the facial image matches either the pre-stored facial data corresponding to the decrypted vehicle body data or the pre-stored facial data corresponding to the license plate data" can be understood as follows: The facial image corresponding to the license plate data is used as input. A match is considered a match if it matches either the pre-stored facial data corresponding to the vehicle body data or the pre-stored facial data corresponding to the license plate data in the database; otherwise, a mismatch is considered a non-match. It's important to note that the currently collected facial image represents the identity of the current person. Whether we allow the current person to enter depends on whether their facial data is already registered. However, simply matching the facial data with registered facial data may lead to errors due to facial recognition inaccuracies. Therefore, we narrow down the registered facial data to either the facial data corresponding to the vehicle body or the facial data corresponding to the license plate. In other words, this facial recognition requires not only facial matching but also matching of the vehicle body or license plate for a successful facial recognition. This allows us to lower the facial recognition threshold, thereby increasing the convenience and speed of facial recognition. For example, in environments with masks, hats, low light, or strong light, if face recognition is still performed using traditional face recognition thresholds, it may fail to identify faces. However, the above method can lower the recognition threshold, thereby increasing the success rate of face recognition.

[0087] The statement "If a match is found, proceed to S500 and update the pre-stored facial data corresponding to the pre-stored decrypted vehicle body data and the pre-stored facial data corresponding to the pre-stored license plate data based on the facial image" can be understood as follows: If the above match is found, it means that the current person is indeed an owner or person in the park, and the database of pre-stored facial data should be updated accordingly using the currently collected facial image to improve the accuracy of the next recognition.

[0088] The phrase "if there is no match, an alarm signal will be issued" can be understood as follows: this alarm signal is not only for parking lot managers, but also for users whose facial images have just been collected, so that manual registration for entry can be carried out, or the vehicle can be temporarily charged for entry, or entry can be denied.

[0089] In step S420, "determining whether the face data corresponding to the decrypted vehicle body data matches the face image" can be understood as follows: whether the face data corresponding to the vehicle body matches the face image refers to the above description of "determining whether the face image matches one of the pre-stored face data corresponding to the decrypted vehicle body data or the pre-stored face data corresponding to the license plate data".

[0090] The description of "S500 raising the barrier and recording the entry time when a vehicle enters the park; and outputting a payment QR code based on the entry time, exit time, and pre-stored parking price when the vehicle leaves the park, and raising the barrier after obtaining the payment information corresponding to the payment QR code" can be understood as follows: When entering the park, S500 can directly raise the barrier and record the entry time; when leaving the park, S500 first calculates the fee, then scans the code, and raises the barrier after successful scanning, or first calculates the fee and then determines whether payment is required. If there is an account balance, ETC, long-term rental user, or user who has purchased a parking space, the barrier can be raised directly.

[0091] If there is a mismatch, after the step of issuing an alarm signal, the process may further include: deleting the facial image, vehicle body data, and license plate data related to triggering the alarm signal from the pre-stored registered database.

[0092] See Figure 1 In some embodiments, determining whether the decrypted vehicle body data matches the pre-stored vehicle body data corresponding to the pre-stored license plate data includes:

[0093] Determine whether the color and shape of the decrypted vehicle body data match the color and shape of the pre-stored vehicle body data corresponding to the pre-stored license plate data. If they do not match, issue an alarm signal.

[0094] If a match is found, it is determined whether the time interval between the identification of the license plate data and the last identification of another license plate data is less than a first preset threshold. If it is less, the process jumps to S500. If it is not less, a trajectory point is configured for the license plate in the vehicle body data. The position where the trajectory point appears this time is configured as the first position, and the position where the trajectory point is stationary for a preset time is configured as the second position. A first vector is constructed from the first position to the second position. It is determined whether the first vector is similar to the second vector of the pre-stored vehicle body data corresponding to the pre-stored license plate data. If they are similar, the process jumps to S500. If they are not similar, an alarm signal is issued.

[0095] This invention focuses on a detailed description of the matching method between vehicle body data and pre-stored vehicle body data. Specifically, it first uses traditional facial recognition to check if the vehicle's color and shape match. If they don't match, an alarm is triggered, indicating that the vehicle may be using a cloned license plate to avoid paying parking fees. However, even with the same color and shape, it's possible that someone stole the car, or someone outside the park rented, borrowed, or stole it. Since such vehicles are registered but the drivers aren't, it still poses a security risk. Therefore, even if a match is found, we should still examine the vehicle based on driving habits. For example, when turning into the park at the same intersection, each driver's first vector will change due to different safety thresholds for the left and right sides and different stopping times. Comparing this first vector with the pre-stored second vector of the same vehicle allows for identification of whether the driver is the original owner, thus ensuring both convenient entry into the park and the safety of screening for suspicious individuals. In constructing the first vector and pre-storing the second vector, it can be understood that the data on the vehicle body is a short video clip before the license plate is scanned. The earliest time of this video must be earlier than the appearance time of the license plate, and the end time of this video must be later than the time it takes for the license plate to be recognized and the barrier to be raised. During this process, we need to exclude situations where vehicles are queuing in a straight line during traffic jams, making it impossible to obtain the driver's driving trajectory when turning, thus preventing the identification of unregistered drivers based on driving behavior, or in other words, preventing the refusal of drivers with abnormal behavior from entering the park. That is, we can compare the time between consecutively scanning two license plates with a first preset threshold. If the time is too short, it indicates a traffic jam queue. This should be considered to reduce congestion and avoid unnecessary comparisons, allowing passage directly. S500 should not only include its description but also include directly recording the time and raising the barrier immediately upon entering the park.

[0096] Among them, the above-mentioned "determining whether the decrypted vehicle body data matches the pre-stored vehicle body data corresponding to the pre-stored license plate data" is located in S300, and its specific determination conditions are consistent with or basically consistent with the "determining whether the decrypted vehicle body data matches the face image" in S420.

[0097] The phrase "determining whether the color and shape of the decrypted vehicle body data match the color and shape of the pre-stored vehicle body data corresponding to the pre-stored license plate data" can be understood as using an algorithm similar to facial recognition to identify whether the vehicle's color and shape are consistent with the pre-stored data. However, due to the more complex angles of the vehicle, the requirements for color overlap are higher, while the requirements for shape overlap are lower. The specific settings can be adjusted by those skilled in the art.

[0098] The phrase "If a match is found, determine whether the time interval between the identification of the license plate data and the previous identification of another license plate data is less than a first preset threshold" can be understood as follows: In order to avoid the situation where the video duration of the vehicle body data collected during queuing is too short, resulting in the inability to detect the behavior trajectory, if it is too short, it is determined to be less than. If it is less than, the subsequent judgment of driving habits is waived and the process directly jumps to S500; otherwise, the driving habits are further examined.

[0099] The first preset threshold can be 3 seconds to 10 minutes, preferably 1 minute. That is to say, if the previous license plate was captured 1 minute ago and then the license plate data is captured again, it will definitely be a situation where there is no queue or traffic jam.

[0100] The statement, "If it is not less than, then in the vehicle body data, configure trajectory points for the license plate, configure the position where the trajectory point appears this time as the first position, configure the position where the trajectory point is stationary for a preset time as the second position, and construct a first vector from the first position to the second position," can be understood as: generating a first vector for the more important trajectory in the vehicle body data as a subsequent judgment standard. Typically, a park has a perpendicular entrance on a straight road, and vehicles must make a large turn to enter. Each person's turn is affected by factors such as the turning radius, the timing of the turn, and vehicle performance, so the position of a license plate at the camera's starting position and the position where it is stationary waiting for the barrier to open will be different. Analyzing this as a vector is similar to collecting a new, unique biometric code for each person, like a fingerprint, which can be used to further verify the user's identity. Alternatively, even so, if the first vector and the second vector are significantly different, it can be judged as abnormal behavior and an alarm can be triggered.

[0101] The statement, "Determine whether the first vector is similar to the second vector of the pre-stored vehicle body data corresponding to the pre-stored license plate data. If similar, jump to S500; if not similar, issue an alarm signal," can be understood as follows: The database pre-stores a second vector corresponding to this license plate. This second vector can be multiple, representing the trajectory vectors of its multiple forms. The method for determining whether the first vector and the second vector match can be as follows:

[0102] Method 1:

[0103] Determine whether there exists a second vector whose length differs from that of the first vector by less than a first percentage; if not, then they are considered dissimilar.

[0104] If it exists, then determine whether the existing second vector has an angle that differs from the first vector by less than a first percentage; if it does not exist, it is determined to be dissimilar; if it exists, it is determined to be similar.

[0105] Method 2:

[0106] Determine whether there exists at least one second vector whose angle differs from that of the first vector by less than a first percentage; if not, then they are determined to be dissimilar.

[0107] If it exists, then determine whether the length of the second vector that exists is less than a first percentage difference from the length of the first vector; if it does not exist, it is determined to be dissimilar; if it exists, it is determined to be similar.

[0108] This invention, through the above method, can determine whether a second vector, whose length and angle differ from the first vector by less than a first percentage, from among many second vectors pre-stored in the database is similar, thereby determining whether to jump to S500 or directly issue an alarm signal. The above comparison method is relatively conventional and common, similar to fingerprint recognition.

[0109] The phrase "length difference less than the first percentage" means that the absolute value of the difference between the length of the first vector and the length of the second vector is less than the first percentage of the length of the first vector.

[0110] The phrase "angle difference less than the first percentage" means that the absolute value of the angle between the first vector and the horizontal and the angle between the second vector and the horizontal is less than the percentage of the angle between the first vector and the horizontal.

[0111] The first percentage can be 1% to 50%, preferably 3%.

[0112] Method 3:

[0113] Build a 3D map to adjust the matching degree with the second vector:

[0114] Using the viewfinder of the video that acquires the vehicle body data of the first vector as the first map base, the height and radius of the cone are generated by multiplying the first vector by a length weighting coefficient, and the two cones are respectively placed at the starting point and the ending point of the first vector, wherein the center of the base of the cone coincides with the starting point and the ending point of the first vector, and a first three-dimensional map of the first vector is generated on the first map base.

[0115] Using the viewfinder of the video with the pre-stored second vector vehicle body data as the second map base, the height and radius of the cone are generated by multiplying the second vector by the length weight coefficient, and the two cones are respectively placed at the starting point and the ending point of the second vector, wherein the center of the base of the cone coincides with the starting point and the ending point of the second vector, and a second three-dimensional map of the second vector is generated on the map base.

[0116] Determine whether there exists at least one second 3D map whose non-overlapping volume with the first 3D map is less than a first percentage of the total volume of the first 3D map. If such a volume exists, the map is considered similar; otherwise, it is considered dissimilar.

[0117] In order to integrate the differences between the first vector and the second vector in terms of starting point, key position, and length, this invention constructs a three-dimensional map of a cone with equal length and radius, compares the non-overlapping volumes, and then compares it with the first percentage, so as to intuitively determine whether there is a second vector similar to the first vector.

[0118] The first percentage can be 1% to 50%, preferably 3%.

[0119] See Figure 1 In some embodiments, determining whether the decrypted vehicle data matches the facial image, and if they match, includes:

[0120] Determine whether the color and shape of the decrypted vehicle body data match the color and shape of the pre-stored vehicle body data corresponding to the pre-stored license plate data. If they do not match, issue an alarm signal.

[0121] If a match is found, it is determined whether the time interval between the identification of the license plate data and the last identification of another license plate data is less than a first preset threshold (because the trajectory points collected in a short time due to traffic jams and queuing are not representative). If it is less than the threshold, the process jumps to S500. If it is not less than the threshold, a trajectory point is configured for the license plate in the vehicle body data. The position where the trajectory point appears this time is configured as the first position, and the position where the trajectory point is stationary for a preset time is configured as the second position. A first vector is constructed from the first position to the second position. It is determined whether the first vector is similar to the second vector of the pre-stored vehicle body data corresponding to the pre-stored face data that matches the face image. If they are similar, the process jumps to S500. If they are not similar, an alarm signal is issued.

[0122] The present invention constructs a second vector from pre-stored vehicle body data in a database, which is generated during the process of retrieving matching facial data from a facial image. This second vector is compared with the first vector to determine whether the barrier can be raised or an alarm can be triggered.

[0123] See Figure 1 In some embodiments, the update steps of S410 ("If a match is found, proceed to S500 and update the pre-stored face data corresponding to the pre-stored decrypted vehicle body data and the pre-stored face data corresponding to the pre-stored license plate data according to the face image") and S420 ("If a match is found, proceed to S500 and update the license plate data according to the decrypted vehicle body data and the face image") include the following steps:

[0124] S431. Determine whether the total number of updates to the license plate data, vehicle body data, and face data of the user corresponding to the face image within a first preset time exceeds a first threshold. If not, execute the update step; if so, then: the step of "determining whether the first vector is similar to the second vector of the pre-stored vehicle body data corresponding to the pre-stored license plate data; if similar, jump to S500; if not similar, issue an alarm signal." includes:

[0125] Using the viewfinder of the video that acquires the vehicle body data of the first vector as the first map base, the height and radius of the cone are generated by multiplying the first vector by a length weighting coefficient, and the two cones are respectively placed at the starting point and the ending point of the first vector, wherein the center of the base of the cone coincides with the starting point and the ending point of the first vector, and a first three-dimensional map of the first vector is generated on the first map base.

[0126] Using the viewfinder of the video with the pre-stored second vector vehicle body data as the second map base, the height and radius of the cone are generated by multiplying the second vector by the length weight coefficient, and the two cones are respectively placed at the starting point and the ending point of the second vector, wherein the center of the base of the cone coincides with the starting point and the ending point of the second vector, and a second three-dimensional map of the second vector is generated on the map base.

[0127] Determine whether there exists a second percentage in at least one of the second three-dimensional maps that does not overlap with the first three-dimensional map and has a volume smaller than the total volume of the first three-dimensional map. The second percentage is calculated as the first percentage * the first threshold / the total number of updates. If such a percentage exists, the map is considered similar. If not, the map is considered dissimilar. Then, delete the pre-stored license plate data, vehicle body data, and face data of the user corresponding to the face image.

[0128] This invention adds a more stringent step to the original method of deleting pre-stored license plate, vehicle, and facial data based on excessive updates within a short period. This involves constructing a more rigorous method to determine the overlap ratio when comparing the overlap area between the first and second 3D map models. Specifically, instead of the original manually set threshold of a first percentage, if the total number of updates exceeds a first threshold, the threshold is lowered to a second percentage. Therefore, in this case, the user is only flagged as suspicious, without immediately deleting all pre-stored license plate, vehicle, and facial data. Instead, the more stringent vehicle data threshold is applied the next time the user re-enters the site. Based on the matching conditions, the system determines whether to raise the barrier or issue an alarm. On one hand, this method essentially increments the first threshold, giving the user an extra opportunity to update their data without significantly affecting the standard first threshold judgment logic. This protects data from easy deletion, ensuring data security and reliability. On the other hand, this entry also serves as a new entry of vehicle data, specifically driving behavior habits. Even with stricter judgment conditions in the future, if the driving behavior remains consistent, the system will still successfully determine that the 3D map model matches and raise the barrier, thus ensuring both the safety and convenience of the judgment.

[0129] In any security authentication, complex decision criteria are more secure but inconvenient; simple decision criteria are convenient but insecure. This invention utilizes the aforementioned cyclical method to increase both the security and convenience of the invention.

[0130] The length weighting coefficient can be 0.1 to 10, preferably 0.5. That is, in each first three-dimensional map and second three-dimensional map, the length and radius of two cones are constructed with half the vector length. Constructing the weighting coefficient in this way can build a three-dimensional map model with undulations that is not particularly large, so as to compare the overlapping volumes of the two three-dimensional map models and find a more matching driving behavior habit.

[0131] Of course, as a variation of the above embodiment, the first percentage can be changed to the second percentage instead of changing the length weight coefficient to the first threshold and the total number of updates. This allows for comparison with a 3D map model that has a smaller total volume, and increases the percentage of the difference, which is not conducive to the first percentage being smaller.

[0132] See Figure 1In some embodiments, the update steps of S410 ("If a match is found, proceed to S500 and update the pre-stored face data corresponding to the pre-stored decrypted vehicle body data and the pre-stored face data corresponding to the pre-stored license plate data according to the face image") and S420 ("If a match is found, proceed to S500 and update the license plate data according to the decrypted vehicle body data and the face image") include the following steps:

[0133] S431. Determine whether the total number of updates to the license plate data, vehicle body data, and facial data of the user corresponding to the facial image within a first preset time exceeds a first threshold. If not, execute the update step; if so, then: in the step "S300, the vehicle body data and park entry time are encrypted in a first asymmetric form based on the hash value of the license plate data", the following steps are included:

[0134] S301. Use the SHA256 algorithm to obtain the first hash value of the license plate data;

[0135] S302, Take the last n bits of the first hash value 2 The number of bits, where n = n + 1, and the n 2 No more than 256;

[0136] S303, vehicle body data, and / or entry time are configured into an n*n grid, and the sequence number of each cell is configured sequentially to obtain the first n... 2 Grid;

[0137] S304. Compare the last n bits of the first hash value from back to front. 2 The first hash value is sorted by the size of the first bit and the last n bits of the hash value in descending order. 2 The first sequence is generated by bits, where, in the case of the same numerical value, the later one in the first hash value is considered larger;

[0138] S305, the first n 2 The grid numbers are adjusted according to the first sequence to obtain the second n. 2 The grid is considered as completing the first asymmetric encryption.

[0139] This invention uses a preset fixed value for the base 'n', such as 3. Initially, data is sent sequentially using a 9x9 grid, essentially a first asymmetric encryption. This is followed by a first asymmetric decryption to reassemble the 9x9 grid. If a user updates too frequently within a short period, the 'n' for that user is permanently incremented by one, changing the grid from 9x9 to 16x16. This ensures that each user's 'n' is different, increasing the difficulty of hacking. It also provides a more secure and robust data transmission and encryption method for users who have exhibited abnormal behavior, guaranteeing data security. Furthermore, it offers convenience to other users who have not experienced abnormal behavior, reducing the system's computational load.

[0140] See Figure 1 In some embodiments, the updates in S410 ("if a match is found, proceed to S500 and update the pre-stored face data corresponding to the pre-stored decrypted vehicle body data and the pre-stored face data corresponding to the pre-stored license plate data according to the face image") and S420 ("if a match is found, proceed to S500 and update the license plate data according to the decrypted vehicle body data and the face image") include the following steps:

[0141] S431. Determine whether the total number of updates of the license plate data, vehicle body data, and face data of the user corresponding to the face image within a first preset time exceeds a first threshold. If so, delete the pre-stored license plate data, vehicle body data, and face data of the user corresponding to the face image.

[0142] This invention updates the database only when a mismatch is followed by a match. Furthermore, if a user updates their data too frequently, it may be due to counterfeiting, failure to register relevant procedures, or other potential security violations causing excessive data updates in a short period. In such cases, all pre-stored data should be deleted to ensure the security of the park.

[0143] The first preset time period can be from 1 day to 1 year, preferably 1 month. The first threshold can be from 1 time to 100 times, preferably 5 times. That is to say, if a user's license plate data, vehicle body data, and facial data are updated more than 5 times within 1 month, all of their pre-stored data should be deleted to ensure the security of the park.

[0144] See Figure 1 In some embodiments, the park is an industrial park or an office park.

[0145] like Figure 1 As shown, the present invention provides a campus identity authentication management system, including...

[0146] The camera system is used to acquire vehicle license plate data, vehicle body data, and entry time.

[0147] A processor, which performs the following steps:

[0148] S200: Determine whether the license plate data matches the pre-stored registered license plate. If it matches, proceed to S300; if it does not match, proceed to S420.

[0149] S300: Determine whether the decrypted vehicle body data matches the pre-stored vehicle body data corresponding to the pre-stored license plate data. If they match, proceed to S500; otherwise, proceed to S410.

[0150] S410. Determine whether the face image matches one of the pre-stored face data corresponding to the decrypted vehicle body data or the pre-stored face data corresponding to the license plate data. If they match, proceed to S500.

[0151] S420. Determine whether the face data corresponding to the decrypted vehicle body data matches the face image. If they match, proceed to S500.

[0152] S500: Raise the barrier when the vehicle enters the park; raise the barrier after obtaining the payment information corresponding to the payment QR code when the vehicle leaves the park.

[0153] A lever raising system, which is used to raise the lever upon receiving a lever raising command.

[0154] like Figure 1 As shown, the present invention provides a computing device.

[0155] Including processor and memory;

[0156] The memory is used to store computer instructions;

[0157] The processor is used to execute computer instructions stored in the memory, causing the computing device to perform the steps of a campus identity authentication management method.

[0158] The present invention provides a computer program product, the computer program product including computer instructions, the computer instructions instructing a computing device to perform the steps of a campus identity authentication management method.

[0159] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for managing identity authentication in a park, characterized in that: include S100: Obtain vehicle license plate data, vehicle body data, and entry time; S200: Determine whether the license plate data matches the pre-stored registered license plate. If it matches, proceed to S300; if it does not match, proceed to S420. S300: The vehicle body data and entry time are encrypted using the first asymmetric encryption method based on the hash value of the license plate data. The server decrypts the vehicle body data using the first asymmetric encryption method based on the hash value of the license plate data to obtain the decrypted vehicle body data. The server then determines whether the decrypted vehicle body data matches the pre-stored vehicle body data corresponding to the pre-stored license plate data. If they match, the server jumps to S500. If they do not match, the server jumps to S410. S410. Determine whether the face image matches one of the pre-stored face data corresponding to the decrypted vehicle body data or the pre-stored face data corresponding to the license plate data. If they match, jump to S500 and update the pre-stored face data corresponding to the decrypted vehicle body data and the pre-stored face data corresponding to the license plate data according to the face image. S420: Determine whether the face data corresponding to the decrypted vehicle body data matches the face image. If they match, proceed to S500 and update the license plate data based on the decrypted vehicle body data and the face image. S500: The barrier is raised when a vehicle enters the park; the barrier is raised again after obtaining payment information corresponding to the payment QR code when the vehicle leaves the park. The determination of whether the decrypted vehicle body data matches the pre-stored vehicle body data corresponding to the pre-stored license plate data includes: Determine whether the color and shape of the decrypted vehicle body data match the color and shape of the pre-stored vehicle body data corresponding to the pre-stored license plate data; If a match is found, it is determined whether the time interval between the identification of the license plate data and the last identification of another license plate data is less than a first preset threshold. If it is less, the process jumps to S500. If it is not less, a trajectory point is configured for the license plate in the vehicle body data. The position where the trajectory point appears this time is configured as the first position, and the position where the trajectory point is stationary for a preset time is configured as the second position. A first vector is constructed from the first position to the second position. It is determined whether the first vector is similar to the second vector of the pre-stored vehicle body data corresponding to the pre-stored license plate data. If they are similar, the process jumps to S500. If they are not similar, an alarm signal is issued. The step of determining whether the decrypted vehicle data matches the facial image, if they match, includes: Determine whether the color and shape of the decrypted vehicle body data match the color and shape of the pre-stored vehicle body data corresponding to the pre-stored license plate data. If they do not match, issue an alarm signal. If a match is found, it is determined whether the time interval between the identification of the license plate data and the last identification of another license plate data is less than a first preset threshold. If it is less, the process jumps to S500. If it is not less, a trajectory point is configured for the license plate in the vehicle body data. The position where the trajectory point appears this time is configured as the first position, and the position where the trajectory point is stationary for a preset time is configured as the second position. A first vector is constructed from the first position to the second position. It is determined whether the first vector is similar to the second vector of the pre-stored vehicle body data corresponding to the pre-stored face data that matches the face image. If they are similar, the process jumps to S500. If they are not similar, an alarm signal is issued.

2. The campus identity authentication management method according to claim 1, characterized in that: The update steps of S410 ("If a match is found, proceed to S500 and update the pre-stored face data corresponding to the decrypted vehicle body data and the pre-stored face data corresponding to the license plate data according to the face image") and S420 ("If a match is found, proceed to S500 and update the license plate data according to the decrypted vehicle body data and the face image") include the following steps: S431. Determine whether the total number of updates of the license plate data, vehicle body data, and face data of the user corresponding to the face image within the first preset time exceeds the first threshold. If not, execute the update step. If so, then: the step of "determining whether the first vector is similar to the second vector of the pre-stored vehicle body data corresponding to the pre-stored license plate data; if similar, jumping to S500; if not similar, issuing an alarm signal" includes: Using the viewfinder of the video that acquires the vehicle body data of the first vector as the first map base, the height and radius of the cone are generated by multiplying the first vector by a length weighting coefficient, and the two cones are respectively placed at the starting point and the ending point of the first vector, wherein the center of the base of the cone coincides with the starting point and the ending point of the first vector, and a first three-dimensional map of the first vector is generated on the first map base. Using the viewfinder of the video with the pre-stored second vector vehicle body data as the second map base, the height and radius of the cone are generated by multiplying the second vector by the length weight coefficient, and the two cones are respectively placed at the starting point and the ending point of the second vector, wherein the center of the base of the cone coincides with the starting point and the ending point of the second vector, and a second three-dimensional map of the second vector is generated on the map base. Determine whether there exists a second percentage in at least one of the second three-dimensional maps that does not overlap with the first three-dimensional map and has a volume smaller than the total volume of the first three-dimensional map. The second percentage is calculated as the first percentage * the first threshold / the total number of updates. If such a percentage exists, the map is considered similar. If not, the map is considered dissimilar. Then, delete the pre-stored license plate data, vehicle body data, and face data of the user corresponding to the face image.

3. The campus identity authentication management method according to claim 1, characterized in that: The update steps in S410 ("If a match is found, proceed to S500 and update the pre-stored face data corresponding to the pre-stored decrypted vehicle body data and the pre-stored face data corresponding to the pre-stored license plate data according to the face image") and S420 ("If a match is found, proceed to S500 and update the license plate data according to the decrypted vehicle body data and the face image") include the following steps: S431. Determine whether the total number of updates of the license plate data, vehicle body data, and face data of the user corresponding to the face image within a first preset time exceeds a first threshold. If so, delete the pre-stored license plate data, vehicle body data, and face data of the user corresponding to the face image.

4. A campus identity authentication management system, characterized in that: include The camera system is used to acquire vehicle license plate data, vehicle body data, and entry time. A processor, which performs the following steps: S200: Determine whether the license plate data matches the pre-stored registered license plate. If it matches, proceed to S300; if it does not match, proceed to S420. S300: The vehicle body data and entry time are encrypted using the first asymmetric encryption method based on the hash value of the license plate data. The server decrypts the vehicle body data using the first asymmetric encryption method based on the hash value of the license plate data to obtain the decrypted vehicle body data. The server then determines whether the decrypted vehicle body data matches the pre-stored vehicle body data corresponding to the pre-stored license plate data. If they match, the server jumps to S500. If they do not match, the server jumps to S410. S410. Determine whether the face image matches one of the pre-stored face data corresponding to the decrypted vehicle body data or the pre-stored face data corresponding to the license plate data. If they match, jump to S500 and update the pre-stored face data corresponding to the decrypted vehicle body data and the pre-stored face data corresponding to the license plate data according to the face image. S420: Determine whether the face data corresponding to the decrypted vehicle body data matches the face image. If they match, proceed to S500 and update the license plate data based on the decrypted vehicle body data and the face image. S500: The barrier is raised when a vehicle enters the park; the barrier is raised again after obtaining payment information corresponding to the payment QR code when the vehicle leaves the park. A boom control system, used to raise the boom upon receiving a boom command; The determination of whether the decrypted vehicle body data matches the pre-stored vehicle body data corresponding to the pre-stored license plate data includes: Determine whether the color and shape of the decrypted vehicle body data match the color and shape of the pre-stored vehicle body data corresponding to the pre-stored license plate data; If a match is found, it is determined whether the time interval between the identification of the license plate data and the last identification of another license plate data is less than a first preset threshold. If it is less, the process jumps to S500. If it is not less, a trajectory point is configured for the license plate in the vehicle body data. The position where the trajectory point appears this time is configured as the first position, and the position where the trajectory point is stationary for a preset time is configured as the second position. A first vector is constructed from the first position to the second position. It is determined whether the first vector is similar to the second vector of the pre-stored vehicle body data corresponding to the pre-stored license plate data. If they are similar, the process jumps to S500. If they are not similar, an alarm signal is issued.

5. A computing device, characterized in that, Including processor and memory; The memory is used to store computer instructions; The processor is configured to execute computer instructions stored in the memory, causing the computing device to perform the steps of the campus identity authentication management method according to any one of claims 1 to 3.

6. A computer program product, the computer program product comprising computer instructions, the computer instructions instructing a computing device to perform the steps of a campus identity authentication management method according to any one of claims 1 to 3.

Images