Parking lot anti-following and fee-avoiding control system and method based on multi-vision and distance detection

Abstract

This invention discloses a parking lot anti-following-toll-evasion control system and method based on multi-view vision and spacing detection. The system includes an exit channel detection unit, a payment status recognition unit, a following-toll-evasion determination unit, and a gate enforcement control unit. The following-toll-evasion determination unit achieves accurate identification of following-toll-evasion behavior through a multi-condition joint judgment model. The gate enforcement control unit injects a high-priority forced closing signal into the gate controller to cover the anti-collision function and adopts a safe deceleration and descent strategy to achieve accurate interception. At the same time, it automatically captures images of vehicles evading tolls and uploads them to the cloud as evidence for recovery. This application can improve the interception rate of following-toll-evasion in parking lots, reduce the false judgment rate, and has the advantages of high security, low transformation cost, and complete evidence. It is suitable for exit management of various types of enclosed parking lots.

Description

Technical Field

[0001] This invention relates to the field of intelligent parking management technology, and in particular to a parking lot anti-following-and-escape-paying control system based on multi-view vision and spacing detection, and a parking lot anti-following-and-escape-paying control method based on multi-view vision and spacing detection. Background Technology

[0002] Parking fee management is a crucial aspect of urban traffic management and commercial operations. Currently, the vast majority of parking lot exits use barrier gate systems to control vehicle entry and exit, in conjunction with license plate recognition or IC card reading systems for fee management. To ensure driving safety, existing barrier gate equipment is generally equipped with anti-collision sensors (including inductive loops, infrared beam sensors, or pressure sensors). Their working principle is as follows: when the barrier gate arm is descending, if the anti-collision sensor detects an obstacle (including a vehicle) within the area covered by the gate arm, it immediately stops descending and reverses to raise, preventing the barrier gate from damaging the vehicle.

[0003] However, the aforementioned anti-smashing function has been exploited by some unscrupulous drivers, forming a common parking fee evasion tactic – "following the car to evade payment." The specific pattern is as follows: the vehicle evading payment waits at the exit for a vehicle that is paying normally to clear the way. After the vehicle in front finishes paying, the barrier lifts, and the vehicle evading payment follows closely behind (with a very small distance between vehicles, usually within 1 to 3 meters) and quickly enters the exit lane. Because the barrier's anti-smashing sensor continuously detects a vehicle in the lane, the barrier cannot be lowered, allowing the vehicle evading payment to smoothly exit the parking lot while the barrier remains raised, without paying any parking fees.

[0004] The existing technical solutions for addressing the aforementioned toll evasion behaviors have the following shortcomings: (1) Traditional solutions rely on manual monitoring: relying on parking lot managers for on-site monitoring, but manual judgment and response are slow, and it is difficult to achieve 24-hour uninterrupted monitoring, resulting in high management costs and low actual interception rate.

[0005] (2) Delayed closing scheme for the barrier gate: Some parking lots set a fixed delay before closing the barrier gate after the vehicle in front passes through. However, the delay time is fixed and cannot adapt to different exit channel lengths and vehicle speeds. If the delay is too long, it will not be able to prevent following vehicles. If the delay is too short, there is a safety hazard of smashing the vehicle.

[0006] (3) Pure camera image recognition solution: Existing image recognition-based solutions are usually only used for license plate recognition, lack accurate measurement of the real-time distance between multiple vehicles in the channel, and are not deeply integrated with the gate anti-smashing control logic, so they cannot realize real-time judgment of following behavior and active control of the gate.

[0007] (4) None of the existing solutions have resolved the contradiction between the anti-smashing function and the following vehicle interception, that is, how to forcibly prevent the following vehicle from leaving while ensuring safety without canceling the anti-smashing function (to prevent injury). Summary of the Invention

[0008] The present invention aims to at least partially solve one of the technical problems in the aforementioned technologies. Therefore, one objective of the present invention is to propose a parking lot anti-following-toll-evasion control system based on multi-view vision and distance detection, which can improve the interception rate of following vehicles to avoid toll evasion and reduce the false alarm rate.

[0009] The second objective of this invention is to propose a parking lot anti-following payment control method based on multi-view vision and spacing detection.

[0010] To achieve the above objectives, a first aspect of the present invention proposes a parking lot anti-following and evasion control system based on multi-view vision and spacing detection, comprising: an exit lane detection unit, which is set within the parking lot exit lane to acquire the position, speed, and distance between adjacent vehicles for all vehicles within the lane; a payment status recognition unit, which is communicatively connected to the parking lot fee management system to acquire the payment status of vehicles entering the exit lane in real time and bind the vehicle payment status with the corresponding vehicle identifier; and a following determination unit, which is connected to both the exit lane detection unit and the payment status recognition unit. The system is connected to the following vehicle controller and is used to determine whether a following vehicle is engaging in toll evasion based on the distance between adjacent vehicles, relative speed, the payment status of the preceding vehicle, and the payment record of the following vehicle, so as to obtain the toll evasion determination result; the barrier gate forced control unit is hardware connected to the following vehicle determination unit and the anti-smashing signal input terminal of the barrier gate controller respectively. When it is determined that there is toll evasion based on the following vehicle toll evasion determination result, the barrier gate forced closing command with a priority higher than the anti-smashing signal is injected into the barrier gate controller, so as to control the barrier gate to fall within a preset delay time after the preceding vehicle has completely passed through the barrier gate coverage area, and at the same time, the barrier gate falling speed is dynamically adjusted according to the distance between the following vehicle and the barrier gate during the falling process.

[0011] According to an embodiment of the present invention, a parking lot anti-following and payment evasion control system based on multi-view vision and spacing detection includes an exit lane detection unit, a payment status recognition unit, a following vehicle determination unit, and a barrier gate enforcement control unit. The exit lane detection unit is located within the parking lot exit lane area to acquire the position, speed, and distance between adjacent vehicles for all vehicles within the lane area. The payment status recognition unit is communicatively connected to the parking lot fee management system to acquire the payment status of vehicles entering the exit lane in real time and bind the vehicle payment status to the corresponding vehicle identifier. The following vehicle determination unit is connected to both the exit lane detection unit and the payment status recognition unit to determine the distance between adjacent vehicles. The system uses vehicle spacing, relative speed, the toll payment status of the preceding vehicle, and the toll payment record of the following vehicle to determine whether a vehicle is attempting to evade tolls. The gate's forced control unit is hardware-connected to both the following vehicle determination unit and the gate controller's anti-collision signal input. When the following vehicle evasion is determined to exist based on the toll evasion determination result, the control unit injects a forced closing command with a higher priority than the anti-collision signal into the gate controller. This command controls the gate to descend within a preset delay after the preceding vehicle has completely passed through the gate's coverage area. Simultaneously, the descending speed is dynamically adjusted based on the distance between the following vehicle and the gate. This improves the interception rate of following vehicles attempting to evade tolls and reduces the false alarm rate.

[0012] In addition, the parking lot anti-following payment evasion control system based on multi-view vision and distance detection proposed in the above embodiments of the present invention may also have the following additional technical features: Optionally, the parking lot anti-following-toll-evasion control system based on multi-view vision and spacing detection also includes an alarm and recording unit. The alarm and recording unit is connected to the following-toll-evasion determination unit. When it is determined that there is a following-toll-evasion behavior based on the following-toll-evasion determination result, the alarm is activated simultaneously with sound and light alarms, and multiple frames of image evidence of the evading vehicle are captured. The license plate number, evasion time, lane number and image evidence are encrypted, stored and then uploaded to the cloud parking management platform.

[0013] Optionally, the exit channel detection unit includes a first inductive loop, a second inductive loop, a third inductive loop, a binocular camera, a millimeter-wave radar sensor, and an edge computing module; The first, second, and third inductive loops correspond to the first, second, and third detection areas, respectively. The first detection area is located directly below the gate arm, the second detection area is located behind the first detection area, and the third detection area is located outside the gate arm, so as to determine the detection area of ​​the vehicle through the inductive loop signals. The binocular camera is installed on the top of the passage, and the millimeter-wave radar sensor is installed on the side wall of the passage, so as to use a deep neural network to perform real-time target detection on the camera image, identify the outline bounding box and centroid coordinates of all vehicles within the passage range, and calculate the distance and speed of adjacent vehicles by the difference of centroid coordinates between consecutive frames. The edge computing module is used to perform time synchronization and spatial registration of the position, speed and distance between adjacent vehicles of all vehicles within the passage range.

[0014] Optionally, the payment status recognition unit includes a license plate recognition module, which is used to collect vehicle images and perform OCR recognition on the collected vehicle images to obtain the license plate number; and send the license plate number to the database of the toll management system to query the payment status of the corresponding vehicle, thereby completing the binding of the license plate and the payment status.

[0015] Optionally, the following vehicle determination unit has a built-in multi-condition joint determination model. The multi-condition joint determination model is used to determine the following vehicle as a toll evasion behavior when multiple preset conditions are met at the same time, otherwise it is determined as normal passage. The multiple preset conditions include the preceding vehicle having paid the toll and the barrier gate being raised, the longitudinal distance between the following vehicle and the preceding vehicle being less than a preset safety threshold, and the following vehicle having a speed greater than 0 and no corresponding toll payment record.

[0016] Optionally, the barrier gate forced control unit controls the barrier gate to fall within a preset delay time after the third ground induction coil in the third detection area confirms that the vehicle in front has completely left the passage, and injects a high-priority forced closing signal into the barrier gate controller to cover the original anti-smashing sensor signal.

[0017] Optionally, during the lowering process of the barrier gate, the lowering speed is dynamically adjusted based on the distance between the following vehicle and the end of the barrier arm detected by the millimeter-wave radar sensor; Specifically, if the distance between the following vehicle and the end of the gate arm is greater than the first preset threshold, the gate will descend at the standard speed; if the distance between the following vehicle and the end of the gate arm is greater than or equal to the second preset threshold and less than or equal to the first preset threshold, the gate will descend at a low speed of 30% to 50% of the standard speed; if the distance between the following vehicle and the end of the gate arm is less than the second preset threshold, the descent will stop.

[0018] Optionally, the alarm and recording unit includes an audible and visual alarm, a high-definition snapshot camera, and a data storage module; The sound and light alarm is installed within the exit passage area and is used to emit a dedicated toll evasion alarm sound and flashing light; the high-definition capture camera is aimed at the exit passage and automatically captures no less than 3 frames of close-up images of the vehicle's front and license plate after the toll evasion is detected; the data storage module is used to store the license plate number, toll evasion time, passage number and image evidence in an encrypted manner to the local storage device and upload them to the cloud parking management platform in real time.

[0019] Optionally, the parking lot anti-following toll evasion control system based on multi-view vision and spacing detection also includes a multi-vehicle scene coordination module. When there are two or more vehicles in the exit lane at the same time, the multi-vehicle scene coordination module establishes a tracking object for each vehicle independently, maintains the payment status, location and speed information of each vehicle, and independently executes the following toll evasion judgment logic for each pair of adjacent vehicles.

[0020] To achieve the above objectives, a second aspect of the present invention proposes a parking lot anti-following payment evasion control method based on multi-view vision and spacing detection, applied to the parking lot anti-following payment evasion control system based on multi-view vision and spacing detection described in any of the first aspects, comprising the following steps: performing OCR recognition on the acquired vehicle image to obtain the corresponding license plate information, and querying the payment status of the corresponding vehicle in the database of the payment management system according to the license plate information; if payment has been made, sending a gate-raising command to the barrier controller, and simultaneously activating the exit channel detection unit to enter the working state; exit channel detection... The unit acquires the position, speed, and distance between adjacent vehicles for all vehicles within the channel range. Based on the distance between adjacent vehicles, relative speed, payment status of the preceding vehicle, and payment record of the following vehicle, it determines whether the following vehicle is engaging in toll evasion behavior, thus obtaining a toll evasion determination result. When toll evasion behavior is determined to exist based on the toll evasion determination result, a forced closing command with a priority higher than the anti-smashing signal is injected into the barrier gate controller. This allows the barrier gate to be controlled to fall within a preset delay after the preceding vehicle has completely passed through the area covered by the barrier arm. Simultaneously, the falling speed of the barrier gate is dynamically adjusted based on the distance between the following vehicle and the barrier arm during the falling process. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the overall architecture of a parking lot anti-following toll evasion control system based on multi-view vision and spacing detection according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the outlet channel sensor layout according to an embodiment of the present invention, wherein, Figure 2 (a) is a top view of the exit passage. Figure 2 (b) is a side view of the exit passage; Figure 3 This is a schematic diagram of the forced control timing of a barrier gate according to an embodiment of the present invention, wherein, Figure 3 (a) shows the position of the barrier gate arm. Figure 3(b) shows the longitudinal position of the vehicle in front. Figure 3 (c) represents the change in longitudinal spacing between the front and rear vehicles. Figure 3 (d) is the timing sequence of the anti-smashing signal and the forced shutdown signal; Figure 4 This is a schematic diagram of a safe deceleration descent strategy according to an embodiment of the present invention, wherein, Figure 4 (a) is a schematic diagram of the three stages of the barrier gate arm's descent. Figure 4 (b) is a schematic diagram of the curve of the gate arm's falling speed versus the gate arm's height above the ground; Figure 5 This is a flowchart illustrating a parking lot anti-following evasion control method based on multi-view vision and spacing detection according to an embodiment of the present invention. Figure 6 This is a flowchart illustrating the logic for determining toll evasion by following another vehicle and controlling the gate according to a specific embodiment of the present invention. Detailed Implementation

[0022] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0023] To better understand the above technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present invention and to fully convey the scope of the invention to those skilled in the art.

[0024] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0025] Figure 1 This is a schematic diagram of the overall architecture of a parking lot anti-following toll evasion control system based on multi-view vision and distance detection according to an embodiment of the present invention, as shown below. Figure 1 As shown, the parking lot anti-following toll evasion control system includes: an exit channel detection unit 10, a payment status recognition unit 20, a following vehicle determination unit 30, and a barrier gate forced control unit 40.

[0026] The system includes an exit channel detection unit 10, which is set within the parking lot exit channel area to obtain the position, speed, and distance between adjacent vehicles for all vehicles within the channel area; a payment status recognition unit 20, which is connected to the parking lot fee management system to obtain the payment status of vehicles entering the exit channel in real time and bind the vehicle payment status with the corresponding vehicle identifier; a following vehicle determination unit 30, which is connected to both the exit channel detection unit 10 and the payment status recognition unit 20, to determine whether a following vehicle is engaging in fare evasion based on the distance between adjacent vehicles, relative speed, the payment status of the preceding vehicle, and the payment record of the following vehicle; and a barrier gate forced control unit 40, which is hardware connected to both the following vehicle determination unit 30 and the anti-smashing signal input terminal of the barrier gate controller. When the following vehicle evasion determination result indicates the existence of fare evasion, the control unit injects a forced closing command with a priority higher than the anti-smashing signal into the barrier gate controller to control the barrier gate to fall within a preset delay after the preceding vehicle has completely passed through the barrier gate coverage area. During the falling process, the control unit dynamically adjusts the falling speed of the barrier gate based on the distance between the following vehicle and the barrier gate.

[0027] Specifically, the exit lane detection unit 10 is responsible for collecting the physical status information of vehicles within the exit lane. The hardware components include: three sets of inductive loops arranged sequentially along the exit lane (corresponding to the first detection area A, the second detection area B, and the third detection area C, respectively); a binocular vision camera installed on the top of the lane (resolution no less than 1080P, frame rate no less than 30fps); and a millimeter-wave radar sensor installed on the side wall of the lane (operating frequency 77GHz, ranging accuracy better than 5cm, speed measurement accuracy better than 0.1m / s). After time synchronization and spatial registration of the data from the three types of sensors by the edge computing unit, the unit outputs the position coordinates (accuracy no less than 10cm), velocity vectors, and vehicle spacing (accuracy no less than 5cm) of each vehicle within the lane. When a vehicle enters the exit lane, the payment status recognition unit 20 reads the license plate using a license plate recognition module (OCR accuracy no less than 99.5%), completes the payment status query within 200ms, and binds the "paid / unpaid" status marker to the license plate, pushing it to the following vehicle determination unit in real time. The following vehicle determination unit 30 receives real-time data from the exit channel detection unit and the payment status recognition unit, performs logical operations based on a multi-condition joint determination model, and outputs the following vehicle evasion determination result within the determination period (not exceeding 100ms). The barrier gate enforcement control unit 40 receives the determination result from the following vehicle determination unit, connects to the existing barrier gate controller via a standard RS-485 or relay hardware interface, and executes forced closing commands and speed reduction control commands. This unit is equipped with an independent power supply module to ensure independent operation even in the event of a parking management system failure.

[0028] As one embodiment, the exit channel detection unit 10 includes a first inductive loop, a second inductive loop, a third inductive loop, a binocular camera, a millimeter-wave radar sensor, and an edge computing module. The first, second, and third inductive loops correspond to a first detection area, a second detection area, and a third detection area, respectively. The first detection area is located directly below the barrier gate arm, the second detection area is located behind the first detection area, and the third detection area is located outside the barrier gate arm, allowing the inductive loop signals to determine the vehicle's location within the detection area. The binocular camera is installed at the top of the channel, and the millimeter-wave radar sensor is installed on the side wall of the channel. A deep neural network is used to perform real-time target detection on the camera footage, identifying the outline bounding boxes and centroid coordinates of all vehicles within the channel range. The distance and speed between adjacent vehicles are calculated using the difference in centroid coordinates between consecutive frames. The edge computing module is used to perform time synchronization and spatial registration of the positions, speeds, and distances between adjacent vehicles for all vehicles within the channel range.

[0029] Specifically, such as Figure 2 As shown, Figure 2 (a) is a top view of the exit passage. Figure 2 (b) is a side view of the exit lane. Taking a standard one-way single-lane exit lane (lane length 8 meters, width 3.5 meters) as an example, the sensor layout scheme is described: (a) First detection area A: Located in the range from directly below the gate arm to 0.5 meters in front of the gate arm, a first ground induction coil A1 (size 1.5m×0.6m) is deployed to accurately detect whether the gate arm has completely left the projection range of the roof of the vehicle in front and trigger the "vehicle in front has passed safely" confirmation signal.

[0030] (b) Second detection area B: Located 1.5 to 3.5 meters behind the first detection area, a second inductive loop B1 (1.5m × 0.6m in size) is deployed to detect whether subsequent vehicles have entered the exit lane and are in front of the gate coverage area, and to calculate the distance between vehicles A and B in combination with millimeter-wave radar data.

[0031] (c) Third detection area C: Located 0.5 to 1.5 meters outside the gate arm, a third ground induction coil C1 is deployed to confirm that the vehicle in front has completely driven out of the exit channel, as one of the final conditions for triggering the gate to close.

[0032] (d) A binocular vision camera is installed at the top of the passage, 2.5 to 3 meters above the ground, with a field of view covering the entire exit passage (horizontal angle not less than 90 degrees), complementing the millimeter-wave radar: the camera provides vehicle outline and license plate information, while the millimeter-wave radar provides accurate distance and speed data. The two data are fused after time alignment (time stamp accuracy better than 10ms).

[0033] As an example, the payment status recognition unit 20 includes a license plate recognition module, which is used to collect vehicle images and perform OCR recognition on the collected vehicle images to obtain the license plate number; and send the license plate number to the database of the toll management system to query the payment status of the corresponding vehicle, thereby completing the binding of the license plate and the payment status.

[0034] In other words, the license plate number is identified and bound when the vehicle enters the exit lane. By querying the parking fee system database, the payment status is confirmed within no more than 200 milliseconds, and the payment status is synchronized to the following vehicle judgment unit 30 in real time.

[0035] As an example, the following vehicle determination unit 30 has a built-in multi-condition joint determination model. The multi-condition joint determination model is used to determine the following vehicle as a toll evasion behavior when multiple preset conditions are met at the same time, otherwise it is determined as normal passage. The multiple preset conditions include the preceding vehicle's toll payment status being paid and the barrier gate being raised, the longitudinal distance between the following vehicle and the preceding vehicle being less than a preset safety threshold, and the following vehicle's driving speed being greater than 0 and having no corresponding toll payment record.

[0036] Specifically, the following vehicle detection unit 30 comprehensively determines the following vehicle evasion behavior based on the following conditions: (a) the toll payment status of the preceding vehicle is marked as "paid" and the barrier gate is in the rising state; (b) the following vehicle enters the exit channel and the longitudinal distance between it and the preceding vehicle is less than the preset safety threshold. The The value range is 1.5 meters to 3.5 meters; (c) the speed of the vehicle behind is greater than zero and it does not stop to wait for payment, and there is no corresponding payment record for the vehicle behind in the parking lot charging system; (d) when the above conditions (a), (b) and (c) are met at the same time, the following vehicle to evade payment judgment is triggered.

[0037] As an example, the barrier gate forced control unit 40 controls the barrier gate to fall within a preset delay time after the third ground induction coil in the third detection area confirms that the vehicle in front has completely driven out of the passage, and injects a high-priority forced closing signal into the barrier gate controller to cover the original anti-smashing sensor signal.

[0038] As an example, during the lowering process of the barrier gate, the lowering speed is dynamically adjusted based on the distance between the following vehicle and the end of the barrier arm detected by the millimeter-wave radar sensor. Specifically, if the distance between the following vehicle and the end of the barrier arm is greater than a first preset threshold, the barrier gate lowers at the standard speed; if the distance between the following vehicle and the end of the barrier arm is greater than or equal to a second preset threshold and less than or equal to the first preset threshold, the barrier gate lowers at a low speed of 30% to 50% of the standard speed; if the distance between the following vehicle and the end of the barrier arm is less than the second preset threshold, the lowering stops.

[0039] It should be noted that, see Figure 3 ,in, Figure 3 (a) shows the position of the barrier gate arm. Figure 3 (b) shows the longitudinal position of the vehicle in front. Figure 3 (c) represents the change in longitudinal spacing between the front and rear vehicles. Figure 3 (d) The timing sequence of the anti-smashing signal and the forced closing signal shows the time relationship between the passing vehicle, the coverage of the anti-smashing signal, and the forced lowering of the barrier gate. The specific parameters are as follows: At time T0: After the vehicle in front completes payment, the barrier gate begins to lift, taking approximately 1.5 seconds to complete (taking a common barrier gate as an example).

[0040] At time T1: The front of the vehicle reaches the position of the first inductive loop A1. At this time, the barrier gate has been fully raised and the system enters the following vehicle monitoring state.

[0041] At time T2: The rear of the vehicle in front passes the first inductive loop A1, and the first inductive loop A1 changes from "vehicle present" to "no vehicle present," generating a "vehicle in front has passed safely" confirmation signal. At the same time, if the second inductive loop B1 detects the following vehicle (and has triggered the toll evasion mark), the gate enforcement control unit prepares to issue a closing command.

[0042] T3 = T2 + ( (Default 0.5 seconds): The barrier gate forced control unit sends a forced closing command to the barrier gate controller, the anti-smashing signal is overridden, and the barrier gate begins to fall. The setting is adjustable from 0.3 seconds to 1.5 seconds to confirm that the rear of the vehicle in front has completely left the range of the gate, thus avoiding accidentally hitting the vehicle in front.

[0043] At time T4: The barrier gate is fully lowered, and the following vehicle is blocked in the passage. The time from T4 to T3 is the standard lowering time of the barrier gate (usually 1 to 2 seconds), which is about 2 to 4 seconds in deceleration mode.

[0044] See Figure 4 ,in, Figure 4 (a) is a schematic diagram of the three stages of the barrier gate arm's descent. Figure 4 (b) is a schematic diagram of the curve of the lowering speed of the barrier gate arm versus the height of the barrier gate arm above the ground. The safe deceleration lowering strategy shows the correspondence between the position of the following vehicle and the lowering speed of the barrier gate. The detailed implementation plan is as follows: The system moves the gate arm from its initial height H0 (raised position, approximately 2 meters above the ground) to its ground position. The falling process can be divided into three stages: Phase 1 (Free fall phase, H0 to...) ): The gate arm operates at standard speed (Typically 0.5m / s) Normal descent. During this stage, the gate arm is more than 1.5 meters above the ground, posing no risk of hitting vehicles, and descends rapidly to quickly block the toll evasion route.

[0045] Phase Two (Deceleration and Fall Phase) to When millimeter-wave radar detects that the roof of the following vehicle enters the horizontal projection range of the gate arm ( ≤1.0m), switch the barrier gate to low speed. (= 0.4× The vehicle will fall slowly at a speed of approximately 0.2 m / s to avoid impact damage to the roof.

[0046] Phase 3 (Light Touch Hold Phase, H_low): When the barrier arm contacts the roof of the following vehicle (or the rear of the roof, usually 0.3m from the rear of the vehicle), the barrier stops falling and maintains its current height, continuously applying a slight downward pressure (not exceeding 50N) to prevent the vehicle from continuing to move forward, waiting for the driver to stop voluntarily. At the same time, the intercom function is triggered, allowing the parking lot manager to communicate remotely with the driver.

[0047] The above three-stage control is achieved by the variable frequency speed regulation function of the barrier gate motor driver. It can be implemented on modern barrier gate products that support variable frequency speed regulation without replacing the mechanical structure of the barrier gate.

[0048] As an example, such as Figure 1 As shown, the parking lot anti-following toll evasion control system based on multi-view vision and spacing detection also includes an alarm and recording unit 50. The alarm and recording unit 50 is connected to the following vehicle determination unit 30. When it is determined that there is a following vehicle toll evasion behavior based on the following vehicle toll evasion determination result, the alarm will be activated simultaneously with sound and light alarms, and multiple frames of image evidence of the evading vehicle will be captured. The license plate number, evasion time, lane number and image evidence will be encrypted, stored and then uploaded to the cloud parking management platform.

[0049] As one embodiment, the alarm and recording unit 50 includes an audible and visual alarm, a high-definition capture camera, and a data storage module. The audible and visual alarm is installed within the exit passage area and is used to emit a dedicated toll evasion alarm sound and flashing lights. The high-definition capture camera is aimed at the exit passage and automatically captures no less than 3 frames of close-up images of the vehicle's front and license plate after the toll evasion determination is triggered. The data storage module is used to store the license plate number, toll evasion time, passage number, and image evidence in an encrypted manner to the local storage device and upload them to the cloud parking management platform in real time.

[0050] In other words, the alarm and recording unit 50 triggers an audible and visual alarm (alarm sound pressure ≥85dB, strobe light power ≥10W) and activates a high-definition capture camera (resolution not less than 2 million pixels) to capture multiple frames of the toll-evading vehicle, encrypt and store the evidence data on the local NAS and simultaneously upload it to the cloud parking management platform.

[0051] As an example, the parking lot anti-following toll evasion control system based on multi-view vision and spacing detection also includes a multi-vehicle scene coordination module. When there are two or more vehicles in the exit lane at the same time, the multi-vehicle scene coordination module establishes a tracking object for each vehicle independently, maintains the payment status, location and speed information of each vehicle, and independently executes the following toll evasion judgment logic for each pair of adjacent vehicles.

[0052] It should be noted that when there are three or more vehicles in the exit lane at the same time (e.g., two vehicles paying tolls pass through in sequence, and the third vehicle is evading tolls), the multi-vehicle scenario coordination module handles the situation as follows: (a) The system establishes an independent tracking object for each vehicle in the channel, assigns a unique ID, and maintains its respective payment status, location, and speed information.

[0053] (b) The following vehicle determination unit independently executes the determination logic for each pair of adjacent vehicles (in order of driving direction, namely the first vehicle, the second vehicle, the third vehicle, etc.): it performs a determination once for the first vehicle and the second vehicle, a determination once for the second vehicle and the third vehicle, and so on.

[0054] (c) The timing of gate closure shall be based on the combination of "the vehicle in front has completely passed through and the vehicle behind is evading tolls" between the two closest vehicles, taking the most conservative approach. (Maximum value) to ensure the vehicle in front has passed safely before closing the barrier.

[0055] In addition, considering the system's reliability under extreme conditions, this application includes the following fail-safe mechanism: (a) Sensor failure: When a sensor fails (determined by self-test signal), the system is downgraded to "camera only mode" or "radar only mode" and a maintenance alarm is triggered at the same time; if a critical sensor (such as ground induction coil A1) fails completely, the system reverts to manual review mode and an alarm prompts the administrator to intervene.

[0056] (b) Protection against false alarms: To prevent vehicles that are paying tolls and following the vehicle (such as the escort vehicle of a large moving vehicle) from being mistakenly identified as evading tolls, the system has a "whitelist" function. Pre-authorized license plate numbers can be exempted from being identified as evading tolls.

[0057] (c) Power outage protection: The gate control unit is equipped with a UPS uninterruptible power supply to ensure that the system can maintain normal operation for at least 30 minutes after a power outage, so as to prevent the use of power outages to evade fees.

[0058] In summary, the parking lot anti-following toll evasion control system based on multi-view vision and spacing detection according to embodiments of the present invention includes an exit channel detection unit, a payment status recognition unit, a following vehicle determination unit, a barrier gate enforcement control unit, and an alarm and recording unit. These units work collaboratively to form a complete closed loop for detecting and blocking following vehicles toll evasion. First, data from three types of sensors—inductive loops, millimeter-wave radar, and video cameras—are fused to achieve centimeter-level real-time detection of the position, speed, and spacing of multiple vehicles within the exit channel, eliminating blind spots and errors associated with single sensors. Next, a multi-condition joint judgment logic is established, comprehensively considering four dimensions: the payment status of the preceding vehicle, the barrier gate status, the distance to the following vehicle, and the payment record of the following vehicle. This effectively distinguishes between normal following and malicious following toll evasion, reducing the false judgment rate. Finally, upon confirming following toll evasion, the barrier gate enforcement control unit injects a high-priority forced closure signal into the barrier gate controller via a hardware interface, legally covering the anti-following toll evasion measures. The system triggers a sensor signal, causing the barrier to close precisely at a safe moment, cutting off the path for toll evasion. Finally, when a vehicle partially enters the barrier's coverage area, the barrier slowly descends at a low speed (30% to 50% of the standard speed), achieving a safe interception effect. Thus, through multi-condition joint judgment and active barrier control, the interception rate for toll evasion can be increased to over 95%, effectively curbing toll evasion. The safe deceleration strategy, while blocking toll-evading vehicles, avoids physical damage to the vehicles caused by the barrier, complying with parking lot safety operation standards. The multi-condition joint judgment logic integrates multiple dimensions of information such as payment status, distance, and speed, with a false judgment rate of less than 2% for normal following passage (such as two vehicles from the same user exiting together). It automatically records the license plate, time, and image evidence of toll-evading vehicles, supporting subsequent recovery. The barrier's forced control unit of this application integrates with existing barrier systems through a standard hardware interface, eliminating the need to replace the entire barrier equipment, resulting in low modification costs and strong adaptability.

[0059] To achieve the above embodiments, this invention proposes a parking lot anti-following payment evasion control method based on multi-view vision and distance detection, such as... Figure 5 As shown, the parking lot anti-following evasion control method based on multi-view vision and spacing detection includes the following steps: S101 performs OCR recognition on the acquired vehicle image to obtain the corresponding license plate information, and queries the payment status of the corresponding vehicle in the database of the toll management system based on the license plate information. If payment has been made, a lifting command is sent to the barrier gate controller, and the exit channel detection unit is activated to enter the working state.

[0060] S102, the exit channel detection unit obtains the position, speed and distance between adjacent vehicles of all vehicles within the channel range, and determines whether the following vehicle is following the vehicle to evade toll based on the distance between adjacent vehicles, relative speed, payment status of the preceding vehicle and payment record of the following vehicle, so as to obtain the result of the following vehicle to evade toll.

[0061] S103, when it is determined that there is a toll evasion behavior based on the toll evasion judgment result, a forced closing command with a priority higher than the anti-smashing signal is injected into the barrier gate controller to control the barrier gate to fall within a preset delay time after the preceding vehicle has completely passed through the area covered by the barrier arm. At the same time, the falling speed of the barrier gate is dynamically adjusted according to the distance between the following vehicle and the barrier arm during the falling process.

[0062] It should be noted that when a vehicle enters the exit lane, the license plate recognition module identifies the license plate and checks the payment status. Once payment is completed, the barrier gate opens normally, and the exit lane detection unit is activated to begin real-time monitoring of the position, speed, and spacing of all vehicles in the lane. As the vehicle in front passes through the barrier gate and begins to exit, the barrier gate remains open. The exit lane detection unit continuously monitors for following vehicles in the lane; if the distance between following vehicles is less than [a certain value], [the system will detect the following vehicle]. If the toll is not paid, the following vehicle detection unit triggers a toll evasion flag; the barrier gate control unit delays the activation of the following vehicle after the preceding vehicle has completely passed the gate. The system issues a forced closure command, overriding the anti-smashing function and controlling the gate to fall; the alarm and recording unit simultaneously triggers an audible and visual alarm, captures vehicle images, records toll evasion events, and uploads them to the cloud; after a following vehicle is blocked by the gate, the system guides it to make a payment, and after the payment is completed, the gate rises normally to allow passage.

[0063] As a specific example, such as Figure 6 As shown, the complete workflow includes the following steps: S301, the vehicle in front enters the exit lane, the license plate recognition module completes recognition within 200ms, checks the payment status and binds the vehicle ID; When the payment status of S302A is confirmed as "paid", the barrier gate will be raised normally to allow passage, and the system will activate the exit channel detection unit and the following vehicle determination unit.

[0064] When the payment status is confirmed as "unpaid" by the S302B, a payment prompt will be displayed, and the user will be directed to the payment terminal. S303: When the vehicle in front passes through the barrier gate, the inductive loop A1 detects that the front of the vehicle has passed the position of the barrier arm and generates a "pass confirmation" signal for the vehicle in front; the millimeter-wave radar continuously scans the number and spacing of vehicles in the channel.

[0065] S304, the following vehicle determination unit continuously monitors whether a vehicle has entered the second detection area B. If a following vehicle is detected, the following determination logic is immediately executed: Determination condition (a) the preceding vehicle has paid the toll and the barrier gate is in the raised state; Determination condition (b) the longitudinal distance D between the following vehicle and the preceding vehicle < (Default value 2.5m); Judgment condition (c) The following vehicle's speed v>0 and the following vehicle has no payment record in the toll collection system; If conditions (a), (b) and (c) are all met, the following vehicle toll evasion mark is set to "TRUE" and enters S305B.

[0066] S305A: If the following vehicle has already paid the toll, it will be allowed to pass normally, and the process will end.

[0067] S305B, the barrier gate forced control unit monitors the status of the ground induction coil C1. Once C1 confirms that the vehicle in front has completely left the passage (i.e., the barrier arm will not hit the vehicle in front when it falls), it immediately injects a high-priority forced closing signal into the barrier gate controller, overriding the anti-collision sensor signal and starting the barrier gate lowering procedure.

[0068] S306, the barrier gate begins to descend. During the descent, millimeter-wave radar continuously monitors the distance between the following vehicle and the end of the barrier arm. :like If the height exceeds 1.0m, the barrier gate will descend at a standard speed; if For vehicles with a height ≤1.0m and ≥0.3m, the barrier switch to deceleration mode (40% of standard speed), gently contacting / pressing the rear of the vehicle roof (not exceeding the rear bumper) to apply physical resistance rather than forceful impact, achieving a "light blocking" effect; if If the distance is less than 0.3m, the system determines that most of the vehicle has passed through, triggers an alarm and records the event, but no longer controls the barrier gate to prevent accidents.

[0069] The S307 alarm and recording unit simultaneously triggers an audible and visual alarm, and the camera continuously captures images of vehicles evading tolls (including clear license plates), with the data encrypted and uploaded to the cloud.

[0070] S308: When a vehicle following another vehicle is blocked in the exit lane by the barrier gate, the system guides the driver to make the payment through the payment terminal (display screen + QR code) next to the exit lane. After the payment is completed, the barrier gate will open normally to allow passage.

[0071] It should be noted that the system used in the parking lot anti-following-and-escape-payment control method based on multi-view vision and spacing detection in this embodiment is the aforementioned parking lot anti-following-and-escape-payment control system based on multi-view vision and spacing detection. Therefore, the explanation and description of the embodiment of the parking lot anti-following-and-escape-payment control system based on multi-view vision and spacing detection also applies to the parking lot anti-following-and-escape-payment control method based on multi-view vision and spacing detection in this embodiment, and will not be repeated here.

[0072] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied 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.

[0073] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0074] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0075] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0076] It should be noted that any reference signs placed between parentheses in the claims should not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention can be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0077] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0078] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

[0079] In the description of this invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0080] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0081] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0082] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms should not be construed as necessarily referring to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0083] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A parking lot anti-following payment evasion control system based on multi-view vision and spacing detection, characterized in that, include: An exit lane detection unit is installed within the parking lot exit lane area to obtain the position, speed, and distance between adjacent vehicles for all vehicles within the lane area. The payment status identification unit is connected to the parking fee management system and is used to obtain the payment status of vehicles entering the exit channel in real time and bind the vehicle payment status with the corresponding vehicle identifier. The following vehicle determination unit is connected to the exit channel detection unit and the payment status recognition unit respectively. It is used to determine whether the following vehicle is following the vehicle to evade toll based on the distance between adjacent vehicles, relative speed, payment status of the preceding vehicle and payment record of the following vehicle, so as to obtain the following vehicle toll evasion determination result. The barrier gate forced control unit is hardware connected to the following vehicle determination unit and the anti-smashing signal input terminal of the barrier gate controller. When it is determined that there is a following vehicle evasion behavior based on the following vehicle evasion determination result, the forced closing command with a priority higher than the anti-smashing signal is injected into the barrier gate controller. The barrier gate is controlled to fall within a preset delay time after the preceding vehicle has completely passed through the area covered by the barrier arm. At the same time, the falling speed of the barrier gate is dynamically adjusted according to the distance between the following vehicle and the barrier arm during the falling process.

2. The parking lot anti-following evasion control system based on multi-view vision and spacing detection as described in claim 1, characterized in that, Also includes: An alarm and recording unit is connected to the following vehicle determination unit. When the following vehicle evasion determination result determines that there is a following vehicle evasion behavior, the alarm and recording unit will simultaneously activate the sound and light alarm and capture multiple frames of image evidence of the evading vehicle. The license plate number, evasion time, lane number and image evidence will be encrypted, stored and uploaded to the cloud parking management platform.

3. The parking lot anti-following evasion control system based on multi-view vision and spacing detection as described in claim 1, characterized in that, The exit channel detection unit includes a first inductive loop, a second inductive loop, a third inductive loop, a binocular camera, a millimeter-wave radar sensor, and an edge computing module; The first inductive loop, the second inductive loop, and the third inductive loop correspond to the first detection area, the second detection area, and the third detection area, respectively. The first detection area is located directly below the barrier gate arm, the second detection area is located behind the first detection area, and the third detection area is located outside the barrier gate arm, so as to determine the detection area where the vehicle is located through the inductive loop signal. The binocular camera is installed at the top of the channel, and the millimeter-wave radar sensor is installed on the side wall of the channel. The deep neural network is used to perform real-time target detection on the camera image, identify the outline bounding box and centroid coordinates of all vehicles within the channel range, and calculate the distance and speed of adjacent vehicles by the difference of centroid coordinates between consecutive frames. The edge computing module is used to perform time synchronization and spatial registration of the position, speed and distance between adjacent vehicles of all vehicles within the collected channel range.

4. The parking lot anti-following payment evasion control system based on multi-view vision and spacing detection as described in claim 1, characterized in that, The payment status recognition unit includes a license plate recognition module, which is used to collect vehicle images and perform OCR recognition on the collected vehicle images to obtain the license plate number; and send the license plate number to the database of the toll management system to query the payment status of the corresponding vehicle, thus completing the binding between the license plate and the payment status.

5. The parking lot anti-following evasion control system based on multi-view vision and spacing detection as described in claim 1, characterized in that, The following vehicle determination unit has a built-in multi-condition joint determination model. The multi-condition joint determination model is used to determine the following vehicle as a toll evasion behavior when multiple preset conditions are met at the same time, otherwise it is determined as normal passage. The multiple preset conditions include the preceding vehicle having paid the toll and the barrier gate being raised, the longitudinal distance between the following vehicle and the preceding vehicle being less than a preset safety threshold, and the following vehicle having a speed greater than 0 and no corresponding toll payment record.

6. The parking lot anti-following payment evasion control system based on multi-view vision and spacing detection as described in claim 3, characterized in that, The gate control unit controls the gate to fall within a preset delay after the third ground induction coil in the third detection area confirms that the vehicle in front has completely left the passage, and injects a high-priority forced closing signal into the gate controller to cover the original anti-smashing sensor signal.

7. The parking lot anti-following evasion control system based on multi-view vision and distance detection as described in claim 1, characterized in that, During the lowering process of the barrier gate, the lowering speed is dynamically adjusted based on the distance between the following vehicle and the end of the barrier arm detected by the millimeter-wave radar sensor; Specifically, if the distance between the following vehicle and the end of the gate arm is greater than the first preset threshold, the gate will descend at the standard speed; if the distance between the following vehicle and the end of the gate arm is greater than or equal to the second preset threshold and less than or equal to the first preset threshold, the gate will descend at a low speed of 30% to 50% of the standard speed; if the distance between the following vehicle and the end of the gate arm is less than the second preset threshold, the descent will stop.

8. The parking lot anti-following evasion control system based on multi-view vision and spacing detection as described in claim 2, characterized in that, The alarm and recording unit includes an audible and visual alarm, a high-definition snapshot camera, and a data storage module. The sound and light alarm is installed within the exit passage area and is used to emit a dedicated toll evasion alarm sound and flashing light; the high-definition capture camera is aimed at the exit passage and automatically captures no less than 3 frames of close-up images of the vehicle's front and license plate after the toll evasion is detected; the data storage module is used to store the license plate number, toll evasion time, passage number and image evidence in an encrypted manner to the local storage device and upload them to the cloud parking management platform in real time.

9. The parking lot anti-following evasion control system based on multi-view vision and spacing detection as described in claim 2, characterized in that, It also includes a multi-vehicle scenario coordination module. When there are two or more vehicles in the exit channel at the same time, the multi-vehicle scenario coordination module establishes a tracking object for each vehicle independently, maintains the payment status, location and speed information of each vehicle, and independently executes the following vehicle toll evasion judgment logic for each pair of adjacent vehicles.

10. A parking lot anti-following payment evasion control method based on multi-view vision and distance detection, characterized in that, The parking lot anti-following evasion control system based on multi-view vision and spacing detection, applied to any one of claims 1-9, comprises the following steps: The acquired vehicle image is subjected to OCR recognition to obtain the corresponding license plate information. Based on the license plate information, the payment status of the corresponding vehicle is queried in the database of the toll management system. If payment has been made, a lifting command is sent to the gate controller, and the exit channel detection unit is activated to enter the working state. The exit lane detection unit acquires the position, speed and distance between adjacent vehicles for all vehicles within the lane range, and determines whether the following vehicle is engaging in toll evasion based on the distance between adjacent vehicles, relative speed, toll payment status of the preceding vehicle and toll payment record of the following vehicle, so as to obtain the toll evasion determination result. When it is determined that there is a toll evasion behavior based on the toll evasion judgment result, a forced closing command with a priority higher than the anti-smashing signal is injected into the gate controller to control the gate to fall within a preset delay time after the preceding vehicle has completely passed through the gate's coverage area. At the same time, the gate's falling speed is dynamically adjusted according to the distance between the following vehicle and the gate during the falling process.

Images