AI-based parking order intelligent auditing method and system, and storage medium

Abstract

The invention discloses an AI-based parking order intelligent auditing method and system and a storage medium, and belongs to the technical field of data processing, and the method comprises the following steps: after an AI intelligent parking auditing platform receives a vehicle pre-leaving signal, a parking order is automatically generated, and an order verification signal is synchronously triggered and transmitted to an edge master control processor; after receiving the order verification signal, the edge master control processor acquires an entry acquisition image of the vehicle to be driven away, so that parking space information and vehicle appearance parameters of the vehicle to be driven away are obtained through matching, and the parking space is marked as an order parking space; obtaining an order parking space geomagnetic perception influence factor, obtaining an order parking space geomagnetic perception index, and carrying out the judgment to obtain an order parking space recognition result of a vehicle to be driven away; whether manual reexamination needs to be carried out is judged according to the order parking space identification result of the vehicle to be driven away, and the problems of low parking order reexamination identification precision and high misjudgment rate caused by single video detection in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an AI-based parking order intelligent review method, system and storage medium. Background Art

[0002] With the rapid development of the automobile manufacturing industry and the improvement of people's living standards, the utilization rate of cars has greatly increased, resulting in an increasing number of parking order reviews and the level of review requirements. The existing parking order review system generally obtains parking time, owner account information, vehicle history records and other data through cameras and databases to achieve review management.

[0003] For example, the parking fee collection method, device and electronic device disclosed in the invention patent publication with publication number CN111784420B include: obtaining vehicle information of a parked vehicle at a parking space, where the vehicle information has a corresponding user account on a payment platform; publishing a vehicle parking record on the blockchain through a first blockchain node corresponding to the parking fee collection system on the blockchain, where the vehicle parking record includes the parking duration of the parked vehicle at the parking space; and whenever the parking duration reaches a first preset duration, performing an asset transfer operation related to the payment platform based on the vehicle parking record to collect assets corresponding to the first preset duration.

[0004] For example, the invention patent announcement with announcement number: CN113191514B announces a visitor vehicle information reservation and review method and system, including: obtaining first associated user information according to first reservation request information; obtaining first user owner information according to the first associated user information; judging whether the first associated user information and the first user owner information meet the first predetermined condition; obtaining first reservation vehicle information when satisfied; obtaining first vehicle historical path information according to the first reservation vehicle information; obtaining preset time information; obtaining first vehicle preset record; inputting the first vehicle preset record into the first training model to obtain record matching information; obtaining the first vehicle evaluation grade according to the record matching information; when the first vehicle evaluation grade meets the second predetermined condition, obtaining the first vehicle reservation review result according to the first reservation time information.

[0005] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0006] In the prior art, existing parking systems usually only rely on cameras to obtain vehicle information. In actual applications, they are affected by various factors such as light changes, weather conditions, traffic density, and installation location, resulting in poor image quality captured by the camera or inability to accurately identify the vehicle. Therefore, there are problems such as low parking order review and recognition accuracy and high misjudgment rate. Summary of the invention

[0007] The embodiments of the present application solve the problems of low parking order review and recognition accuracy and high misjudgment rate caused by single video detection in the prior art by providing an AI-based intelligent review method, system and storage medium for parking orders, thereby achieving accurate identification of vehicles in complex environments.

[0008] An embodiment of the present application provides an AI-based intelligent review method for parking orders, comprising the following steps: when the AI ​​intelligent parking review platform receives a vehicle pre-departure signal, it automatically generates a parking order and synchronously triggers an order verification signal to be transmitted to the edge master control processor; after the edge master control processor receives the order verification signal, it obtains an entry acquisition image of the pre-departure vehicle, thereby matching the parking space information and vehicle appearance parameters of the pre-departure vehicle, and marking the parking space as an order parking space; obtaining the historical geomagnetic perception change parameters of the order parking space, and analyzing the geomagnetic perception influencing factor of the order parking space, obtaining the geomagnetic perception parameters of the order parking space, and analyzing them with the geomagnetic perception influencing factor of the order parking space to obtain the geomagnetic perception index of the order parking space, thereby determining the order parking space recognition result of the pre-departure vehicle; and determining whether manual review is required based on the order parking space recognition result of the pre-departure vehicle.

[0009] Furthermore, the acquiring of the approaching captured image of the pre-departing vehicle, thereby matching the parking space information and vehicle appearance parameters of the pre-departing vehicle, and marking the parking space as an ordered parking space, specifically comprises the following steps: acquiring the approaching captured image of the pre-departing vehicle, identifying the license plate information, parking space information and vehicle appearance parameters of the pre-departing vehicle, and marking the parking space as an ordered parking space; the vehicle appearance parameters comprise: the maximum vehicle length, maximum vehicle width, maximum vehicle height, wheel size and average grayscale of the vehicle color of the pre-departing vehicle; performing analysis based on the vehicle appearance parameters to obtain vehicle appearance indicators of the pre-departing vehicle; the vehicle appearance indicators of the pre-departing vehicle are used to characterize the appearance recognition of the pre-departing vehicle.

[0010] Furthermore, the acquisition of historical geomagnetic perception change parameters of the order parking space and the analysis to obtain the geomagnetic perception influencing factor of the order parking space specifically comprises the following steps: the historical geomagnetic perception change parameters of the order parking space include the daily average geomagnetic induction delay time and the daily average magnetic field strength change, the daily average magnetic induction strength change, the daily average magnetic flux rate change and the daily average magnetic permeability change; the geomagnetic induction reference set preset in the database is acquired, the geomagnetic induction reference set includes the daily average geomagnetic sensor delay reference value and the daily average magnetic field strength change reference value, the daily average magnetic induction strength change reference value, the daily average magnetic flux rate change reference value and the daily average magnetic permeability change reference value; the geomagnetic induction influencing index of the order parking space is obtained by comparing the historical geomagnetic perception change parameters of the order parking space with the geomagnetic induction reference set; the geomagnetic perception influencing factor of the order parking space is obtained by analysis based on the geomagnetic induction influencing index of the order parking space; the geomagnetic induction influencing index of the order parking space is used to characterize the historical interference intensity of the order parking space.

[0011] Furthermore, the analysis based on the geomagnetic induction influence index of the order parking space is performed to obtain the geomagnetic perception influence factor of the order parking space. The specific steps include: obtaining the geomagnetic induction influence index intervals of each location preset in the database and the geomagnetic induction influence factors corresponding to the geomagnetic induction influence index intervals of each location, matching the geomagnetic induction influence index of the order parking space with the geomagnetic induction influence index intervals of each location preset in the database, if the geomagnetic induction influence index of the order parking space is within a preset geomagnetic induction influence index interval, then obtaining the geomagnetic induction influence factor corresponding to the geomagnetic induction influence index interval as the geomagnetic induction influence factor of the order parking space.

[0012] Furthermore, the order parking space geomagnetic perception index is obtained, and the order parking space identification result of the pre-departing vehicle is determined therefrom, specifically including: obtaining the maximum lateral dimension and the maximum longitudinal dimension of the vehicle based on the vehicle shape parameters; obtaining the geomagnetic perception parameters of the order parking space and analyzing them with the maximum lateral dimension, the maximum longitudinal dimension and the geomagnetic induction influencing factor of the order parking space to obtain the geomagnetic perception index of the order parking space; the geomagnetic perception index of the order parking space is used to characterize the degree of geomagnetic influence of the pre-departing vehicle on the order parking space; the geomagnetic perception parameters of the order parking space include the lateral coverage dimension of the vehicle, the longitudinal coverage dimension of the vehicle, and the magnetic induction intensity. and magnetic field strength; obtain the first geomagnetic perception evaluation threshold and the second geomagnetic perception evaluation threshold preset in the database, and compare them with the geomagnetic perception index of the order parking space to obtain the order parking space recognition result of the pre-leaving vehicle; if the geomagnetic perception index of the order parking space is less than the first geomagnetic perception evaluation threshold or the geomagnetic perception index of the order parking space is greater than the second geomagnetic perception evaluation threshold, then the order parking space recognition result of the pre-leaving vehicle is that the pre-leaving vehicle leaves the order parking space; if the geomagnetic perception index of the order parking space is above the first geomagnetic perception evaluation threshold and below the second geomagnetic perception evaluation threshold, then the order parking space recognition result of the pre-leaving vehicle is that the pre-leaving vehicle does not leave the order parking space.

[0013] Furthermore, the specific method for obtaining the geomagnetic perception index of the ordered parking space is as follows:

[0014]

[0015] In the formula, CW represents the geomagnetic perception index of the parking space, e represents the natural constant, and DC y represents the geomagnetic induction factor of the parking space, L x Indicates the lateral coverage size of the vehicle, ΔL x Indicates the maximum lateral dimension of the vehicle, L y Indicates the longitudinal coverage dimension of the vehicle, ΔL y Indicates the maximum longitudinal dimension of the vehicle, Q a Represents the magnetic induction intensity, ΔQ a Indicates the reference value of magnetic induction intensity, Q b represents the magnetic field strength, ΔQ b Indicates reference value.

[0016] Furthermore, the determining whether manual review is needed based on the identification result of the ordered parking space of the pre-departing vehicle comprises the following specific steps: if the identification result of the ordered parking space of the pre-departing vehicle is that the pre-departing vehicle has left the ordered parking space, performing an appearance inspection on the pre-departing vehicle when it leaves the parking lot, obtaining vehicle appearance review indicators and license plate review information of the pre-departing vehicle, and comparing them with the vehicle appearance indicators and license plate information of the pre-departing vehicle respectively, to determine again whether manual review is needed; if the identification result of the ordered parking space of the pre-departing vehicle is that the pre-departing vehicle has not left the ordered parking space, performing a manual review on the pre-departing vehicle.

[0017] Further, the comparison is made with the vehicle appearance index and license plate information of the vehicle to be driven away respectively to determine whether manual review is required, specifically including: if the vehicle appearance review index of the vehicle to be driven away is consistent with the vehicle appearance index of the vehicle to be driven away and the license plate information of the vehicle to be driven away is consistent with the license plate review information of the vehicle to be driven away, then the intelligent review is passed and no manual review is performed; if the vehicle appearance review index of the vehicle to be driven away is inconsistent with the vehicle appearance index of the vehicle to be driven away or the license plate information of the vehicle to be driven away is inconsistent with the license plate review information of the vehicle to be driven away, then an intelligent review is performed, and the specific process is: obtaining the geomagnetic perception index of the order parking space and the environmental impact parameters within a preset period of time after the order upload time point, and evaluating to obtain the geomagnetic perception secondary evaluation index; A secondary evaluation first threshold and a secondary evaluation second threshold preset in a database are obtained, and compared with the geomagnetic perception secondary evaluation index to obtain a secondary identification result of the order parking space of the pre-departing vehicle; if the geomagnetic perception secondary evaluation index is less than the secondary evaluation first threshold or the geomagnetic perception secondary evaluation index is greater than the secondary evaluation second threshold, then the secondary identification result of the order parking space of the pre-departing vehicle is that the pre-departing vehicle leaves the order parking space; if the geomagnetic perception secondary evaluation index is above the secondary evaluation first threshold and below the secondary evaluation second threshold, then the secondary identification result of the order parking space of the pre-departing vehicle is that the pre-departing vehicle does not leave the order parking space; the environmental impact parameters include the average ambient air pressure, average ambient humidity and average ambient temperature in a preset period of time past the order upload time point.

[0018] Furthermore, the AI-based parking order intelligent review system is characterized by comprising: a signal transmission module, a pre-departure vehicle appearance acquisition module, a geomagnetic induction evaluation module and a manual review and judgment module; the signal transmission module is used to automatically generate a parking order when the AI ​​intelligent parking review platform receives the vehicle pre-departure signal, and synchronously trigger the order verification signal to be transmitted to the edge master control processor; the pre-departure vehicle appearance acquisition module is used to obtain the entry acquisition image of the pre-departure vehicle after the edge master control processor receives the order verification signal, thereby matching the parking space information and vehicle appearance parameters of the pre-departure vehicle, and marking the parking space as the order parking space; the geomagnetic induction evaluation module is used to obtain the historical geomagnetic perception change parameters of the order parking space, and analyze the geomagnetic perception influencing factor of the order parking space, obtain the geomagnetic perception parameters of the order parking space, and analyze them with the geomagnetic perception influencing factor of the order parking space to obtain the geomagnetic perception index of the order parking space, thereby judging the order parking space identification result of the pre-departure vehicle; the manual review and judgment module is used to judge whether manual review is required according to the order parking space identification result of the pre-departure vehicle.

[0019] An embodiment of the present application provides a computer-readable storage medium for storing a program, which, when executed by a processor, implements an AI-based intelligent review method for parking orders.

[0020] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0021] 1. The AI-based intelligent parking order review method provided by the present invention evaluates the vehicle appearance indicators of the pre-departing vehicle and the geomagnetic induction influence indicators of the order parking space, thereby determining the order parking space identification result of the pre-departing vehicle, thereby realizing accurate identification of vehicles in complex environments, and effectively solving the problem of low parking order review and identification accuracy and high misjudgment rate caused by only single video detection in the prior art.

[0022] 2. The present invention obtains the geomagnetic perception parameters of the order parking space and analyzes them with the geomagnetic induction influencing factors of the order parking space, thereby obtaining the geomagnetic perception index of the order parking space, avoiding the potential interference of external factors such as the surrounding electromagnetic field on the geomagnetic measurement, and thus achieving the improvement of the parking space recognition accuracy for the pre-departing vehicle under different environmental conditions.

[0023] 3. By comprehensively evaluating the geomagnetic perception index of the order parking space and the environmental impact parameters in the preset period after the order upload time, the geomagnetic perception secondary evaluation index is obtained, so as to accurately judge the result of the pre-departure vehicle leaving the order parking space, thereby improving the efficiency of parking order review in a complex and changeable environment and reducing the occurrence of misjudgment and missed judgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A flow chart of the method provided in the embodiment of the present application;

[0025] Figure 2 A schematic diagram of the system structure provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] The embodiments of the present application solve the problems of low parking order review and recognition accuracy and high misjudgment rate caused by single video detection in the prior art by providing an AI-based parking order intelligent review method, system and storage medium, thereby achieving accurate identification of vehicles in complex environments.

[0027] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0028] like Figure 1As shown, it is a flow chart of the AI-based intelligent review method for parking orders provided in an embodiment of the present application, and the method includes the following steps: when the AI ​​intelligent parking review platform receives a vehicle pre-departure signal, it automatically generates a parking order and synchronously triggers an order verification signal to be transmitted to the edge master control processor; after the edge master control processor receives the order verification signal, it obtains the entry acquisition image of the pre-departure vehicle, thereby matching the parking space information and vehicle appearance parameters of the pre-departure vehicle, and marking the parking space as an order parking space; obtaining the historical geomagnetic perception change parameters of the order parking space, and analyzing to obtain the geomagnetic perception influencing factor of the order parking space, obtaining the geomagnetic perception parameters of the order parking space, and analyzing them with the geomagnetic perception influencing factor of the order parking space to obtain the geomagnetic perception index of the order parking space, thereby determining the order parking space recognition result of the pre-departure vehicle; judging whether manual review is required based on the order parking space recognition result of the pre-departure vehicle.

[0029] In this embodiment, it should be noted that the edge master control processor, usually referred to as the edge controller (EPI C), is the physical interface between IT (information technology) and OT (operational technology). The edge controller has a built-in PLC (programmable logic controller) or PAC (programmable automation controller) and has advanced programming, communication, visualization and other functions. It integrates PLC, PAC, PC, advanced programming, communication, visualization and other technologies, and has data processing and logic control functions.

[0030] It should also be noted that the present invention automatically generates parking orders through the AI ​​intelligent parking review platform, and synchronously transmits the order verification signal to the edge master control processor, and automatically completes the order verification and vehicle departure judgment by combining the vehicle entry collection image, the order parking space geomagnetic perception parameters and other information, thereby improving the efficiency of parking order review. The geomagnetic perception influencing factor is obtained by analyzing the historical geomagnetic perception change parameters of the order parking space, and the geomagnetic perception index of the order parking space is obtained by analyzing the real-time geomagnetic perception parameters, so as to accurately judge whether the vehicle has left the order parking space, avoid the review misjudgment caused by geomagnetic interference, and automatically trigger the review process according to the judgment result of the vehicle leaving the order parking space. Only when necessary, the abnormal order is submitted for manual review, which greatly reduces the workload of manual review, and at the same time improves the independence and automation level of the intelligent review platform.

[0031] Furthermore, an approaching captured image of the pre-departing vehicle is obtained, thereby matching the parking space information and vehicle appearance parameters of the pre-departing vehicle, and marking the parking space as an ordered parking space. The specific steps include: obtaining an approaching captured image of the pre-departing vehicle, identifying the license plate information, parking space information and vehicle appearance parameters of the pre-departing vehicle, and marking the parking space as an ordered parking space; the vehicle appearance parameters include: the maximum vehicle length, maximum vehicle width, maximum vehicle height, wheel size and average grayscale of the vehicle color of the pre-departing vehicle; analyzing based on the vehicle appearance parameters to obtain the vehicle appearance index of the pre-departing vehicle; the vehicle appearance index of the pre-departing vehicle is used to characterize the appearance recognition of the pre-departing vehicle.

[0032] In this embodiment, it should be noted that when a vehicle enters the parking lot, the AI ​​intelligent parking audit platform will start the video acquisition device (such as a camera) to record the vehicle throughout the entire process, and capture the images of the pre-departure vehicle in the entry video multiple times, and use these images as the entry acquisition images of the pre-departure vehicle. The entry acquisition image is identified using the license plate recognition (LPR) technology to obtain the license plate information of the pre-departure vehicle, and the parking space number in the image is obtained. Using image measurement software (such as Camera Measure, SigmaScan Pro, Digimizer, etc.), the length, width, height, wheel size and other external parameters of the vehicle are measured to obtain the maximum length, maximum width, maximum height and wheel size of the pre-departure vehicle, obtain the vehicle color in the picture, and grayscale the color of the vehicle to obtain the average grayscale of the vehicle color. And the above information is saved to the AI ​​intelligent parking audit platform. It should be noted that the wheel size of the pre-departure vehicle refers to the wheel size of any wheel of the pre-departure vehicle.

[0033] The vehicle shape index of the vehicle to be driven out is obtained by:

[0034] WX=e (A+B+H) +e C +e R ;

[0035] Wherein, WX represents the vehicle shape index of the pre-departure vehicle, e represents the natural constant, A represents the processed value of the maximum length of the pre-departure vehicle, B represents the processed value of the maximum width of the pre-departure vehicle, H represents the processed value of the maximum height of the pre-departure vehicle, C represents the processed value of the wheel size of the pre-departure vehicle, and R represents the processed value of the average grayscale of the vehicle color of the pre-departure vehicle.

[0036] It should be noted that the processed value of the maximum length of the vehicle, the processed value of the maximum width of the vehicle, the processed value of the maximum height of the vehicle, the processed value of the wheel size and the processed value of the average grayscale of the vehicle color of the pre-departing vehicle are obtained through data cleaning and data preprocessing, and the unit uniform removal processing has been performed, so mathematical operations can be performed directly.

[0037] It should also be noted that the vehicle shape index of the pre-departure vehicle is obtained by analyzing the vehicle shape parameters (the maximum length, maximum width, maximum height, wheel size and average grayscale of the vehicle color of the pre-departure vehicle), taking into account the mutual influence relationship between these parameters, for example: length and width determine the floor area, affecting the parking space requirements; height is related to passability and stability; wheel size affects handling and suspension; color grayscale is related to nighttime visibility. These parameters work together to accurately identify vehicles in complex environments in vehicle shape recognition, providing a key basis for intelligent parking management, safety monitoring, etc.

[0038] Furthermore, historical geomagnetic perception change parameters of the order parking space are obtained, and the geomagnetic perception influencing factors of the order parking space are obtained by analysis. The specific steps include: the historical geomagnetic perception change parameters of the order parking space include the average daily geomagnetic induction delay time and the average daily magnetic field strength change, the average daily magnetic induction strength change, the average daily magnetic flux rate change and the average daily magnetic permeability change; the geomagnetic induction reference set preset in the database is obtained, the geomagnetic induction reference set includes the average daily geomagnetic sensor delay reference value and the average daily magnetic field strength change reference value, the average daily magnetic induction strength change reference value, the average daily magnetic flux rate change reference value and the average daily magnetic permeability change reference value; the geomagnetic induction influencing index of the order parking space is obtained by comparing the historical geomagnetic perception change parameters of the order parking space with the geomagnetic induction reference set; the geomagnetic perception influencing factor of the order parking space is obtained by analysis based on the geomagnetic induction influencing index of the order parking space; the geomagnetic induction influencing index of the order parking space is used to characterize the historical interference intensity of the order parking space.

[0039] In this embodiment, it should be noted that the daily average geomagnetic induction delay time can be obtained by connecting the signal output port of the geomagnetic sensor to the input port of the signal receiving or recording device (recording device such as a data acquisition device, an oscilloscope, etc.), setting a suitable signal type and frequency, and then observing and recording the time from the start of the change to the stabilization of the sensor output signal. The daily average magnetic field strength change can be obtained by using a magnetic field measuring instrument such as a Tesla meter or a fluxmeter to continuously monitor the magnetic field strength of the parking space area, and by comparing the magnetic field strength values ​​at different time points. The daily average magnetic induction intensity change can be measured using a magnetic induction intensity measuring instrument (such as a Gauss meter) embedded in the parking space, the daily average magnetic flux rate change can be measured using a fluxmeter or a Hall effect sensor embedded in the parking space, and the daily average magnetic permeability change can be measured using a high-precision magnetic permeability meter and a high-precision fluxgate sensor.

[0040] The geomagnetic induction impact index of the order parking space is obtained by:

[0041]

[0042] In the formula, DC represents the geomagnetic induction impact index of the order parking space, Y represents the average daily geomagnetic induction delay time, ΔY represents the average daily geomagnetic sensor delay reference value, e represents the natural constant, Q c Indicates the daily average magnetic field intensity change, ΔQ c Indicates the reference value of daily average magnetic field intensity change, Q g Indicates the daily average change in magnetic induction intensity, ΔQ g represents the reference value of the daily average magnetic induction intensity change, φ represents the daily average magnetic flux rate change, Δφ represents the reference value of the daily average magnetic flux rate change, ρ represents the daily average magnetic permeability change, Δρ represents the reference value of the daily average magnetic permeability change.

[0043] It should be noted that the geomagnetic induction influence index of the order parking space is obtained by obtaining the historical geomagnetic perception change parameters of the order parking space and analyzing them, taking into account the mutual influence relationship between these parameters. For example, the longer the average daily geomagnetic induction delay time indicates that the response delay of the magnetic field strength change signal is greater, resulting in the delayed recording of the actual perceived magnetic field strength change. That is, when the sensor responds slowly, the measured value of the actual change amplitude becomes smaller. Therefore, the increase in the delay time usually leads to insensitive response to the change in magnetic field strength, which affects the accuracy of the perception result. If the change in magnetic induction intensity sensed within a certain period of time is large, it means that the magnetic field change near the parking space is more significant, which is due to the proximity of the vehicle or the presence of other large-scale metal objects. Therefore, by analyzing the increase and decrease trend of the change in magnetic induction intensity, it can be inferred whether there is a vehicle on the parking space or has left. The higher the magnetic permeability of the material, the less difficult it is for the magnetic field to pass through. The increase in the change in magnetic permeability will lead to an increase in the change in magnetic flux rate. The longer the delay time will weaken the sensitivity of the magnetic field change signal, resulting in a slower response speed of the sensor to the passing of the vehicle. Vehicles in high permeability areas will produce more significant magnetic field changes, while in low permeability areas, the field changes may be weaker.

[0044] Furthermore, the geomagnetic perception influence factor of the order parking space is obtained by analysis based on the geomagnetic induction influence index of the order parking space. The specific steps include: obtaining the geomagnetic induction influence index intervals of each location preset in the database and the geomagnetic induction influence factors corresponding to the geomagnetic induction influence index intervals of each location, matching the geomagnetic induction influence index of the order parking space with the geomagnetic induction influence index intervals of each location preset in the database, if the geomagnetic induction influence index of the order parking space is within a preset geomagnetic induction influence index interval, then obtaining the geomagnetic induction influence factor corresponding to the geomagnetic induction influence index interval as the geomagnetic induction influence factor of the order parking space.

[0045] In this embodiment, through the preset intervals of local magnetic induction influence indexes and the corresponding geomagnetic induction influence factors, the geomagnetic induction influence index of the order parking space can be quickly matched with these preset intervals, so as to quickly determine the geomagnetic induction influence factor of the order parking space. The geomagnetic induction influence factor of the order parking space is obtained by analyzing the historical geomagnetic perception change parameters of the order parking space. Through the preset intervals of local magnetic induction influence indexes and the corresponding geomagnetic induction influence factors, the geomagnetic induction influence index of the order parking space can be quickly matched with these preset intervals, so as to quickly determine the geomagnetic induction influence factor of the order parking space. By introducing the geomagnetic induction influence factor of the order parking space, the influence of environmental interference and parameter fluctuations on the geomagnetic perception analysis results can be effectively reduced. Through the analysis of the geomagnetic induction influence factor of the order parking space, the evaluation of the actual geomagnetic conditions of the order parking space is made more scientific and comprehensive, which is conducive to the calculation of the subsequent geomagnetic perception index of the order parking space and the accurate evaluation of the vehicle order parking space identification results.

[0046] Furthermore, the geomagnetic perception index of the order parking space is obtained, and the identification result of the order parking space of the pre-departing vehicle is determined therefrom, specifically including: obtaining the maximum lateral dimension and the maximum longitudinal dimension of the vehicle based on the vehicle shape parameters; obtaining the geomagnetic perception parameters of the order parking space and analyzing them with the maximum lateral dimension, the maximum longitudinal dimension and the geomagnetic induction influencing factor of the order parking space to obtain the geomagnetic perception index of the order parking space; the geomagnetic perception index of the order parking space is used to characterize the degree of geomagnetic influence of the pre-departing vehicle on the order parking space; the geomagnetic perception parameters of the order parking space include the lateral coverage dimension of the vehicle, the longitudinal coverage dimension of the vehicle, the magnetic induction intensity and the magnetic field Strength; obtain the first threshold value and the second threshold value of geomagnetic perception evaluation preset in the database, and compare them with the geomagnetic perception index of the order parking space to obtain the order parking space recognition result of the pre-leaving vehicle. If the geomagnetic perception index of the order parking space is less than the first threshold value of geomagnetic perception evaluation or the geomagnetic perception index of the order parking space is greater than the second threshold value of geomagnetic perception evaluation, then the order parking space recognition result of the pre-leaving vehicle is that the pre-leaving vehicle leaves the order parking space; if the geomagnetic perception index of the order parking space is above the first threshold value of geomagnetic perception evaluation and below the second threshold value of geomagnetic perception evaluation, then the order parking space recognition result of the pre-leaving vehicle is that the pre-leaving vehicle has not left the order parking space.

[0047] In this embodiment, it should be noted that the order parking space recognition result of the pre-leaving vehicle is obtained by comparing the first threshold value and the second threshold value of the geomagnetic perception evaluation preset in the database with the geomagnetic perception index of the order parking space. This is because the magnetic field changes when the vehicle drives out of the parking space or the vehicle drives into the parking space. Therefore, it is possible to judge whether the pre-leaving vehicle is in the parking space. If the geomagnetic perception index of the order parking space is less than the first threshold value of the geomagnetic perception evaluation, it means that the pre-leaving vehicle leaves the parking space or other smaller vehicles enter the order parking space, or the vehicle entering has a small metal content and the distribution of metal parts is different from that of the pre-leaving vehicle. If the geomagnetic perception index of the order parking space is greater than the second threshold value of the geomagnetic perception evaluation, it means that other larger vehicles enter the order parking space or the vehicle entering has a large metal content and the distribution of metal parts is different from that of the pre-leaving vehicle. Therefore, it can be proved that the vehicle in the order parking space is not the pre-leaving vehicle, that is, the order parking space recognition result of the pre-leaving vehicle is that the pre-leaving vehicle leaves the order parking space.

[0048] In the prior art, video detection is the mainstream parking space detection method, which determines whether the vehicle has left the parking space by analyzing the vehicle's image information. However, the video detection method is limited by image quality, lighting conditions, weather conditions and viewing angles, and misjudgments or missed judgments may occur. Especially at night or in bad weather conditions, the effect of video surveillance is poor. The present invention, by introducing geomagnetic sensing technology, can still provide an effective basis for judgment when visual information is missing or the image quality is low. Geomagnetic sensing is not affected by ambient light and meteorological conditions, so it can ensure stability and accuracy in all weather and all environments.

[0049] Compared with the traditional method of the prior art that only relies on video detection to judge whether a vehicle has left a parking space, the present invention provides a more accurate vehicle status recognition capability by analyzing the geomagnetic sensing parameters of the order parking space, such as the lateral coverage size and longitudinal coverage size of the vehicle, and combining the geomagnetic induction influencing factors for in-depth analysis. Geomagnetic sensing technology can reflect the presence or absence of a vehicle and the changes in the magnetic field around the parking space in real time, while video detection is usually based only on static image information and lacks the accurate capture of dynamic changes generated during the interaction between the vehicle and the parking space.

[0050] The present invention can effectively distinguish whether the vehicle actually leaves the ordered parking space under different circumstances by evaluating the geomagnetic perception index of the ordered parking space. The introduction of geomagnetic perception indicators makes the order review not only rely on visual information, but also comprehensively consider the historical geomagnetic changes of the parking space and the geomagnetic changes caused by the current vehicle, which significantly improves the accuracy of order parking space identification. For example, when a vehicle leaves the parking space, the magnetic field distribution of the vehicle changes. This change can be accurately measured by the equipment, while video detection may lead to wrong judgments due to the limitations of angles, viewing distances or image recognition algorithms. Traditional video surveillance solutions rely more on manual intervention, especially when the recognition results are unclear, manual review is required, which increases management costs. In contrast, the present invention combines the multiple judgment mechanisms of geomagnetic perception technology to provide more accurate decision-making basis when identifying whether the vehicle has left the parking space, thereby reducing the number of manual reviews and improving the automation and intelligence level of parking order review. By combining geomagnetic perception and analysis of geomagnetic induction influencing factors, the present invention has significant advantages in improving the accuracy, intelligence level and reliability of parking order review, can overcome the shortcomings of existing video detection solutions, enhance the dynamic adaptability of review, and can provide stable and accurate order parking space identification results in various complex environments, thereby improving the efficiency and intelligence of parking management.

[0051] It should be noted that geomagnetic induction technology detects changes in the geomagnetic field on the ground and combines sensor data to infer the existence, shape, size and position of objects in the environment. When a vehicle enters a parking space, the metal parts and tires of the vehicle will affect the surrounding geomagnetic field, causing the distribution and intensity of the magnetic field to change. By analyzing these changes, the relevant dimensional parameters of the vehicle (such as lateral and longitudinal dimensions) can be inferred. For example, the lateral coverage dimension refers to the width of the parking space occupied by the vehicle. The width of the vehicle is related to the spatial range of the geomagnetic field it occupies. When the vehicle moves within the range of the geomagnetic sensor, the presence of its metal parts will affect the magnetic field distribution in the area. By comparing the magnetic field data before and after the vehicle enters the field, the change in magnetic field intensity is used to infer the width of the vehicle (the lateral coverage dimension of the vehicle). The longitudinal coverage dimension refers to the length of the parking space occupied by the front and rear of the vehicle, which can be obtained by capturing the impact of the vehicle on the magnetic field through the longitudinally arranged geomagnetic sensors.

[0052] It should be noted that the specific steps for obtaining the geomagnetic perception index of the order parking space are as follows: obtaining the vehicle shape index interval of each pre-departing vehicle and the magnetic induction intensity and magnetic field intensity corresponding to the vehicle shape index of each pre-departing vehicle preset in the database, comparing the vehicle shape index of the pre-departing vehicle with the preset vehicle shape index interval of each pre-departing vehicle, if the vehicle shape index of the pre-departing vehicle is within a preset vehicle shape index interval of the pre-departing vehicle, then obtaining the magnetic induction intensity and magnetic field intensity corresponding to the vehicle shape index interval of the pre-departing vehicle, and marking them as the magnetic induction intensity reference value and the magnetic field intensity reference value respectively; obtaining the geomagnetic perception parameters of the order parking space and analyzing them with the maximum lateral dimension of the vehicle, the maximum longitudinal dimension of the vehicle, the magnetic induction intensity reference value and the magnetic field intensity reference value, and the geomagnetic induction influencing factor of the order parking space to obtain the geomagnetic perception index of the order parking space.

[0053] Furthermore, the geomagnetic perception index of the order parking space is obtained. The specific method is as follows:

[0054]

[0055] In the formula, CW represents the geomagnetic perception index of the parking space, e represents the natural constant, and DC y represents the geomagnetic induction factor of the parking space, L x Indicates the lateral coverage size of the vehicle, ΔL x Indicates the maximum lateral dimension of the vehicle, L y Indicates the longitudinal coverage dimension of the vehicle, ΔL y Indicates the maximum longitudinal dimension of the vehicle, Q a Represents the magnetic induction intensity, ΔQ a Indicates the reference value of magnetic induction intensity, Q b represents the magnetic field strength, ΔQ b Indicates the reference value of magnetic field strength.

[0056] In this embodiment, it should be noted that the steps of obtaining the transverse coverage size and the longitudinal coverage size of the vehicle are as follows: the magnetic field changes caused by the parking of the vehicle are measured and captured by the geomagnetic induction measuring device (such as a Gauss meter high-precision magnetic permeability measuring instrument, a high-precision fluxgate sensor) of the parking lot to obtain magnetic field data, and the magnetic field simulation software (such as ANSYS, Comsol Multiphysics, etc.) is used to perform a baseline simulation of the magnetic field in the parking area without a vehicle, and then the magnetic field data is input into the simulation software to update the simulation, and the real-time magnetic field data is recorded. The magnetic field simulation software will automatically output the characteristics of the simulated magnetic field changes and the distribution of the measured length according to the input magnetic field data, the characteristics of the magnetic field changes, such as the change of the magnetic field intensity gradient, the expansion of the magnetic field distribution area, etc., according to the change of the magnetic field intensity gradient, the expansion size of the magnetic field distribution area, the vehicle's outline size is automatically evaluated by the simulation software to obtain the vehicle's transverse coverage size and the vehicle's longitudinal coverage size. The magnetic induction intensity and the magnetic field intensity can be measured by the geomagnetic induction measuring device (such as a Gauss meter high-precision magnetic permeability measuring instrument, a high-precision fluxgate sensor) of the parking lot.

[0057] It should also be noted that the maximum lateral dimension of the vehicle refers to the maximum lateral length that can be covered by the vehicle in the approach image of the pre-departing vehicle, that is, the maximum lateral projection length, and the maximum longitudinal dimension of the vehicle refers to the maximum longitudinal length that can be covered by the vehicle in the approach image of the pre-departing vehicle, that is, the maximum longitudinal projection length.

[0058] The geomagnetic perception index of the order parking space is obtained by analyzing the geomagnetic perception parameters of the order parking space (the vehicle's lateral coverage size and the vehicle's longitudinal coverage size), taking into account the mutual influence relationship between these parameters. For example, the larger the lateral coverage size, the larger the interaction area between the vehicle and the geomagnetic sensor, the more obvious the disturbance of the vehicle to the geomagnetic field will be, resulting in a stronger change in the induction signal. If the parking space occupied by the vehicle lateral is larger, the geomagnetic induction device will detect a stronger magnetic field change. The larger the longitudinal coverage size, the longer the length of the vehicle covering the parking space, thereby expanding the scope of influence on the magnetic field and increasing the change of the geomagnetic induction signal. Longer vehicles are distributed more widely in the parking space, which will cause the geomagnetic sensor to have a larger change in the magnetic field in the front and rear areas of the vehicle, thereby increasing the value of the geomagnetic perception index.

[0059] Furthermore, it is determined whether manual review is needed based on the identification result of the ordered parking space of the pre-departing vehicle, and the specific steps include: if the identification result of the ordered parking space of the pre-departing vehicle is that the pre-departing vehicle has left the ordered parking space, then a shape detection is performed on the pre-departing vehicle when the pre-departing vehicle leaves the parking lot, and the vehicle shape review index and license plate review information of the pre-departing vehicle are obtained, and compared with the vehicle shape index and license plate information of the pre-departing vehicle respectively, to determine again whether manual review is needed; if the identification result of the ordered parking space of the pre-departing vehicle is that the pre-departing vehicle has not left the ordered parking space, then a manual review is performed on the pre-departing vehicle.

[0060] In this embodiment, when the order parking space identification result of the pre-departing vehicle shows that it has left the order parking space, the appearance detection of the vehicle will be automatically triggered. The maximum length, maximum width, maximum height, wheel size and average grayscale analysis of the vehicle color of the pre-departing vehicle are obtained to obtain the vehicle appearance index before the pre-departing vehicle leaves, and are recorded as the vehicle appearance review index of the pre-departing vehicle. The license plate information is identified and verified by license plate recognition technology (such as OCR technology). If the license plate information is consistent with the vehicle information previously stored in the AI ​​intelligent parking review platform, it indicates that the vehicle has left correctly. If the stored vehicle information is inconsistent, a manual review is required. If the order parking space identification result of the pre-departing vehicle is that the pre-departing vehicle has not left the order parking space, the pre-departing vehicle is manually reviewed. By combining the appearance detection and license plate review information, the AI ​​intelligent parking review platform can accurately determine whether the vehicle has actually left the parking space when it leaves. This process not only relies on geomagnetic induction parameters, but also combines video surveillance data, which improves the adaptability of the AI ​​intelligent parking review platform to complex environments and avoids the limitations of a single technology. When errors occur in the vehicle appearance or license plate information, manual review can intervene in time to ensure the accuracy of the AI ​​intelligent parking review platform's judgment, thereby reducing the risk of wrong judgments. When the vehicle information matches correctly, the AI ​​intelligent parking review platform automatically confirms, avoiding manual review, thereby saving the cost and time of manual review and improving the automation level of parking lot management.

[0061] Further, the vehicle appearance index and license plate information of the vehicle to be driven away are compared respectively to determine whether manual review is required, specifically including: if the vehicle appearance review index of the vehicle to be driven away is consistent with the vehicle appearance index of the vehicle to be driven away and the license plate information of the vehicle to be driven away is consistent with the license plate review information of the vehicle to be driven away, then the intelligent review is passed and no manual review is performed; if the vehicle appearance review index of the vehicle to be driven away is inconsistent with the vehicle appearance index of the vehicle to be driven away or the license plate information of the vehicle to be driven away is inconsistent with the license plate review information of the vehicle to be driven away, then an intelligent review is performed, and the specific process is: obtaining the geomagnetic perception index of the order parking space and the environmental impact parameters within a preset period of time after the order upload time point, and evaluating to obtain the geomagnetic perception secondary evaluation index; The preset secondary evaluation first threshold and secondary evaluation second threshold in the database are obtained, and compared with the geomagnetic perception secondary evaluation index to obtain the secondary recognition result of the order parking space of the pre-departing vehicle; if the geomagnetic perception secondary evaluation index is less than the secondary evaluation first threshold or the geomagnetic perception secondary evaluation index is greater than the secondary evaluation second threshold, then the secondary recognition result of the order parking space of the pre-departing vehicle is that the pre-departing vehicle leaves the order parking space; if the geomagnetic perception secondary evaluation index is above the secondary evaluation first threshold and below the secondary evaluation second threshold, then the secondary recognition result of the order parking space of the pre-departing vehicle is that the pre-departing vehicle does not leave the order parking space; the environmental influencing parameters include the average ambient air pressure, average ambient humidity and average ambient temperature in the preset period after the order upload time point.

[0062] In this embodiment, the geomagnetic perception secondary evaluation index is obtained by:

[0063]

[0064] Wherein, CW2 represents the secondary evaluation index of geomagnetic perception, Q represents the average ambient air pressure in the preset period past the time point of order upload, ΔQ represents the reference value of ambient air pressure, S represents the average ambient humidity in the preset period past the time point of order upload, ΔS represents the reference value of ambient humidity, W represents the average ambient temperature in the preset period past the time point of order upload, ΔW represents the reference value of ambient temperature, e represents the natural constant, and CW represents the geomagnetic perception index of the order parking space.

[0065] It should be noted that the ambient air pressure reference value, ambient humidity reference value and ambient temperature reference value can be obtained through the database. The average ambient air pressure and average ambient humidity in the preset period after the order upload time can be measured by a barometer and a hygrometer. The geomagnetic perception secondary evaluation index obtained by analyzing the average ambient air pressure and average ambient humidity in the preset period after the order upload time and the geomagnetic perception index of the order parking space takes into account the impact of changes in environmental factors on geomagnetic induction. The greater the change in environmental factors, the greater the change in geomagnetic induction.

[0066] like Figure 2 As shown, it is a structural schematic diagram of the AI-based parking order intelligent review system provided in the embodiment of the present application. The AI-based parking order intelligent review system provided in the embodiment of the present application includes: a signal transmission module, a pre-departure vehicle appearance acquisition module, a geomagnetic induction evaluation module and a manual review and judgment module; the signal transmission module is used for automatically generating a parking order when the AI ​​intelligent parking review platform receives a vehicle pre-departure signal, and synchronously triggering the transmission of an order verification signal to the edge master control processor; the pre-departure vehicle appearance acquisition module is used for acquiring the pre-departure vehicle appearance after the edge master control processor receives the order verification signal. The image of the vehicle entering the parking lot is collected, thereby matching the parking space information and vehicle appearance parameters of the vehicle to be driven out, and marking the parking space as the ordered parking space; the geomagnetic induction evaluation module is used to obtain the historical geomagnetic perception change parameters of the ordered parking space, and analyze the geomagnetic perception influencing factors of the ordered parking space, obtain the geomagnetic perception parameters of the ordered parking space, and analyze them with the geomagnetic perception influencing factors of the ordered parking space to obtain the geomagnetic perception indicators of the ordered parking space, thereby judging the identification result of the ordered parking space of the vehicle to be driven out; the manual review judgment module is used to judge whether manual review is needed according to the identification result of the ordered parking space of the vehicle to be driven out.

[0067] In this embodiment, in the actual application of the prior art, there are many problems with the camera, for example, the working effect of the camera is heavily dependent on external lighting conditions. At night or in low light environments, the recognition effect of the camera is greatly reduced, resulting in reduced accuracy of license plate recognition and vehicle shape recognition, or even failure to work. Camera monitoring involves the collection of image and video data. There may be occlusion problems between vehicles in the parking lot, especially interference between large or high vehicles and low vehicles. The camera cannot accurately capture all information. Some parking spaces are in the blind spot of the camera and may not be clearly recognized, resulting in the inability to obtain the occupancy status of the parking space in real time.

[0068] Therefore, this solution uses geomagnetic induction technology to infer the presence of a vehicle by monitoring the changes in the geomagnetic field near the parking space, and further analyzes the parking status based on the geomagnetic induction parameters. Compared with cameras, geomagnetic induction technology does not rely on external lighting, so it is not affected by light changes and weather conditions (such as rainy days, haze, etc.) and the external environment. Geomagnetic induction technology always maintains a stable working state. Unlike cameras, geomagnetic induction technology only monitors changes in the ground magnetic field. It does not need to capture images or videos of vehicles to avoid the inability to obtain information in a timely and accurate manner. The equipment cost of geomagnetic sensors is relatively low, and basically no maintenance is required after installation, reducing operating costs. This solution combines cameras and geomagnetic induction technology to achieve complementary advantages. When the vehicle is scratched or other unexpected situations occur, the unexpected problem can be effectively solved by searching for monitoring. By jointly analyzing the parking situation of the vehicle with monitoring information and geomagnetic induction technology, it is beneficial to the management of the vehicle.

[0069] The geomagnetic perception index of the ordered parking space is obtained by analyzing the changes in the geomagnetic field generated by the vehicle in the parking space. This index mainly reflects the degree of geomagnetic influence on the parking space, and helps to determine whether the pre-departure vehicle has left the ordered parking space. If the vehicle leaves the parking space, the geomagnetic field will change, generating a clear geomagnetic perception signal. By analyzing this change, the occupancy status of the parking space can be accurately determined. The geomagnetic perception index combines multiple parameters (such as the horizontal and longitudinal dimensions of the vehicle) for accurate judgment, avoiding misjudgments and omissions that may occur in single camera or video analysis.

[0070] The secondary evaluation index of geomagnetic perception is a further detailed analysis after the preliminary analysis of geomagnetic perception index. Its main purpose is to improve the accuracy of the evaluation by comprehensively considering environmental factors (such as air pressure, humidity, temperature, etc.) to prevent the problem of misjudgment caused by the influence of external environmental temperature, air pressure and humidity on geomagnetic perception index.

[0071] If the identification result of the ordered parking space of the vehicle to be departed is that the vehicle to be departed has not left the ordered parking space, the vehicle to be departed will be manually reviewed to prevent vehicle violations from affecting the vehicle review.

[0072] Among them, the embodiment of the present application also provides a computer-readable storage medium for storing a program, which, when executed by a processor, implements an AI-based intelligent review method for parking orders.

[0073] To summarize, this embodiment evaluates the vehicle appearance indicators of the pre-departing vehicle and the geomagnetic induction influence indicators of the order parking space, thereby determining the order parking space recognition result of the pre-departing vehicle, thereby achieving accurate recognition of vehicles in complex environments, and effectively solving the problem of low parking order review and recognition accuracy and high misjudgment rate caused by single video detection in the prior art.

[0074] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention 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 code.

[0075] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of 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 produce 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.

[0076] 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.

[0077] 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. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0078] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0079] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. The AI-based parking order intelligent review method is characterized by: The following steps are involved: When the AI ​​intelligent parking review platform receives the vehicle's pre-departure signal, it automatically generates a parking order and simultaneously triggers the order verification signal to be transmitted to the edge master control processor; After receiving the order verification signal, the edge master control processor obtains the approaching image of the vehicle to be driven out, thereby matching the parking space information and vehicle appearance parameters of the vehicle to be driven out, and marks the parking space as the ordered parking space; Obtain the historical geomagnetic perception change parameters of the order parking space, and analyze to obtain the geomagnetic perception influencing factors of the order parking space, obtain the geomagnetic perception parameters of the order parking space, and analyze them with the geomagnetic perception influencing factors of the order parking space to obtain the geomagnetic perception index of the order parking space, thereby determining the order parking space recognition result of the vehicle to be driven out; Determine whether manual review is required based on the order parking space identification results of the vehicle that is expected to leave.

2. The AI-based parking order intelligent review method according to claim 1, characterized in that: The step of acquiring the approaching captured image of the vehicle to be driven out, thereby matching the parking space information and the vehicle appearance parameters of the vehicle to be driven out, and marking the parking space as the ordered parking space specifically includes: Acquire the approaching image of the vehicle to be driven out, identify the license plate information, parking space information and vehicle appearance parameters of the vehicle to be driven out, and mark the parking space as the ordered parking space; The vehicle appearance parameters include: the maximum length of the vehicle to be driven away, the maximum width of the vehicle, the maximum height of the vehicle, the wheel size and the average grayscale of the vehicle color; Analyzing based on the vehicle shape parameters, obtaining the vehicle shape index of the vehicle to be driven away; The vehicle appearance index of the pre-departing vehicle is used to characterize the appearance recognition of the pre-departing vehicle.

3. The AI-based parking order intelligent review method according to claim 2, characterized in that: The steps of obtaining the historical geomagnetic perception change parameters of the ordered parking space and analyzing the geomagnetic perception influencing factors of the ordered parking space include: The historical geomagnetic perception change parameters of the ordered parking space include the daily average geomagnetic induction delay time and the daily average magnetic field intensity change, the daily average magnetic induction intensity change, the daily average magnetic flux rate change and the daily average magnetic permeability change; Obtaining a geomagnetic induction reference set preset in a database, wherein the geomagnetic induction reference set includes a daily average geomagnetic sensor delay reference value and a daily average magnetic field intensity change reference value, a daily average magnetic induction intensity change reference value, a daily average magnetic flux rate change reference value, and a daily average magnetic permeability change reference value; The geomagnetic induction influence index of the ordered parking space is obtained by comparing the historical geomagnetic perception change parameters of the ordered parking space with the geomagnetic induction reference set; Based on the order parking space geomagnetic induction influence index, the order parking space geomagnetic perception influence factor is obtained; The geomagnetic induction impact index of the ordered parking space is used to characterize the intensity of interference to the ordered parking space in history.

4. The AI-based parking order intelligent review method according to claim 3, characterized in that: The specific steps of analyzing the geomagnetic induction influence index of the order parking space to obtain the geomagnetic perception influence factor of the order parking space include: Obtain the local magnetic induction influence index intervals preset in the database and the geomagnetic induction influence factors corresponding to the local magnetic induction influence index intervals, match the geomagnetic induction influence index of the order parking space with the local magnetic induction influence index intervals preset in the database; if the geomagnetic induction influence index of the order parking space is within a preset geomagnetic induction influence index interval, obtain the geomagnetic induction influence factor corresponding to the geomagnetic induction influence index interval as the geomagnetic induction influence factor of the order parking space.

5. The AI-based parking order intelligent review method according to claim 1, characterized in that: The step of obtaining the geomagnetic sensing index of the order parking space and determining the identification result of the order parking space of the vehicle to be driven out specifically includes: Based on the vehicle appearance parameters, the maximum lateral dimension and the maximum longitudinal dimension of the vehicle are obtained by statistics; Obtain the geomagnetic perception parameters of the order parking space and analyze them with the maximum lateral dimension of the vehicle, the maximum longitudinal dimension of the vehicle, and the geomagnetic induction influencing factor of the order parking space to obtain the geomagnetic perception index of the order parking space; The geomagnetic perception index of the ordered parking space is used to characterize the degree of geomagnetic influence of the vehicle to be driven out at the ordered parking space; The geomagnetic perception parameters of the ordered parking space include the vehicle lateral coverage size, the vehicle longitudinal coverage size, the magnetic induction intensity and the magnetic field intensity; The first geomagnetic perception evaluation threshold and the second geomagnetic perception evaluation threshold preset in the database are obtained, and compared with the geomagnetic perception index of the order parking space to obtain the order parking space recognition result of the pre-leaving vehicle. If the geomagnetic perception index of the order parking space is less than the first geomagnetic perception evaluation threshold or the geomagnetic perception index of the order parking space is greater than the second geomagnetic perception evaluation threshold, then the order parking space recognition result of the pre-leaving vehicle is that the pre-leaving vehicle leaves the order parking space; if the geomagnetic perception index of the order parking space is above the first geomagnetic perception evaluation threshold and below the second geomagnetic perception evaluation threshold, then the order parking space recognition result of the pre-leaving vehicle is that the pre-leaving vehicle has not left the order parking space.

6. The AI-based parking order intelligent review method according to claim 5, characterized in that: The specific method for obtaining the geomagnetic perception index of the order parking space is as follows: In the formula, CW represents the geomagnetic perception index of the parking space, e represents the natural constant, and DC y represents the geomagnetic induction factor of the parking space, L x Indicates the lateral coverage size of the vehicle, ΔL x Indicates the maximum lateral dimension of the vehicle, L y Indicates the longitudinal coverage dimension of the vehicle, ΔL y Indicates the maximum longitudinal dimension of the vehicle, Q a Represents the magnetic induction intensity, ΔQ a Indicates the reference value of magnetic induction intensity, Q b represents the magnetic field strength, ΔQ b Indicates reference value.

7. The AI-based parking order intelligent review method according to claim 5, characterized in that: The specific steps of judging whether manual review is required based on the order parking space identification result of the vehicle to be driven out include: If the result of the identification of the ordered parking space of the pre-leaving vehicle is that the pre-leaving vehicle has left the ordered parking space, the pre-leaving vehicle is subjected to appearance inspection when it leaves the parking lot, and the vehicle appearance review index and license plate review information of the pre-leaving vehicle are obtained, and the vehicle appearance review index and license plate review information of the pre-leaving vehicle are respectively compared with the vehicle appearance index and license plate information of the pre-leaving vehicle, and it is determined again whether manual review is required; If the identification result of the ordered parking space of the vehicle to be departed is that the vehicle to be departed has not left the ordered parking space, the vehicle to be departed will be manually reviewed.

8. The AI-based parking order intelligent review method according to claim 7, characterized in that: The comparison with the vehicle appearance index and license plate information of the vehicle to be driven out respectively to determine whether manual review is required specifically includes: If the vehicle appearance review index of the vehicle to be driven away is consistent with the vehicle appearance index of the vehicle to be driven away, and the license plate information of the vehicle to be driven away is consistent with the license plate review information of the vehicle to be driven away, the intelligent review is passed and no manual review is performed; If the vehicle appearance review index of the vehicle to be driven out is inconsistent with the vehicle appearance index of the vehicle to be driven out, or the license plate information of the vehicle to be driven out is inconsistent with the license plate review information of the vehicle to be driven out, an intelligent review is performed. The specific process is: obtaining the geomagnetic perception index of the order parking space and the environmental impact parameters within a preset period of time after the order upload time point, and evaluating to obtain the geomagnetic perception secondary evaluation index; Obtain a second evaluation first threshold and a second evaluation second threshold preset in a database, and compare them with the geomagnetic perception second evaluation index to obtain a second recognition result of the order parking space of the pre-departing vehicle; if the geomagnetic perception second evaluation index is less than the second evaluation first threshold or the geomagnetic perception second evaluation index is greater than the second evaluation second threshold, then the second recognition result of the order parking space of the pre-departing vehicle is that the pre-departing vehicle has left the order parking space; if the geomagnetic perception second evaluation index is above the second evaluation first threshold and below the second evaluation second threshold, then the second recognition result of the order parking space of the pre-departing vehicle is that the pre-departing vehicle has not left the order parking space; The environmental impact parameters include the average ambient air pressure, average ambient humidity and average ambient temperature within a preset period of time past the order upload time point.

9. AI-based parking order intelligent review system, characterized by: include: Signal transmission module, pre-departure vehicle appearance acquisition module, geomagnetic induction evaluation module and manual review and judgment module; The signal transmission module is used to automatically generate a parking order when the AI ​​intelligent parking review platform receives the vehicle pre-departure signal, and synchronously trigger the order verification signal to be transmitted to the edge master control processor; The pre-departure vehicle appearance acquisition module is used to acquire the approaching captured image of the pre-departure vehicle after the edge master control processor receives the order verification signal, thereby matching the parking space information and vehicle appearance parameters of the pre-departure vehicle, and marking the parking space as the ordered parking space; The geomagnetic induction evaluation module is used to obtain the historical geomagnetic perception change parameters of the order parking space, and analyze to obtain the geomagnetic perception influencing factor of the order parking space, obtain the geomagnetic perception parameters of the order parking space, and analyze them with the geomagnetic perception influencing factor of the order parking space to obtain the geomagnetic perception index of the order parking space, thereby determining the order parking space identification result of the vehicle to be driven out; The manual review judgment module is used to judge whether manual review is required based on the order parking space identification result of the vehicle to be driven out.

10. A computer-readable storage medium for storing a program, characterized in that: When the program is executed by the processor, the AI-based intelligent parking order review method as described in any one of claims 1 to 8 is implemented.

Images