Barrier gate control method and device, management and control all-in-one machine and storage medium

Abstract

The embodiment of the invention provides a barrier gate control method and device, a management and control all-in-one machine and a storage medium, and relates to the technical field of video monitoring, the method is applied to a monocular camera arranged on the front face of the management and control all-in-one machine, and the management and control all-in-one machine further comprises a binocular camera arranged on the side face; the method comprises the following steps: acquiring each abnormal event reported by a binocular camera and the occurrence time of each abnormal event; determining an initial control mode about the barrier gate based on the license plate recognition result; if the initial control mode is to open a barrier gate, whether an abnormal event exists in a lane area indicated by the detection area currently or not is analyzed based on the occurrence time of the currently obtained abnormal event; if the abnormal event exists, correcting the initial control mode based on the current abnormal event to obtain a target control mode, and controlling the barrier gate based on the target control mode; and if not, controlling the barrier gate based on an initial control mode. According to the scheme, the intelligent degree of barrier gate control can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of video monitoring, and particularly to a method and device for controlling a barrier gate, a control and management integrated machine, and a storage medium. Background Art

[0002] In a vehicle access scenario with a barrier gate, it is necessary to capture images of vehicles entering or exiting and identify the license plates of the vehicles in order to control the barrier gate at the access.

[0003] In related technologies, usually a capture camera is set in the access scenario. The capture camera monitors the monitoring area in front of the barrier gate, captures images of the vehicles in the monitoring area, and identifies the license plates of the vehicles.

[0004] However, controlling the barrier gate only based on the license plate identified by the capture camera has insufficient intelligence in controlling the barrier gate. Summary of the Invention

[0005] The purpose of the embodiments of the present application is to provide a method and device for controlling a barrier gate, a control and management integrated machine, and a storage medium to improve the intelligence level of controlling the barrier gate. The specific technical solutions are as follows:

[0006] In a first aspect, the embodiments of the present application provide a method for controlling a barrier gate, which is applied to a monocular camera arranged on the front of a control and management integrated machine. The control and management integrated machine further includes a binocular camera arranged on the side. The monocular camera is used to capture images of vehicles that touch a preset trigger line for license plate recognition. In the acquisition screen of the binocular camera, a polygonal detection area is pre-calibrated. The detection area includes a part of the lane area in front of the barrier gate and a part of the lane area behind the barrier gate. The trigger line is located within the detection area and is not less than a predetermined distance from the designated edge of the detection area. The designated edge is the edge that the vehicle first touches when driving in. The method includes:

[0007] Obtain each abnormal event reported by the binocular camera and the occurrence time of each abnormal event. The binocular camera is used to monitor abnormal events in the lane area indicated by the detection area and report the monitored abnormal events;

[0008] In response to obtaining a license plate recognition result, based on the license plate recognition result, determine an initial control mode for the barrier gate;

[0009] If the initial control mode is to open the barrier gate, based on the occurrence time of the abnormal events obtained currently, analyze whether there are abnormal events in the lane area indicated by the detection area currently;

[0010] If an abnormal event currently exists, correct the initial control method based on the existing abnormal event to obtain a target control method, and control the barrier gate based on the target control method; and if not, control the barrier gate based on the initial control method;

[0011] wherein, the target control method is a control method for dealing with the existing abnormal event.

[0012] Optionally, analyzing whether there is currently an abnormal event in the lane area indicated by the detection area based on the occurrence time of the abnormal event currently obtained includes:

[0013] Determine whether there is an abnormal event in the currently obtained abnormal events whose time difference between the occurrence time and the current time is less than a predetermined time threshold; if so, determine that there is currently an abnormal event in the lane area indicated by the detection area; otherwise, determine that there is currently no abnormal event in the lane area indicated by the detection area.

[0014] Optionally, the abnormal event monitoring includes monitoring one or more of the following events:

[0015] Monitoring of lane state abnormal events, vehicle state abnormal events, and barrier gate state abnormal events;

[0016] wherein, the lane state abnormal event includes: a following-too-close event in the lane area in front of the barrier gate, and / or, a congestion event in the lane area entered after passing through the barrier gate;

[0017] The vehicle state abnormal event includes a false license plate event.

[0018] Optionally, correcting the initial control method based on the existing abnormal event to obtain a target control method includes:

[0019] In the case where the currently existing abnormal event is a congestion event, change the initial control method to a method for keeping the barrier gate closed and outputting a first alarm message through the management and control integrated machine to obtain a target control method; wherein, the first alarm message is a message for indicating that congestion has occurred and indicating to wait.

[0020] Optionally, the method further includes:

[0021] Obtain an elimination notice reported by the binocular camera for the monitored abnormal event;

[0022] When the currently existing abnormal event is a congestion event, after the step of controlling the barrier gate based on the target control method, if a cancellation notice of the currently existing abnormal event is obtained, the barrier gate is controlled to open for the vehicle indicated by the license plate recognition result.

[0023] Optionally, the modifying the initial control method based on the currently existing abnormal event to obtain a target control method includes:

[0024] When the currently existing abnormal event is a fake license plate event, changing the initial control method to a method for keeping the barrier gate closed and outputting a second alarm message through the management and control integrated machine, so as to obtain a target control method; wherein, the second alarm message is a message for indicating that the current license plate is a fake license plate.

[0025] Optionally, the modifying the initial control method based on the currently existing abnormal event to obtain a target control method includes:

[0026] When the currently existing abnormal event is a following-too-close event, changing the initial control method to a method for opening the barrier gate and extending the duration of the barrier gate in the open state, so as to obtain a target control method.

[0027] Optionally, the modifying the initial control method based on the currently existing abnormal event to obtain a target control method includes:

[0028] When the currently existing abnormal event is a barrier gate status abnormal event, changing the initial control method to a method for performing abnormal recovery processing on the barrier gate and opening the barrier gate after the abnormal recovery processing, so as to obtain a target control method.

[0029] Optionally, the identification method of the congestion event includes:

[0030] Monitoring whether the boundary of the congestion monitoring area in the detection area overlaps with the vehicle; wherein, the congestion monitoring area includes a partial lane area behind the barrier gate;

[0031] If there is an overlap, monitoring whether the congestion monitoring area meets the congestion condition, and if it meets the congestion condition, identifying that there is a congestion event; wherein, the congestion condition includes: there are adjacent vehicles with the number of spaced pixels less than the target pixel number; the target pixel number is the pixel number of a specified proportion of the pixel contour of the target vehicle, and the target vehicle is the vehicle closer to the barrier gate among the adjacent vehicles.

[0032] Optionally, the identification method of the barrier gate status abnormal event includes:

[0033] Based on a pre-trained gate state monitoring model, identify the current image containing the gate captured by the binocular camera to obtain an identification result indicating whether the gate is in an open state or a closed state;

[0034] According to the obtained identification result indicating whether the gate is in an open state or a closed state, and the target gate state corresponding to the current image, determine whether the gate is in an abnormal state. If so, it is identified that there is a gate state abnormal event; wherein, the target gate state corresponding to the current image is: the state that the gate should have when the current image represents that there is vehicle passage or no vehicle passage.

[0035] In a second aspect, an embodiment of the present application provides a gate control device, which is applied to a monocular camera disposed on the front of a control integrated machine. The control integrated machine further includes a binocular camera disposed on the side. The monocular camera is used to capture an image of a vehicle that touches a preset trigger line for license plate recognition; a polygonal detection area is pre-calibrated in the acquisition screen of the binocular camera. The detection area includes a part of the lane area in front of the gate and a part of the lane area behind the gate. The trigger line is located within the detection area and is not less than a predetermined distance from the specified edge of the detection area. The specified edge is the edge that the vehicle first touches when driving in; the device includes:

[0036] An acquisition module, configured to acquire each abnormal event reported by the binocular camera and the occurrence time of each abnormal event; the binocular camera is used to monitor abnormal events in the lane area indicated by the detection area and report the monitored abnormal events;

[0037] A determination module, configured to, in response to analyzing and obtaining a license plate recognition result, based on the license plate recognition result, determine an initial control mode for the gate;

[0038] An analysis module, configured to, if the initial control mode is to open the gate, based on the occurrence time of the abnormal events currently acquired, analyze whether there are currently abnormal events in the lane area indicated by the detection area;

[0039] A control module, configured to, if there are, based on the currently existing abnormal events, correct the initial control mode to obtain a target control mode, and based on the target control mode, control the gate; and, if not, control the gate based on the initial control mode;

[0040] Wherein, the target control mode is a control mode for coping with the existing abnormal events.

[0041] In a third aspect, an embodiment of the present application provides a control integrated machine. The control integrated machine includes a monocular camera disposed on the front and a binocular camera disposed on the side. The monocular camera is configured to capture an image of a vehicle that touches a preset trigger line for license plate recognition. In the acquisition screen of the binocular camera, a polygonal detection area is pre-calibrated. The detection area includes a partial lane area in front of the gate and a partial lane area behind the gate. The trigger line is located within the detection area and is not less than a predetermined distance from a designated edge of the detection area. The designated edge is the edge that is first touched when the vehicle enters. The monocular camera includes:

[0042] a memory for storing a computer program;

[0043] a processor configured to implement any of the gate control methods when executing the program stored on the memory.

[0044] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium. A computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, any of the gate control methods is implemented.

[0045] The embodiment of the present application further provides a computer program product containing instructions. When it runs on a computer, it causes the computer to execute any of the above-mentioned gate control methods.

[0046] Advantageous effects of the embodiment of the present application:

[0047] The gate control method provided by the embodiments of the present application is applied to a monocular camera arranged on the front of a control integrated machine. The control integrated machine further includes a binocular camera arranged on the side. The monocular camera is used to capture images of vehicles that touch a preset trigger line for license plate recognition. In the acquisition screen of the binocular camera, a polygonal detection area is pre-calibrated. The detection area includes a partial lane area in front of the gate and a partial lane area behind the gate. The trigger line is located within the detection area and is not less than a predetermined distance from the designated edge of the detection area. The designated edge is the edge that the vehicle first touches when entering. If the monocular camera analyzes and obtains a license plate recognition result, based on the license plate recognition result, an initial control mode for the gate is determined. When the initial control mode of the gate determined based on the license plate recognition result of the monocular camera is to open the gate, the present application does not directly open the gate. Instead, based on the occurrence time of the abnormal event reported by the binocular camera, it analyzes whether there is an abnormal event in the lane area indicated by the detection area currently. When there is an abnormal event, the initial control mode is corrected according to the existing abnormal event to cope with the existing abnormal event. If there is no abnormal event, the gate can be directly controlled based on the initial control mode. It can be seen that through the present application, the intelligent level of gate control can be improved. In addition, the binocular camera is used to monitor abnormal events in the lane area indicated by the detection area and report the monitored abnormal events. And through the setting of the lane area included in the detection area and the setting of the positional relationship with the trigger line, the vehicle entry process can be monitored completely, reducing the missed detection of abnormal events and further improving the intelligent level of gate control.

[0048] Of course, implementing any product or method of the present application does not necessarily require achieving all the above-mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other embodiments can also be obtained based on these drawings.

[0050] Figure 1 It is a schematic diagram of setting up a capture camera to monitor the entrance and exit scene in the prior art provided by the embodiments of the present application;

[0051] Figure 2 It is a schematic diagram of the acquisition screen of the capture camera in the prior art provided by the embodiments of the present application;

[0052] Figure 3 It is a schematic diagram of monitoring the entire lane area through the control integrated machine provided by the embodiments of the present application;

[0053] Figure 4 Schematic diagram of a control integrated machine provided by an embodiment of the present application;

[0054] Figure 5 Flow schematic diagram of a barrier control method provided by an embodiment of the present application;

[0055] Figure 6 Schematic diagram of a monitoring area for too-close following vehicles provided by an embodiment of the present application;

[0056] Figure 7 Schematic diagram of a monitoring line for configuring the barrier in a closed state provided by an embodiment of the present application;

[0057] Figure 8 Schematic diagram of a barrier control device provided by an embodiment of the present application;

[0058] Figure 9 Block diagram of a monocular camera in a control integrated machine provided by an embodiment of the present application. Detailed implementation manners

[0059] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art based on the present application belong to the scope of protection of the present application.

[0060] In a vehicle access scenario with a barrier, it is necessary to capture images of vehicles entering or exiting; as Figure 1 shown, currently, a capture camera 02 is usually installed in the access scenario. The capture camera 02 monitors a monitoring area 01 in front of the barrier 03 and performs license plate recognition. Among them, the monitoring area 01 is approximately 2 - 5 m in the longitudinal direction and approximately 4 m in the transverse direction.

[0061] Vehicles may generate abnormal events (such as: reversing, turning back, staying, etc.) in the lane area of the access scenario. As the captured image of the capture camera is shown in Figure 2 the figure, it can be seen that the field of view of the capture camera only includes a part of the lane area in front of the entrance (or exit) barrier and cannot cover the entire lane area; and the monocular camera can only monitor and identify vehicles, cannot monitor and analyze abnormal events in the entire lane area, cannot alarm for abnormal events, and only controls the barrier relying on the license plate recognized by the capture camera, which is obviously insufficient in the intelligent control of the barrier.

[0062] As Figure 3As shown in the figure, in this application, the entire lane area is monitored by the integrated control machine 300. The captured images of the binocular camera include part of the lane area in front of the gate and part of the lane area behind the gate in the entrance and exit scenarios ( Figure 3 Only the required field of view range of the entrance scenario is given exemplarily. The required field of view range of the exit scenario is similar to that of the entrance scenario).

[0063] Currently, it is usually necessary to additionally install a monitoring camera on the safety island in the entrance and exit scenarios to cover the lane area of the entrance and exit scenarios and monitor and analyze abnormal events in the lane area. However, manual assistance is still required for the monitoring and analysis of abnormal events. It can be seen that in the existing method, there are many devices installed on the safety island in the entrance and exit scenarios, and manual assistance is required, resulting in low monitoring efficiency and insufficient intelligence in the control of the gate.

[0064] Based on this, the embodiments of this application provide a gate control method, device, integrated control machine, and storage medium to improve the intelligence level of gate control.

[0065] First, a gate control method provided by this application will be introduced below.

[0066] Among them, a gate control method provided by the embodiments of this application can be applied to the monocular camera on the front of the integrated control machine. The integrated control machine is provided with a monocular camera on the front and a binocular camera on the side. That is, the integrated control machine can be an electronic device integrated with a monocular camera and a binocular camera for monitoring the entrance and exit scenarios. Compared with the situation where the monocular camera and the binocular camera are separated into two devices, it can save the configuration of wiring or wireless communication, etc. The monocular camera and the binocular camera in the integrated control machine can communicate quickly (the binocular camera can quickly report the monitored abnormal events to the monocular camera); as Figure 4 shown, the integrated control machine provided by the embodiments of this application includes a monocular camera A arranged on the front and a binocular camera B arranged on the side. The monocular camera is used to capture images for license plate recognition (its field of view range can include the monitoring area in front of the gate). The captured images of the binocular camera are pre-calibrated with a polygonal detection area. The detection area includes part of the lane area in front of the gate and the lane area behind the gate to monitor and report abnormal events in the lane area indicated by the entire detection area. Among them, the integrated control machine can also display prompt information such as "Welcome" and "6000 parking spaces remaining".

[0067] In addition, the entrance and exit scenarios involved in this application can be scenarios including one-way exits and / or entrances, or entrance and exit scenarios with two-way traffic (in the case of two-way traffic scenarios, a control integrated machine needs to be set in the exit direction and the entrance direction respectively; alternatively, the control integrated machine includes two monocular cameras and one binocular camera, and the field of view ranges of the two monocular cameras respectively include the monitoring areas in the exit direction and the entrance direction). This application does not make any limitations in this regard. The so-called monitoring area in front of the gate is the area where vehicle identification is required for the vehicle to pass through the gate when driving in the current traffic direction.

[0068] It should be noted that the binocular camera can monitor various abnormal events in the lane area indicated by the detection area. The abnormal events can include: reverse driving, turning back, staying, occupying the lane, false license plate, following too closely, congestion, abnormal gate state, etc. The binocular camera can also count the number of vehicles driving in the lane area indicated by the detection area, etc. This application does not make any limitations in this regard. Among various abnormal events, there may be some types of abnormal events that affect the control method of the monocular camera of the control integrated machine for the gate.

[0069] In addition, a gate control method provided by an embodiment of this application can be applied to any entrance and exit scenario with a gate, such as: parking lots, community entrances and exits, etc. This application does not make any limitations in this regard. And it should be noted that the above-mentioned monocular camera and binocular camera are arranged on different sides of the body of the control integrated machine: the monocular camera is located on the front and the binocular camera is located on the side; for the control integrated machine, the front of the body is the side that can face the vehicle before the vehicle drives to the gate; as Figure 4 shown, for the control integrated machine, the side that displays prompt information such as "Welcome" and "6000 parking spaces remaining" is the front of the control integrated machine.

[0070] The execution subject of a gate control method provided by an embodiment of this application can be a gate control device. The gate control device can be a functional software or functional module set in the monocular camera. The monocular camera can combine the monitoring results (abnormal events) of the binocular camera to control the gate.

[0071] A gate control method provided by an embodiment of this application is applied to a monocular camera arranged on the front of a control integrated machine. The control integrated machine further includes a binocular camera arranged on the side. The monocular camera is used to capture images of vehicles that touch a preset trigger line for license plate recognition; a detection area of a polygon is pre-calibrated in the acquisition screen of the binocular camera. The detection area includes a partial lane area in front of the gate and a partial lane area behind the gate. The trigger line is located within the detection area and is not less than a predetermined distance from the specified edge of the detection area. The specified edge is the edge that the vehicle first touches when driving in. The method includes:

[0072] Obtain each abnormal event reported by the binocular camera and the occurrence time of each abnormal event; the binocular camera is used to monitor abnormal events in the lane area indicated by the detection area and report the detected abnormal events.

[0073] In response to analyzing and obtaining a license plate recognition result, based on the license plate recognition result, determine an initial control method for the barrier gate.

[0074] If the initial control method is to open the barrier gate, based on the occurrence time of the abnormal events obtained currently, analyze whether there are currently any abnormal events in the lane area indicated by the detection area.

[0075] If there are any, based on the currently existing abnormal events, correct the initial control method to obtain a target control method, and control the barrier gate based on the target control method; and, if there are none, control the barrier gate based on the initial control method.

[0076] Wherein, the target control method is a control method for coping with the existing abnormal events.

[0077] The barrier gate control method provided by the embodiments of the present application is applied to a monocular camera arranged on the front of a management and control integrated machine. The management and control integrated machine further includes a binocular camera arranged on the side. The monocular camera is used to capture images of vehicles that touch a preset trigger line for license plate recognition; a detection area of a polygon is pre-calibrated in the acquisition screen of the binocular camera. The detection area includes a part of the lane area in front of the barrier gate and a part of the lane area behind the barrier gate. The trigger line is located within the detection area and is not less than a predetermined distance from the designated edge of the detection area. The designated edge is the edge that the vehicle first touches when entering. If the monocular camera analyzes and obtains a license plate recognition result, based on the license plate recognition result, determine an initial control method for the barrier gate. When the initial control method for the barrier gate determined based on the license plate recognition result of the monocular camera is to open the barrier gate, the present application does not directly open the barrier gate, but analyzes whether there are currently any abnormal events in the lane area indicated by the detection area based on the occurrence time of the abnormal events reported by the binocular camera, and when there are abnormal events, correct the initial control method according to the currently existing abnormal events to cope with the existing abnormal events; if there are no abnormal events, the barrier gate can be directly controlled based on the initial control method. It can be seen that through the present application, the intelligent degree of barrier gate control can be improved. In addition, the binocular camera is used to monitor abnormal events in the lane area indicated by the detection area and report the detected abnormal events, and through the setting of the lane area included in the detection area and the setting of the positional relationship with the trigger line, the vehicle entry process can be monitored completely, reducing the missed detection of abnormal events and further improving the intelligent degree of barrier gate control.

[0078] The following is an exemplary introduction to a barrier control method provided by an embodiment of the present application in conjunction with the accompanying drawings.

[0079] As Figure 5 shown, a barrier control method provided by an embodiment of the present application is applied to a monocular camera disposed on the front of a management and control integrated machine. The management and control integrated machine further includes a binocular camera disposed on the side. The monocular camera is used to capture images of vehicles that touch a preset trigger line for license plate recognition. A polygonal detection area is pre-calibrated in the acquisition screen of the binocular camera. The detection area includes a partial lane area in front of the barrier and a partial lane area behind the barrier. The trigger line is located within the detection area and is not less than a predetermined distance from the designated edge of the detection area. The designated edge is the edge that the vehicle first touches when entering. The method may include the following steps:

[0080] S501: Obtain each abnormal event reported by the binocular camera and the occurrence time of each abnormal event;

[0081] wherein, the binocular camera is used to monitor abnormal events in the lane area indicated by the detection area and report the monitored abnormal events;

[0082] In the present application, a polygonal detection area is pre-calibrated in the acquisition screen of the binocular camera disposed on the side of the management and control integrated machine. The detection area includes a partial lane area in front of the barrier and a partial lane area behind the barrier. The binocular camera can monitor abnormal events in the lane area indicated by the detection area and report the monitored abnormal events. Among them, the binocular camera can monitor abnormal events such as backing, turning back, staying, occupying the road, false license plate, following too closely, congestion, and abnormal barrier state. The specific monitoring method will be introduced in subsequent embodiments and will not be elaborated here.

[0083] It should be noted that the binocular camera can report the monitored abnormal events in real time. The management and control integrated machine can receive the monocular camera, which can obtain each abnormal event reported by the binocular camera and the occurrence time of each abnormal event, so that the monocular camera can control the barrier subsequently and improve the intelligent level of barrier control.

[0084] In addition, the polygon of the boundary of the detection area in the acquisition screen of the binocular camera is drawn according to the lane areas before and after the barrier. It can be a polygon detection area calibrated manually; or, after identifying the lane area and the barrier in the acquisition screen of the binocular camera and determining the lane areas before and after the barrier, it is automatically calibrated. The present application does not limit the calibration method and shape of the detection area.

[0085] It should be emphasized that in the acquisition screen of the binocular camera in this application, a polygonal detection area is pre-calibrated. The binocular camera can monitor and report abnormal events in the lane area indicated by the detection area. The detection area includes a part of the lane area in front of the gate and a part of the lane area behind the gate; in order to completely monitor the vehicle's entry process, the trigger line can be located within the detection area (the trigger line is located in a part of the lane area in front of the gate so that the vehicle can be identified before passing through the gate), and the distance between the trigger line and the designated edge that the vehicle first contacts when entering is not less than a predetermined distance (for example: the distance of a parking space), that is, the binocular camera can completely monitor the process of the vehicle moving from the designated edge to the trigger line. Without considering the complete monitoring of the vehicle's entry process, the trigger line can also coincide with the designated edge.

[0086] S502: In response to analyzing and obtaining the license plate recognition result, based on the license plate recognition result, determine the initial control mode for the gate;

[0087] A preset trigger line is marked within the field of view of the monocular camera. The monocular camera captures an image when it detects that the vehicle touches the trigger line (the so-called touch means that the vehicle touches the trigger line in the acquisition screen of the monocular camera, not necessarily in actual contact with the trigger line) for license plate recognition; after the monocular camera analyzes the captured image to obtain the license plate recognition result, it can determine the initial control mode for the gate based on the license plate recognition result.

[0088] Among them, in different entrance and exit scenarios, the initial control mode for the gate is different. For example: for the entrance and exit scenario of a parking lot, there is no need to pay attention to whether the vehicle represented by the license plate has the right to enter the parking lot, and only the license plate vehicle needs to be recognized to enter; that is, in the parking lot scenario, the initial control mode for the gate can be to open the gate. For the entrance and exit scenario of a community, it is necessary to pay attention to whether the vehicle represented by the license plate has the right to enter the community, and it is also necessary to recognize whether the license plate has the right to enter the community. If it has the right, the initial control mode for the gate can be to open the gate; otherwise, the initial control mode for the gate can be to keep the gate closed.

[0089] It should be noted that this application does not limit the execution order of steps S501 and S502. It is also possible to execute step S502 first, then execute step S501, or execute steps S501 and S502 simultaneously.

[0090] S503: If the initial control mode is to open the gate, based on the occurrence time of the abnormal event obtained currently, analyze whether there is an abnormal event in the lane area indicated by the detection area currently;

[0091] When the initial control mode is to open the gate, this application does not directly open the gate. Instead, based on the occurrence time of the abnormal events obtained currently, it analyzes whether there are currently abnormal events in the lane area indicated by the detection area, so as to correct the control mode of the gate when there are abnormal events and improve the intelligence level of the gate control.

[0092] Exemplarily, in one implementation, the analyzing whether there are currently abnormal events in the lane area indicated by the detection area based on the occurrence time of the abnormal events obtained currently includes:

[0093] Determine whether there is an abnormal event in the currently obtained abnormal events whose time difference between the occurrence time and the current time is less than a predetermined time threshold; if so, determine that there are currently abnormal events in the lane area indicated by the detection area; otherwise, determine that there are no currently abnormal events in the lane area indicated by the detection area.

[0094] The monocular camera can determine whether there is an abnormal event in the currently obtained abnormal events whose time difference between the occurrence time and the current time is less than a predetermined time threshold, so as to determine whether there are currently abnormal events in the lane area indicated by the detection area; wherein, the predetermined time threshold can be flexibly adjusted based on the actual scenario, for example: 1S; this application does not make a limitation on this.

[0095] It can be seen that in this application, the monocular camera can accurately analyze whether there are currently abnormal events in the lane area indicated by the detection area based on the occurrence time of the abnormal events reported by the binocular camera and in combination with the predetermined time threshold, so as to precisely control the gate and improve the intelligence level of the gate control.

[0096] S504: If so, based on the currently existing abnormal events, correct the initial control mode to obtain a target control mode, and control the gate based on the target control mode; and, if not, control the gate based on the initial control mode;

[0097] Wherein, the target control mode is a control mode for dealing with the existing abnormal events.

[0098] If it is recognized that there are currently abnormal events in the lane area indicated by the detection area, then it is necessary to control the gate for dealing with the abnormal events in response to the abnormal events; that is, this application can correct the initial control mode based on the currently existing abnormal events (the ways of correcting the initial control mode are different for different abnormal events) to obtain a target control mode for dealing with the existing abnormal events, and control the gate based on the target control mode; if there are no abnormal events, the gate can be directly controlled based on the initial control mode.

[0099] Exemplarily, in one implementation, if there is an abnormal event, the target control mode can be to keep the barrier gate closed, or perform abnormal recovery processing; that is, in the presence of an abnormal event, the barrier gate cannot be opened according to the initial control mode at this time, and the barrier gate can be kept closed, or abnormal recovery processing can be performed on the barrier gate. Among them, the specific implementation of obtaining the target control mode by correcting the initial control mode will be introduced in detail in the subsequent embodiments in combination with specific abnormal events, and will not be elaborated here.

[0100] It should be additionally noted that the initial control mode can also be to keep the barrier gate closed. Generally, when an abnormal event is recognized, such as abnormal events like fake license plates, following too closely, or congestion, at this time, the initial control mode is corrected, and the obtained target control mode is also to keep the barrier gate closed. The correction of the initial control mode when it is to keep the barrier gate closed in this application will not be elaborated.

[0101] The binocular camera can also identify and report the emergency events and the occurrence time of the emergency events in the lane area indicated by the detection area. If the initial control mode is to keep the barrier gate closed, when it is recognized based on the content reported by the binocular camera that there is an emergency event currently, the initial control mode can be corrected based on the current emergency event to obtain the specified control mode, and the barrier gate can be controlled based on the specified control mode. Exemplarily, the emergency event can be recognizing the presence of an emergency rescue vehicle, such as an ambulance, a fire truck, or a rescue and relief vehicle, etc. At this time, the initial control mode of keeping the barrier gate closed can be corrected to obtain the specified control mode of opening the barrier gate.

[0102] It can be seen that this application can identify abnormal events and emergency events in the entrance and exit scenarios, and control the barrier gate based on the abnormal events and emergency events to deal with abnormal events and emergency events, and improve the intelligent level of the control of the barrier gate.

[0103] In the technical solution of this application, operations such as obtaining, storing, using, processing, transmitting, providing, and disclosing various abnormal events, the occurrence time of abnormal events, license plate recognition results, etc. are all carried out under the premise of obtaining user authorization.

[0104] The gate control method provided by the embodiment of the present application is applied to a monocular camera arranged on the front of a management and control integrated machine. The management and control integrated machine further includes a binocular camera arranged on the side. The monocular camera is used to capture images of vehicles that touch a preset trigger line for license plate recognition. In the acquisition screen of the binocular camera, a polygonal detection area is pre-calibrated. The detection area includes a part of the lane area in front of the gate and a part of the lane area behind the gate. The trigger line is located within the detection area and is not less than a predetermined distance from the designated edge of the detection area. The designated edge is the edge that the vehicle first touches when driving in. If the monocular camera analyzes and obtains a license plate recognition result, based on the license plate recognition result, an initial control method for the gate is determined. When the initial control method for the gate determined based on the license plate recognition result of the monocular camera is to open the gate, the present application does not directly open the gate. Instead, based on the occurrence time of the abnormal event reported by the binocular camera, it analyzes whether there is an abnormal event in the lane area indicated by the detection area at present, and when there is an abnormal event, it corrects the initial control method according to the existing abnormal event to cope with the existing abnormal event. If there is no abnormal event, the gate can be directly controlled based on the initial control method. It can be seen that the present application can improve the intelligent level of gate control. In addition, the binocular camera is used to monitor abnormal events in the lane area indicated by the detection area and report the monitored abnormal events. And through the setting of the lane area included in the detection area and the setting of the positional relationship with the trigger line, the vehicle driving-in process can be completely monitored, reducing the missed detection of abnormal events and further improving the intelligent level of gate control.

[0105] Optionally, in another embodiment of the present application, the abnormal event monitoring includes the monitoring of one or more of the following events:

[0106] Monitoring of lane state abnormal events, monitoring of vehicle state abnormal events, and monitoring of gate state abnormal events;

[0107] Among them, the lane state abnormal events include: the event of following too closely in the lane area in front of the gate, and / or, the congestion event in the lane area entered after passing through the gate;

[0108] The vehicle state abnormal event includes the fake license plate event.

[0109] In the present application, the binocular camera can monitor and report abnormal events in the lane area indicated by the detection area. The abnormal event monitoring can include one or more of: monitoring of lane state abnormal events, monitoring of vehicle state abnormal events, and monitoring of gate state abnormal events. That is, the binocular camera can monitor events such as lane state abnormality, vehicle state abnormality, and gate state abnormality in the lane area indicated by the detection area.

[0110] The abnormal lane state may include a following-too-close event within the lane area in front of the barrier gate, and / or a congestion event within the lane area entered after passing through the barrier gate; the abnormal vehicle state includes a false license plate event, which can be understood as an event where there is a license plate but no real vehicle, such as: a license plate photographed by a mobile phone, a license plate of a toy car, a license plate printed on paper or acrylic board, a license plate held by a person, etc.

[0111] It should be noted that the abnormal events in this application may include abnormal events that affect the barrier gate control method and abnormal events that do not affect the barrier gate control method; the abnormal events shown above are abnormal events that affect the barrier gate control method. The binocular camera can also identify abnormal events such as turning back, reversing, staying, and occupying the road and report them, and the management and control integrated machine can give an alarm, etc.; and, the binocular camera can also count the vehicles passing through the barrier gate, etc., so that the management and control integrated machine can display the number of vehicles passing through the barrier gate (or display the remaining parking space number, etc.).

[0112] Exemplarily, in one implementation manner, the modifying the initial control method based on the currently existing abnormal events to obtain a target control method includes:

[0113] When the currently existing abnormal event is a congestion event, changing the initial control method to a method for keeping the barrier gate closed and outputting a first alarm message through the management and control integrated machine to obtain a target control method; wherein, the first alarm message is a message for indicating that congestion has occurred and indicating to wait.

[0114] When the currently existing abnormal event is a congestion event, that is, a congestion event within the lane area entered after passing through the barrier gate, the number of vehicles entering the barrier gate should be controlled. The initial control method can be changed to keep the barrier gate closed, and a first alarm message indicating that congestion has occurred and indicating to wait is output through the management and control integrated machine to obtain a target control method. It can be seen that this application can modify the initial control method for a congestion event, avoid vehicles from continuing to pass through the barrier gate, and give an alarm to cope with the congestion event.

[0115] Adaptively, the method further includes:

[0116] Obtaining an elimination notice reported by the binocular camera for the monitored abnormal event;

[0117] When the currently existing abnormal event is a congestion event, after the step of controlling the barrier gate based on the target control method, if an elimination notice of the currently existing abnormal event is obtained, then for the vehicle indicated by the license plate recognition result, control the barrier gate to open.

[0118] For a congestion event, the present application can also obtain an elimination notice reported by a binocular camera for the monitored congestion event, that is, after detecting a congestion event and controlling the barrier gate based on the target control mode of the congestion event, the binocular camera can detect the elimination of the congestion event and report an elimination notice for the congestion event; for the vehicle indicated by the license plate recognition result, if the vehicle can pass through the barrier gate (i.e., has the permission to pass through the barrier gate), the monocular camera can control the barrier gate to open so that the vehicle can pass through the barrier gate normally.

[0119] It can be seen that for a congestion event, the present application can intelligently control the barrier gate to eliminate the congestion event; and after the congestion event is eliminated, it can intelligently control the barrier gate to open so that the vehicle can pass through the barrier gate normally.

[0120] Exemplarily, in another implementation manner, the modifying the initial control mode based on the currently existing abnormal event to obtain a target control mode includes:

[0121] When the currently existing abnormal event is a false license plate event, changing the initial control mode to a mode for keeping the barrier gate closed and outputting a second alarm message through the management and control integrated machine to obtain a target control mode; wherein, the second alarm message is a message for indicating that the current license plate is a false license plate.

[0122] When the currently existing abnormal event is a false license plate event, that is, when the binocular camera does not detect the existence of a real vehicle but there is a license plate, and the license plate cannot enter the barrier gate, the initial control mode can be changed to keep the barrier gate closed and output a second alarm message indicating that the current license plate is a false license plate through the management and control integrated machine to obtain a target control mode. It can be seen that for a false license plate event, the present application can modify the initial control mode, prevent a vehicle with a false license plate from passing through the barrier gate, and give an alarm to cope with the false license plate event.

[0123] Exemplarily, in another implementation manner, the modifying the initial control mode based on the currently existing abnormal event to obtain a target control mode includes:

[0124] When the currently existing abnormal event is a following-vehicle-too-close event, changing the initial control mode to a mode for opening the barrier gate and extending the duration of the barrier gate in the open state to obtain a target control mode.

[0125] When the currently existing abnormal event is a following-too-closely event, that is, the vehicles in the lane area in front of the barrier are following too closely, the monocular camera may not be able to recognize the license plate of the following vehicle in the following-too-closely event. The binocular camera can supplement the license plate of the following vehicle. If the vehicle represented by the license plate of the following vehicle has the permission to pass through the barrier, the barrier can be kept continuously open; if the vehicle represented by the license plate of the following vehicle does not have the permission to pass through the barrier, due to the following-too-closely phenomenon, the following vehicle may follow the preceding vehicle through the barrier. To ensure safety, the barrier can be kept continuously open; that is, the initial control method can be changed to a method for opening the barrier and extending the duration of the barrier being in the open state, so as to obtain the target control method; that is, the target control method is to keep the barrier continuously open.

[0126] When the currently existing abnormal event is a following-too-closely event, the target control method may further include: outputting a third alarm message through the management and control integrated machine, and the third alarm message is used to indicate that there is a following-too-closely phenomenon currently, so as to remind adjacent vehicles to keep a safe distance. It can be seen that the present application can correct the initial control method for the following-too-closely event, ensure that the following vehicle can pass through the barrier safely, and give an alarm to cope with the following-too-closely event.

[0127] Exemplarily, in another implementation manner, the modifying the initial control method based on the currently existing abnormal event to obtain the target control method includes:

[0128] When the currently existing abnormal event is a barrier status abnormal event, changing the initial control method to a method for performing abnormal recovery processing on the barrier and opening the barrier after the abnormal recovery processing, so as to obtain the target control method.

[0129] When the currently existing event is a barrier status abnormal event and the barrier needs to be abnormally recovered before it can be opened, the initial control method can be changed to perform abnormal recovery processing on the barrier (for example: restart the barrier) and open the barrier after the abnormal recovery processing, so as to obtain the target control method. The present application can correct the initial control method for the barrier status abnormal event, perform abnormal recovery processing on the barrier, and open the barrier after the abnormal recovery processing, ensuring that the vehicle can pass through the barrier normally to cope with the barrier status abnormal event.

[0130] It can be seen that the present application can flexibly change the initial control method for following-too-closely events, congestion events, fake license plate events, and barrier status abnormal events to cope with the existing abnormal events and improve the intelligent level of barrier control.

[0131] Optionally, in another embodiment of the present application, the recognition method for the congestion event includes:

[0132] Monitor whether the boundary of the congestion monitoring area within the detection area overlaps with a vehicle; wherein, the congestion monitoring area includes a partial lane area after the gate.

[0133] If there is an overlap, monitor whether the congestion monitoring area meets the congestion condition. If the congestion condition is met, it is recognized that there is a congestion event; wherein, the congestion condition includes: there are adjacent vehicles with the number of pixels between them less than the target number of pixels; the target number of pixels is the number of pixels that is a specified proportion of the pixel contour of the target vehicle, and the target vehicle is the vehicle among the adjacent vehicles that is closer to the gate.

[0134] When the binocular camera recognizes a congestion event, it first monitors whether the boundary of the congestion monitoring area within the detection area overlaps with a vehicle; if there is an overlap, it indicates that there may be congestion in the congestion monitoring area, and further monitors whether the congestion monitoring area meets the congestion condition that there are adjacent vehicles with the number of pixels between them less than the target number of pixels. If so, it is recognized that there is a congestion event.

[0135] Wherein, the congestion monitoring area in this application includes a partial lane area that needs to be entered after passing through the gate; the congestion monitoring area of the binocular camera in this application can be pre-configured. For example: manually draw the boundary line of the congestion monitoring area, or, through the area recognition algorithm, recognize the partial lane area behind the gate of the binocular camera, and use the recognized calibration box as the boundary line of the congestion monitoring area; the congestion monitoring area can be flexibly configured according to different entrance and exit scenarios, and this application does not limit this.

[0136] Moreover, the target number of pixels in this application is the number of pixels that is a specified proportion of the pixel contour of the target vehicle. The specified proportion can be 30%. The target vehicle is the vehicle among the adjacent vehicles that is closer to the gate. Thus, for any pair of adjacent vehicles, using the number of pixels that is a specified proportion of the pixel contour of the vehicle among the adjacent vehicles that is closer to the gate as the standard for judging whether there is congestion, it is possible to accurately identify congestion for adjacent vehicles at different distances from the gate within the congestion monitoring area, that is, the adjacent vehicles closer to the gate are judged for congestion according to the pixel distance closer to the gate, and the adjacent vehicles farther away are judged for congestion according to the pixel distance farther away, and it is possible to accurately identify congestion events in the congestion monitoring area within the detection area.

[0137] Optionally, in another embodiment of this application, the recognition method of the gate status abnormal event includes:

[0138] Based on a pre-trained gate status monitoring model, recognize the current image including the gate currently collected by the binocular camera to obtain the recognition result that the gate is in the open state or the closed state.

[0139] Based on the recognition result of whether the barrier gate is in the open state or the closed state, and the target barrier gate state corresponding to the current image, determine whether the barrier gate is in an abnormal state. If so, it is recognized that there is an abnormal event of the barrier gate state; wherein, the target barrier gate state corresponding to the current image is: the state that the barrier gate should have when there is vehicle passage or no vehicle passage represented by the current image.

[0140] When the binocular camera recognizes an abnormal event of the barrier gate state, it first recognizes the current image containing the barrier gate collected by the binocular camera based on a pre-trained barrier gate state recognition model to obtain the recognition result of whether the barrier gate is in the open state or the closed state; then, based on the obtained recognition result and the state that the barrier gate should have when there is vehicle passage or no vehicle passage represented by the current image corresponding to the current image, determine whether the barrier gate is in an abnormal state. If so, that is, the barrier gate state is abnormal, it is recognized that there is an abnormal event of the barrier gate state.

[0141] This application can pre-train a barrier gate state monitoring model. Exemplarily, using the images collected by the binocular camera when there is vehicle passage and when there is no vehicle passage, configure a monitoring line for the barrier gate to be in the closed state (that is, a monitoring line representing the position of the barrier gate when the barrier gate is in the closed state), and use the images collected by the binocular camera when there is vehicle passage and when there is no vehicle passage with the configured monitoring line to train the barrier gate state monitoring model, and store the trained barrier gate state monitoring model locally in the binocular camera.

[0142] When the binocular camera recognizes an abnormal event of the barrier gate state, it uses a pre-trained barrier gate state monitoring model to recognize the state of the barrier gate in the current image to obtain the recognition result of whether the barrier gate is in the open state or the closed state; then, based on this recognition result and the target barrier gate state corresponding to the current image, determine whether the barrier gate is in an abnormal state. For example: the current image is an image without vehicle passage, and the result obtained based on the pre-trained barrier gate state monitoring model is that the barrier gate is in the open state, and the target barrier gate state corresponding to the current image is the closed state, then it is determined that the barrier gate is in an abnormal state (that is, the barrier gate is in the open state when there is no vehicle passage), and it is recognized that there is an abnormal event of the barrier gate state. Another example: the current image is an image with vehicle passage, and the initial control method determined by the monocular camera based on the license plate recognition result of the current image is to open the barrier gate (that is, this vehicle has the permission to pass through the barrier gate), and the result obtained based on the barrier gate state monitoring model is that the barrier gate is in the closed state (the results obtained from the images for a continuous period of time (such as 5s, etc.) are all that the barrier gate is in the closed state, that is, the barrier gate remains in the closed state for a continuous period of time); the target barrier gate state corresponding to the current image is the open state, then it is determined that the barrier gate is in an abnormal state (that is, when there is a vehicle that can pass waiting to pass through the barrier gate, the barrier gate remains in the closed state), and it is recognized that there is an abnormal event of the barrier gate state.

[0143] This application can identify the status of the barrier in the current image based on a pre-trained barrier status monitoring model. By comparing the identified result with the target barrier status corresponding to the current image, it can accurately identify whether there is an abnormal barrier status event currently. In the case of an abnormal barrier status event, it can perform accurate intelligent control on the barrier.

[0144] The following introduces the barrier control method provided by this application in combination with another embodiment.

[0145] This application provides an entrance and exit panoramic stitching and capturing device (i.e., a control and management integrated machine), which can realize vehicle capturing and recognition in the entrance and exit scene, monitor various abnormal events in the lane area, and report alarms; this control and management integrated machine is as Figure 4 shown. This device consists of three image sensors (a monocular camera and a binocular camera), including a monocular camera A set on the front of the device, which is used to capture passing vehicles, recognize license plates, control the barrier, etc. in the entrance and exit scene; a binocular camera B set on the side of the device, which is used to identify and alarm abnormal events in the lane area of the entrance and exit scene. The events that the binocular camera B can identify include: following too closely, congestion, occupying the road, turning back, abnormal barrier status, channel counting, fake license plates, etc.

[0146] Among them, for the event of following too closely, as Figure 6 shown, a polygonal area is defined as the following-too-closely monitoring area 610. Based on the object segmentation algorithm and the object monitoring algorithm (the specific algorithm can be similar to the existing technology and will not be elaborated here), object segmentation and monitoring are performed on the passing vehicles in the entrance and exit scene. If there are multiple vehicle objects in the following-too-closely monitoring area at the same time (for example: the following-too-closely monitoring area includes the front of the leading vehicle and the rear of the following vehicle), it is determined that there is an event of following too closely and reported. Among them, the control and management integrated machine 640 monitors the entire scene; the detection area 600 of the polygon in the acquisition picture of the binocular camera in this application is the dotted area as Figure 6 shown. The detection area 600 includes a part of the lane area in front of the barrier and a part of the lane area behind the barrier. The following-too-closely monitoring area 610 of the following-too-closely event belongs to the part of the lane area in front of the barrier. This following-too-closely monitoring area 610 is a polygonal area and can be configured according to actual needs. In the case where the following-too-closely monitoring area 610 is the part of the lane area in front of the barrier, the position of the following-too-closely monitoring area 610 in this application is not limited. And, as Figure 6 shown, the trigger line 630 can be located within the detection area 600, and the distance between it and the specified edge 620 that the vehicle first touches when entering is not less than the predetermined distance.

[0147] For a congestion event, a polygonal monitoring area is delimited as the congestion monitoring area. Based on the object segmentation algorithm, the passing vehicles in the entrance and exit scene are segmented and monitored. If there is an overlap between the left and right boundaries of the congestion monitoring area and any vehicle, and the number of pixels between adjacent vehicles within the congestion monitoring area is less than the target pixel number (the pixel number of 30% of the vehicle contour of the vehicle closer to the gate among the adjacent vehicles), it is determined that there is a congestion event in the entrance and exit scene.

[0148] For a U-turn event, a polygonal monitoring area is delimited as the U-turn monitoring area. Based on the object segmentation algorithm, the passing vehicles in the entrance and exit scene are segmented and monitored. If a vehicle leaves the U-turn monitoring area in the opposite direction after entering 50% of the U-turn monitoring area (this ratio can be set according to actual needs and is not limited), it is determined that there is a U-turn event.

[0149] For an occupation event, a polygonal monitoring area is delimited as the occupation monitoring area. Based on the object segmentation algorithm and the object classification algorithm, various objects are segmented and monitored. If the same object is in the occupation monitoring area for a long time (a continuous period of time, such as 10s, 1min, etc.), it is determined that the object has an occupation event; and based on the object classification algorithm, the object is classified into types such as motor vehicles, non-motor vehicles, and pedestrians, and an occupation alarm is output.

[0150] For the abnormal gate state event, the binocular camera locally trains the gate state monitoring model to generate the corresponding model file. Among them, as Figure 7 shown, for the scenario where the control integrated machine 700 detects the entire lane area, this application can use the images collected by the binocular camera when there is vehicle passing and when there is no vehicle passing to configure the monitoring line ( Figure 7 the black arrow in) when the gate 710 is in the closed state. Using the images collected by the binocular camera with the configured monitoring line when there is vehicle passing and when there is no vehicle passing, the gate state monitoring model is trained, and the trained gate state monitoring model is stored locally in the binocular camera. When identifying the abnormal gate state event, the pre-trained gate state monitoring model is used to identify the gate state of the current image to obtain the recognition result of whether the gate is in the open state or the closed state; then, based on this recognition result and the target gate state corresponding to the current image, it is determined whether the gate is in an abnormal state. Exemplarily, when there is vehicle passing (the vehicle has the permission to pass through the gate), the gate remains in the non-lifting state (i.e., the closed state) for a continuous period of time; or when there is no vehicle passing, the gate remains in the lifting state (i.e., the open state) for a continuous period of time, it is determined that there is an abnormal gate state event.

[0151] For channel counting, a polygonal monitoring area is defined as the vehicle counting area. Based on the object segmentation algorithm, vehicles are monitored and segmented, and the vehicle targets are continuously monitored. When a vehicle crosses the center line of the vehicle counting area, it is determined that the vehicle enters or leaves (it can be determined whether the vehicle enters or leaves according to the vehicle passing direction and the passing direction of the current lane).

[0152] For the event of a fake license plate, a polygonal monitoring area is defined as the fake license plate monitoring area. Based on the object segmentation algorithm, vehicles are monitored and segmented. When a vehicle is currently monitored to drive in, the monocular camera identifies the license plate, and the binocular camera conducts auxiliary monitoring regarding the fake license plate (or the binocular camera can independently monitor the event of a fake license plate). It is confirmed whether there is a real vehicle. When the binocular camera monitors that there is no real vehicle, it is characterized that the current capture of the monocular camera is caused by a fake license plate, and it is determined that there is an event of a fake license plate, and the monocular camera of the control integration machine is not allowed to open the gate, that is, the gate is kept in the closed state.

[0153] Among them, the above various monitoring areas can be adjusted according to the actual situation. The too-close following monitoring area belongs to a part of the lane area in front of the gate, the congestion monitoring area belongs to a part of the lane area behind the gate, the U-turn monitoring area belongs to a part of the lane area in front of the gate or a part of the lane area behind the gate, the occupancy monitoring area belongs to a part of the lane area in front of the gate or a part of the lane area behind the gate, the vehicle counting area belongs to a part of the lane area including the front of the gate, and the fake license plate monitoring area belongs to a part of the lane area in front of the gate. This application does not make any limitations in this regard.

[0154] The gate control method provided by the embodiment of the present application is applied to the monocular camera on the front of the control integration machine. The control integration machine further includes a binocular camera arranged on the side. In addition to the capture function, the control integration machine can also monitor abnormal events in the lane area of the entrance and exit scene and report them. The control integration machine can intervene immediately based on the abnormal events, improve the intelligent degree of the gate control, and improve the lane passing efficiency. Moreover, the control integration machine integrates multiple functions such as vehicle capture and event monitoring, reducing the redundant equipment installation on the safety island in the entrance and exit scene.

[0155] Based on the above method embodiment, the embodiment of the present application further provides a gate control device, which is applied to the monocular camera arranged on the front of the control integration machine. The control integration machine further includes a binocular camera arranged on the side. The monocular camera is used to capture images of vehicles that touch the preset trigger line for license plate recognition; a polygonal detection area is pre-calibrated in the captured image of the binocular camera. The detection area includes a part of the lane area in front of the gate and a part of the lane area behind the gate. The trigger line is located within the detection area and is not less than a predetermined distance from the specified edge of the detection area. The specified edge is the edge that the vehicle first touches when driving in; as Figure 8 shown, the device includes:

[0156] An acquisition module 810 is configured to acquire each abnormal event reported by the binocular camera and the occurrence time of each abnormal event; the binocular camera is configured to monitor abnormal events in the lane area indicated by the detection area and report the monitored abnormal events.

[0157] A determination module 820 is configured to, in response to obtaining a license plate recognition result through analysis, determine an initial control mode for the gate based on the license plate recognition result.

[0158] An analysis module 830 is configured to, if the initial control mode is to open the gate, analyze whether there is currently an abnormal event in the lane area indicated by the detection area based on the occurrence time of the abnormal event that has been currently acquired.

[0159] A control module 840 is configured to, if there is an abnormal event, correct the initial control mode based on the currently existing abnormal event to obtain a target control mode, and control the gate based on the target control mode; and if there is no abnormal event, control the gate based on the initial control mode.

[0160] Wherein, the target control mode is a control mode for coping with the existing abnormal event.

[0161] The gate control device provided by the embodiment of the present application is applied to a monocular camera arranged on the front of a management and control integrated machine. The management and control integrated machine further includes a binocular camera arranged on the side. The monocular camera is used to capture images of vehicles that touch a preset trigger line for license plate recognition. A polygonal detection area is pre-calibrated in the acquisition screen of the binocular camera. The detection area includes a part of the lane area in front of the gate and a part of the lane area behind the gate. The trigger line is located within the detection area and is not less than a predetermined distance from the designated edge of the detection area. The designated edge is the edge that the vehicle first touches when entering. If the monocular camera analyzes and obtains a license plate recognition result, based on the license plate recognition result, an initial control method for the gate is determined. When the initial control method for the gate determined based on the license plate recognition result of the monocular camera is to open the gate, the present application does not directly open the gate. Instead, based on the occurrence time of the abnormal event reported by the binocular camera, it analyzes whether there is an abnormal event in the lane area indicated by the detection area currently, and when there is an abnormal event, it corrects the initial control method according to the existing abnormal event to cope with the existing abnormal event. If there is no abnormal event, the gate can be directly controlled based on the initial control method. It can be seen that the present application can improve the intelligence level of gate control. In addition, the binocular camera is used to monitor abnormal events in the lane area indicated by the detection area and report the detected abnormal events. And through the setting of the lane area included in the detection area and the setting of the positional relationship with the trigger line, the vehicle entry process can be monitored completely, reducing the missed detection of abnormal events and further improving the intelligence level of gate control.

[0162] Optionally, the analysis module is specifically configured to:

[0163] Determine whether there is an abnormal event in the currently acquired abnormal events whose time difference between the occurrence time and the current time is less than a predetermined time threshold; if so, determine that there is an abnormal event in the lane area indicated by the detection area currently; otherwise, determine that there is no abnormal event in the lane area indicated by the detection area currently.

[0164] Optionally, the abnormal event monitoring includes the monitoring of one or more of the following events:

[0165] Monitoring of lane state abnormal events, vehicle state abnormal events, and gate state abnormal events;

[0166] Among them, the lane state abnormal event includes: an event of following too closely in the lane area in front of the gate, and / or, a congestion event in the lane area entered after passing through the gate;

[0167] The vehicle state abnormal event includes a fake license plate event.

[0168] Optionally, correcting the initial control mode based on the currently existing abnormal event to obtain a target control mode includes:

[0169] When the currently existing abnormal event is a congestion event, changing the initial control mode to a mode for keeping the gate closed and outputting a first alarm message through the management and control integrated machine to obtain a target control mode; wherein, the first alarm message is a message for indicating that congestion has occurred and indicating to wait.

[0170] Optionally, the device further includes a congestion event elimination module for:

[0171] Obtaining an elimination notice for the monitored abnormal event reported by the binocular camera;

[0172] When the currently existing abnormal event is a congestion event, after the step of controlling the gate based on the target control mode, if an elimination notice for the currently existing abnormal event is obtained, then for the vehicle indicated by the license plate recognition result, controlling the gate to open.

[0173] Optionally, correcting the initial control mode based on the currently existing abnormal event to obtain a target control mode includes:

[0174] When the currently existing abnormal event is a fake license plate event, changing the initial control mode to a mode for keeping the gate closed and outputting a second alarm message through the management and control integrated machine to obtain a target control mode; wherein, the second alarm message is a message for indicating that the current license plate is a fake license plate.

[0175] Optionally, correcting the initial control mode based on the currently existing abnormal event to obtain a target control mode includes:

[0176] When the currently existing abnormal event is a following-too-close event, changing the initial control mode to a mode for opening the gate and extending the duration of the gate in the open state to obtain a target control mode.

[0177] Optionally, correcting the initial control mode based on the currently existing abnormal event to obtain a target control mode includes:

[0178] When the currently existing abnormal event is a gate status abnormal event, changing the initial control mode to a mode for performing abnormal recovery processing on the gate and opening the gate after the abnormal recovery processing to obtain a target control mode.

[0179] Optionally, the identification method of the congestion event includes:

[0180] Monitor whether there is an overlap between the boundary of the congestion monitoring area within the detection area and the vehicle; wherein, the congestion monitoring area includes a partial lane area behind the barrier.

[0181] If there is an overlap, monitor whether the congestion monitoring area meets the congestion condition. If the congestion condition is met, it is recognized that there is a congestion event; wherein, the congestion condition includes: there are adjacent vehicles with the number of pixels between them less than the target number of pixels; the target number of pixels is the number of pixels that is a specified proportion of the pixel contour of the target vehicle, and the target vehicle is the vehicle closer to the barrier among the adjacent vehicles.

[0182] Optionally, the recognition method for the abnormal barrier state event includes:

[0183] Based on a pre-trained barrier state monitoring model, recognize the current image containing the barrier currently captured by the binocular camera to obtain the recognition result of whether the barrier is in the open state or the closed state.

[0184] According to the obtained recognition result of whether the barrier is in the open state or the closed state, and the target barrier state corresponding to the current image, determine whether the barrier is in an abnormal state. If so, it is recognized that there is an abnormal barrier state event; wherein, the target barrier state corresponding to the current image is: the state that the barrier should have when the current image represents that there is vehicle passage or there is no vehicle passage.

[0185] The embodiment of the present application further provides a control integrated machine. The control integrated machine includes a monocular camera arranged on the front and a binocular camera arranged on the side. The monocular camera is used to capture an image of a vehicle that touches a preset trigger line for license plate recognition; a detection area with a polygon pre-calibrated in the acquisition picture of the binocular camera. The detection area includes a partial lane area in front of the barrier and a partial lane area behind the barrier. The trigger line is located within the detection area and is not less than a predetermined distance from the specified edge of the detection area. The specified edge is the edge that the vehicle first touches when driving in; as Figure 9 shown, the monocular camera includes:

[0186] A memory 901 for storing a computer program.

[0187] A processor 902 for implementing any of the above-mentioned barrier control methods when executing the program stored on the memory 901.

[0188] And the above-mentioned control integrated machine may further include a communication bus and / or a communication interface. The processor 902, the communication interface, and the memory 901 complete mutual communication through the communication bus.

[0189] The communication bus mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience in representation, only a thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0190] The communication interface is used for communication between the above electronic device and other devices.

[0191] The memory can include a Random Access Memory (RAM), and can also include a Non-Volatile Memory (NVM), such as at least one disk memory. Optionally, the memory can also be at least one storage device located far from the aforementioned processor.

[0192] The above-mentioned processor can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0193] In another embodiment provided by this application, a computer-readable storage medium is also provided. A computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the steps of any of the above gate control methods are implemented.

[0194] In another embodiment provided by this application, a computer program product containing instructions is also provided. When it runs on a computer, the computer is made to execute any of the gate control methods in the above embodiments.

[0195] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from a website, computer, server, or data center to another website, computer, server, or data center in a wired manner (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that includes one or more integrated available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a solid-state disk (SSD), etc.

[0196] It should be noted that in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including", or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or device that includes a series of elements includes not only those elements but also other elements that are not explicitly listed, or also includes elements that are inherent to such process, method, article, or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article, or device that includes the element.

[0197] Each embodiment in this specification is described in a related manner. The same or similar parts between the embodiments can be referred to each other, and the differences between each embodiment and other embodiments are emphasized. In particular, for the device and control integrated machine embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0198] The above are only the preferred embodiments of the present application and are not intended to limit the protection scope of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application are all included in the protection scope of the present application.

Claims

1. A barrier gate control method, characterized in that, Applied to a monocular camera set on the front of a control integrated machine, the control integrated machine further includes a binocular camera set on the side. The monocular camera is used to capture images of vehicles that touch a preset trigger line for license plate recognition. In the acquisition screen of the binocular camera, a polygonal detection area is pre-calibrated. The detection area includes a partial lane area in front of the gate and a partial lane area behind the gate. The trigger line is located within the detection area and is not less than a predetermined distance from the designated edge of the detection area. The designated edge is the edge that the vehicle first touches when entering. The method includes: Obtain each abnormal event reported by the binocular camera and the occurrence time of each abnormal event. The binocular camera is used to monitor abnormal events in the lane area indicated by the detection area and report the monitored abnormal events. In response to analyzing and obtaining a license plate recognition result, based on the license plate recognition result, determine an initial control mode for the gate. If the initial control mode is to open the gate, based on the occurrence time of the abnormal events obtained currently, analyze whether there are currently any abnormal events in the lane area indicated by the detection area. If there are, based on the currently existing abnormal events, correct the initial control mode to obtain a target control mode, and control the gate based on the target control mode. And, if not, control the gate based on the initial control mode. Wherein, the target control mode is a control mode for dealing with the existing abnormal events.

2. The method according to claim 1, characterized in that The analyzing whether there are currently any abnormal events in the lane area indicated by the detection area based on the occurrence time of the abnormal events obtained currently includes: Determine whether there are abnormal events in the currently obtained abnormal events whose time difference from the current time is less than a predetermined time length threshold. If there are, determine that there are currently abnormal events in the lane area indicated by the detection area. Otherwise, determine that there are currently no abnormal events in the lane area indicated by the detection area.

3. The method according to claim 1, characterized in that, The abnormal event monitoring includes monitoring one or more of the following events: Monitoring of lane status abnormal events, monitoring of vehicle status abnormal events, and monitoring of gate status abnormal events. Among them, the lane status abnormal events include: a following-too-close event in the lane area in front of the gate, and / or, a congestion event in the lane area entered after passing through the gate. The vehicle status abnormal events include fake license plate events.

4. The method according to claim 3, characterized in that, The correcting the initial control mode to obtain a target control mode based on the currently existing abnormal events includes: In the case where the currently existing abnormal event is a congestion event, change the initial control mode to a mode for keeping the gate closed and outputting a first alarm message through the control integrated machine to obtain a target control mode. Wherein, the first alarm message is a message for indicating that congestion has occurred and indicating to wait.

5. The method according to claim 4, wherein The method further includes: Obtain an elimination notice for the monitored abnormal events reported by the binocular camera. When the currently existing abnormal event is a congestion event, after the step of controlling the barrier gate based on the target control method, if a cancellation notice of the currently existing abnormal event is obtained, the barrier gate is controlled to open for the vehicle indicated by the license plate recognition result.

6. The method according to claim 3, wherein The modification of the initial control method based on the currently existing abnormal event to obtain the target control method includes: When the currently existing abnormal event is a fake license plate event, the initial control method is changed to a method for keeping the barrier gate closed and outputting a second alarm message through the management and control integrated machine, so as to obtain the target control method; wherein, the second alarm message is a message for indicating that the current license plate is a fake license plate.

7. The method according to claim 3, characterized in that The modification of the initial control method based on the currently existing abnormal event to obtain the target control method includes: When the currently existing abnormal event is a following-too-close event, the initial control method is changed to a method for opening the barrier gate and extending the duration of the barrier gate in the open state, so as to obtain the target control method.

8. The method according to claim 3, wherein The modification of the initial control method based on the currently existing abnormal event to obtain the target control method includes: When the currently existing abnormal event is a barrier gate status abnormal event, the initial control method is changed to a method for performing abnormal recovery processing on the barrier gate and opening the barrier gate after the abnormal recovery processing, so as to obtain the target control method.

9. The method according to claim 3, wherein The identification method of the congestion event includes: Monitoring whether the boundary of the congestion monitoring area within the detection area overlaps with a vehicle; wherein, the congestion monitoring area includes a partial lane area behind the barrier gate; If there is an overlap, monitoring whether the congestion monitoring area meets the congestion condition, and if it meets the congestion condition, it is identified that there is a congestion event; wherein, the congestion condition includes: there are adjacent vehicles with the number of pixels between them less than the target number of pixels; the target number of pixels is the number of pixels of a specified proportion of the pixel contour of the target vehicle, and the target vehicle is the vehicle closer to the barrier gate among the adjacent vehicles.

10. The method according to claim 3, characterized in that, The identification method of the barrier gate status abnormal event includes: Based on a pre-trained barrier gate status monitoring model, identifying the currently captured current image containing the barrier gate by the binocular camera to obtain an identification result of whether the barrier gate is in the open state or the closed state; According to the obtained identification result of whether the barrier gate is in the open state or the closed state and the target barrier gate status corresponding to the current image, determining whether the barrier gate is in an abnormal state, and if so, identifying that there is a barrier gate status abnormal event; wherein, the target barrier gate status corresponding to the current image is: the status that the barrier gate should have when the current image represents that there is vehicle passage or no vehicle passage.

11. A barrier gate control device, characterized in that, Applied to a monocular camera set on the front of a control integrated machine, the control integrated machine further includes a binocular camera set on the side. The monocular camera is used to capture images of vehicles that touch a preset trigger line for license plate recognition. In the acquisition screen of the binocular camera, a polygonal detection area is pre-calibrated. The detection area includes a partial lane area in front of the gate and a partial lane area behind the gate. The trigger line is located within the detection area and is not less than a predetermined distance from the designated edge of the detection area. The designated edge is the edge that is first touched when the vehicle enters. The device includes: An acquisition module, configured to acquire each abnormal event reported by the binocular camera and the occurrence time of each abnormal event. The binocular camera is used to monitor abnormal events in the lane area indicated by the detection area and report the detected abnormal events. A determination module, configured to, in response to analyzing and obtaining a license plate recognition result, determine an initial control mode for the gate based on the license plate recognition result. An analysis module, configured to, if the initial control mode is to open the gate, analyze whether there is an abnormal event in the lane area indicated by the detection area based on the occurrence time of the abnormal events that have been acquired currently. A control module, configured to, if there is an abnormal event, correct the initial control mode based on the existing abnormal event to obtain a target control mode, and control the gate based on the target control mode. And, if there is no abnormal event, control the gate based on the initial control mode. Wherein, the target control mode is a control mode for coping with the existing abnormal event.

12. A control integrated machine, characterized in that, The control integrated machine includes a monocular camera set on the front and a binocular camera set on the side. The monocular camera is used to capture images of vehicles that touch a preset trigger line for license plate recognition. In the acquisition screen of the binocular camera, a polygonal detection area is pre-calibrated. The detection area includes a partial lane area in front of the gate and a partial lane area behind the gate. The trigger line is located within the detection area and is not less than a predetermined distance from the designated edge of the detection area. The designated edge is the edge that is first touched when the vehicle enters. The monocular camera includes: A memory, configured to store a computer program. A processor, configured to, when executing the program stored on the memory, implement the method according to any one of claims 1-10.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1-10 is implemented.

Images