Parking lot management method and system and computer readable storage medium
Through identification algorithms and automated management technology, the problem of inefficient management of traditional parking lots is solved, intelligent management and resource optimization of parking lots are realized, and service efficiency and accuracy are improved.
Patent Information
- Application Number
- CN202411563802.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-05
- Publication Date
- 2024-12-03
AI Technical Summary
Traditional parking lot management relies on a large number of manual operations, is inefficient and is easily affected by human factors, making it difficult to meet the needs of car owners for convenient and fast service, and lack of real-time data support, resulting in inaccurate utilization of parking resources and increasing the problem of parking difficulties.
By obtaining the current event information of the parking lot, determining the target recognition algorithm, identifying abnormal events and performing corresponding management actions, realizing automated management, reducing manual intervention, and improving management efficiency and accuracy.
It realizes intelligent management of parking lots, improves management efficiency and accuracy, reduces the workload of staff, enhances real-time monitoring and data support for parking lot events, and optimizes the utilization of parking space resources.
Smart Images

Figure CN119068425A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of automated management, and in particular, to a parking lot management method, system and computer-readable storage medium. Background Art
[0002] In traditional parking lot management, key links such as vehicle entry and exit, parking space allocation, and fee settlement mostly rely on manual operations. In traditional parking lot operations, manual command occupies a dominant position. From the guidance of vehicle entry, parking space allocation, to the confirmation of vehicle departure charges, a large amount of manual participation is required. This method is not only inefficient, but also prone to human factors leading to errors and delays. It is also difficult to meet the expectations of car owners for convenient and fast services. At the same time, due to the lack of real-time and comprehensive data support, it is difficult for parking lot managers to accurately allocate and optimize parking resources, further exacerbating the problem of parking difficulties. Summary of the invention
[0003] In view of this, an object of the embodiments of the present application is to provide a parking lot management method, system and computer-readable storage medium, which can improve management efficiency and accuracy.
[0004] In a first aspect, an embodiment of the present application provides a parking lot management method, comprising: obtaining current event information in a target area; wherein the target area is one or more monitoring areas in the parking lot; determining a target recognition algorithm based on the current event information; determining whether there is an abnormal event in the target area by recognizing the current event information through the target recognition algorithm; and in the event that the abnormal event exists in the target area, performing corresponding management actions based on the abnormal event.
[0005] In the above implementation process, by determining the corresponding target recognition algorithm based on the current event information obtained, and identifying the current event information based on the corresponding target recognition algorithm to determine whether there is an abnormal event in the target area, and then performing the corresponding management action, it is possible to automatically manage the target area according to the real-time event information of the target area, which can reduce the workload of parking lot staff, realize intelligent management of parking lots, and improve management efficiency and accuracy.
[0006] In one embodiment, determining a target recognition algorithm based on the current event information includes: determining an object type of a target object in the current event information; wherein the current event information includes one or more target items; the target items correspond to one or more object types; determining the number of objects corresponding to each object type in the current event; and determining the target recognition algorithm based on the object type and / or the number of objects.
[0007] In the above implementation process, by determining the object type and / or the number of objects according to actual conditions and then determining the corresponding target recognition algorithm, the efficiency and accuracy of the target recognition algorithm can be improved.
[0008] In one embodiment, the target item includes a person; determining the target recognition algorithm according to the object type includes: if the object type is a person, determining the target recognition algorithm to be a person wandering algorithm; identifying the current event information through the target recognition algorithm to determine whether there is an abnormal event in the target area includes: obtaining a first residence time of the person in the target area; when the first residence time exceeds a first set time threshold, determining that an abnormal event exists in the target area.
[0009] In the above implementation process, when the object type is identified as a person, the first residence time of the person in the target area is obtained, and then whether there is any abnormality in the person's stay in the target area is determined based on the person's residence time in the target area, thereby realizing automatic monitoring of abnormal conditions of people and improving the safety of people in the target area.
[0010] In one embodiment, the target item includes a vehicle; the target recognition algorithm is determined based on the object type and the object quantity, including: if the object type is a vehicle, and there is only one vehicle, then the target recognition algorithm is determined to be one or more of a return algorithm, a vehicle retention algorithm, a vehicle type recognition algorithm, or a non-motor vehicle recognition algorithm; if the object type is a vehicle, and there are multiple vehicles, then the target recognition algorithm is determined to be one or more of a vehicle retention algorithm, a lane congestion algorithm, a traffic statistics algorithm, a vehicle type recognition algorithm, a vehicle following algorithm, and a non-motor vehicle recognition algorithm.
[0011] In the above implementation process, when the target objects include vehicles, the corresponding target recognition algorithm is determined according to the vehicles and their corresponding quantities, thereby improving the accuracy of the target recognition algorithm.
[0012] In one embodiment, the identification of the current event information by the target recognition algorithm to determine whether there is an abnormal event in the target area includes: if the target recognition algorithm is a return algorithm, then when the departure data of the vehicle is obtained, determining whether there is a departing vehicle in the target area according to the current event information; when the departing vehicle exists in the target area, determining that there is an abnormal event in the target area; and / or if the target recognition algorithm is a vehicle retention algorithm, obtaining a second residence time of the vehicle in the target area; when the second residence time exceeds a second set time threshold, determining that there is an abnormal event in the target area; and / or if the target recognition algorithm is a vehicle type recognition algorithm, extracting vehicle features of the vehicle in the target area image; when the vehicle features do not meet the target features, determining that there is an abnormal event in the target area; and / or if the target recognition algorithm is a non-motor vehicle recognition algorithm, extracting the target area image. one or more of the vehicle model information, vehicle model information, license plate information, vehicle logo information and vehicle shape information of the vehicle in the target area image; when one or more of the vehicle model information, vehicle model information, license plate information, vehicle logo information and vehicle shape information is non-motor vehicle information, it is determined that there is an abnormal event in the target area; and / or if the target recognition algorithm is a lane congestion algorithm, the number of vehicles in the target area is obtained, and when the number of vehicles exceeds a number threshold, it is determined that there is an abnormal event in the target area; and / or if the target recognition algorithm is a traffic flow statistics algorithm, the number of vehicles in the target area within a period of time is obtained, and when the number of vehicles exceeds a traffic flow threshold, it is determined that there is an abnormal event in the target area; and / or if the target recognition algorithm is a vehicle following algorithm, the vehicle distance between adjacent vehicles is obtained, and when the vehicle distance exceeds a distance threshold, it is determined that there is an abnormal event in the target area.
[0013] In the above implementation process, when different target recognition algorithms are matched according to the object type and object type, the current actual situation of the vehicle is determined based on the different target recognition algorithms, automatic monitoring of various conditions of the vehicle is achieved, and the application scenarios of the target recognition algorithm are increased.
[0014] In one embodiment, the target object includes a barrier gate; determining the target recognition algorithm according to the object type includes: if the object type is a barrier gate, determining that the target recognition algorithm is a barrier gate state algorithm; identifying the current event information by the target recognition algorithm to determine whether there is an abnormal event in the target area includes: if the target recognition algorithm is a barrier gate state algorithm, obtaining the position information and shape information of the barrier gate; if there are no passing objects around the barrier gate and the duration of the barrier gate in a direction perpendicular to the ground exceeds a first time threshold, determining that the barrier gate is in a non-falling state; if there are passing objects around the barrier gate and the duration of the barrier gate in a direction parallel to the ground exceeds a second time threshold, determining that the barrier gate is in a non-opening state; if the difference between the shape information of the barrier gate and the target shape exceeds a difference range, determining that the barrier gate is in a damaged state; wherein the abnormal events include the non-falling state, the non-opening state and the damaged state.
[0015] In the above implementation process, when there is a barrier in the target area, the corresponding barrier status algorithm can also be matched. The current status of the barrier is determined by the barrier status algorithm, and then it is determined whether there is an abnormal situation in the target area. The objects monitored by the parking lot management method are increased, the application scenarios of the parking lot management method can be improved, and intelligent management of the parking lot can be realized.
[0016] In one embodiment, the current event information is a target area image; obtaining the current event information in the target area includes: monitoring the target area; when the target object appears in the target area, obtaining the width of the target object; when the width of the target object exceeds a width threshold range, obtaining the target area image.
[0017] In the above implementation process, by setting the target area image to be acquired only when the width of the target object exceeds the width threshold range, distant objects can be filtered out, the acquisition of invalid target area images can be reduced, and the accuracy of target area image acquisition can be improved.
[0018] In one embodiment, before obtaining the current event information in the target area, the method also includes: dividing the target area into one or more identification areas, and one or more shielding areas through detection lines; wherein the identification area is configured with one or more target recognition algorithms; obtaining the current event information in the target area includes: obtaining the current event information in the identification area; determining the target recognition algorithm based on the current event information.
[0019] In the above implementation process, by dividing the target area into the recognition area and the shielding area, when extracting the current event information, only the recognition area needs to be extracted, thereby reducing the amount of current event information obtained and improving processing efficiency.
[0020] In one embodiment, the target area includes one or more parking spaces, and the detection line is configured to mark one or more of the parking spaces as one or more parking areas; the identification area includes one or more of the parking areas; determining the target recognition algorithm according to the object type includes: if the object type is a vehicle, and the vehicle is parked in the parking area, determining the target recognition algorithm to be a cross-parking algorithm; identifying the current event information through the target recognition algorithm to determine whether there is an abnormal event in the target area includes: determining the positional relationship between the detection line and the vehicle; if at least one of the detection lines passes through the vehicle's positive projection area on the ground, determining that an abnormal event exists in the target area.
[0021] In the above implementation process, when the vehicle is located in the parking area, the straddle parking algorithm is used to determine whether the vehicle is straddling parking, so as to monitor irregular parking, realize automatic monitoring of the vehicle's parking situation, and increase the application scenarios of the target recognition algorithm.
[0022] In one embodiment, when the abnormal event exists in the target area, a corresponding management action is performed according to the abnormal event, including: determining a corresponding management action according to the abnormal type of the abnormal event; notifying a target person or handling the abnormal event through the management action; wherein the management action includes one or more of issuing an alarm, push notification, on-site command, abnormal marking, and data storage.
[0023] In the above implementation process, by determining the corresponding management action according to the abnormal type of the abnormal event, and then notifying the target personnel or handling the abnormal event through the management action, the parking lot can be managed automatically, the workload of the parking lot staff can be reduced, and the parking lot management efficiency can be improved.
[0024] In a second aspect, an embodiment of the present application further provides a parking lot management system, comprising: an image acquisition device and an electronic device; the image acquisition device is connected to the electronic device; the image acquisition device is used to obtain a target area image and send the target area image to the electronic device; the electronic device is used to execute the vehicle identification method in the above-mentioned first aspect, or any possible implementation of the first aspect.
[0025] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: a processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the machine-readable instructions are executed by the processor to perform the steps of the method in the above-mentioned first aspect, or any possible implementation of the first aspect.
[0026] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the parking lot management method in the above-mentioned first aspect, or any possible implementation of the first aspect are executed.
[0027] In order to make the above-mentioned objects, features and advantages of the present application more obvious and understandable, embodiments are given below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.
[0029] Figure 1 A schematic diagram of interaction with a parking lot management system provided in an embodiment of the present application; Figure 2 A block diagram of an electronic device provided in an embodiment of the present application; Figure 3 A flowchart of a parking lot management method provided in an embodiment of the present application; Figure 4 A schematic diagram of the functional modules of the parking lot management device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0030] The technical solutions in the embodiments of the present application will be described below in conjunction with the accompanying drawings in the embodiments of the present application.
[0031] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.
[0032] Urban parking entrance and exit management systems are widely used in commercial buildings, industrial parks, educational institutions, and residential communities. At present, traditional parking lot management is still mainly based on manual management. The management strategies of some parking lots that use automated management are mainly as follows: After the automated entrance and exit management system obtains input signals from the reading device, the controller compares and judges the input information based on the pre-set access permissions and other related information. When it meets the requirements, it records the access information (such as card number, location, time, entry or exit, etc.), and outputs signals to the actuator (such as the gate) to open and close it, and feeds back the working status to the controller, completing an operation.
[0033] The inventors of this application have discovered through a long period of research that the current parking lot entrance and exit management system in the industry uses a single decision-making process that is not deeply associated with the vehicles and traffic flow conditions in the lot, and is unable to form a unified control strategy for the entire area and each node when independently identifying and managing vehicles entering and leaving the lot.
[0034] In view of this, the present application proposes a parking lot management method, which determines a corresponding target recognition algorithm based on the current event information obtained, and identifies the current event information based on the corresponding target recognition algorithm to determine whether there is an abnormal event in the target area, and then performs a corresponding management action, thereby realizing automatic management of the target area according to the real-time event information of the target area, reducing the workload of parking lot staff, realizing intelligent management of parking lots, and improving management efficiency and accuracy.
[0035] To facilitate understanding of this embodiment, a parking lot management system that executes a parking lot management method disclosed in an embodiment of the present application is first introduced in detail.
[0036] like Figure 1 , which is a schematic diagram of the interaction of the parking lot management system provided by the embodiment of the present application, includes: an image acquisition device 200 and an electronic device 100. The image acquisition device 200 is connected to the electronic device 100.
[0037] The image acquisition device 200 is used to acquire the target area image and send the target area image to the electronic device 100. The image acquisition device 200 is also used to monitor the target area.
[0038] Optionally, the image acquisition device 200 may be configured to monitor the target area in real time, or to monitor the target area at preset time intervals. The way in which the image acquisition device 200 monitors the target area may be selected according to actual conditions.
[0039] The above-mentioned image acquisition device 200 can be a camera, a camera or other device, and the image acquisition device 200 can be selected according to actual conditions.
[0040] The electronic device 100 here may be a network server, a database server, a personal computer (PC), a tablet computer, a smart phone, a personal digital assistant (PDA), etc. The electronic device 100 may be selected according to actual conditions.
[0041] The electronic device 100 is used to execute the parking lot management method in the embodiment of the present application.
[0042] To facilitate understanding of this embodiment, the electronic device 100 that executes the parking lot management method disclosed in the embodiment of the present application is introduced in detail below.
[0043] like Figure 2 , which is a block diagram of an electronic device. The electronic device 100 may include a memory 111 and a processor 113. A person skilled in the art may understand that Figure 2 The structure shown is only for illustration and does not limit the structure of the electronic device 100. For example, the electronic device 100 may further include Figure 2 More or fewer components as shown, or with Figure 2 Different configurations are shown.
[0044] The memory 111 and the processor 113 are directly or indirectly electrically connected to each other to achieve data transmission or interaction. For example, these elements can be electrically connected to each other via one or more communication buses or signal lines. The processor 113 is used to execute the executable module stored in the memory.
[0045] The memory 111 may be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable read-only memory (EEPROM), etc. The memory 111 is used to store programs, and the processor 113 executes the program after receiving an execution instruction. The method executed by the electronic device 100 defined by the process disclosed in any embodiment of the present application can be applied to the processor 113, or implemented by the processor 113.
[0046] The processor 113 may be an integrated circuit chip with signal processing capability. The processor 113 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may 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 gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application may be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
[0047] The electronic device 100 in this embodiment can be used to execute each step in each method provided in the embodiments of the present application. The implementation process of the parking lot management method is described in detail below through several embodiments.
[0048] See also Figure 3 , is a flow chart of the parking lot management method provided by the embodiment of the present application. Figure 3 The specific process shown is described in detail.
[0049] Step S201, obtaining current event information in the target area.
[0050] The target area is one or more monitoring areas in the parking lot, such as parking spaces, parking lot entrances and exits, and passages in the parking lot. The monitoring area can be selected according to actual conditions.
[0051] The current event information here refers to the event information at the current moment.
[0052] Optionally, the event information may include: the object type of the target object, the residence time of the target object, the occupied space of the target object, etc. The event information may be selected according to actual conditions.
[0053] The target object may be a person, a vehicle, a cargo, etc. The target object may be adjusted according to the actual situation.
[0054] Optionally, the current event information can be acquired in real time, at preset time intervals, or when the target area changes. The time for acquiring the current event information can be selected according to actual conditions.
[0055] The above-mentioned current event information can be obtained through an image acquisition device or an electronic device.
[0056] Specifically, when monitoring the target area, the image acquisition device can directly obtain current event information in the target area and send the current event information to the electronic device.
[0057] The image acquisition device may also acquire a target area image of the target area and send the target area image to the electronic device, and the electronic device acquires current event information by analyzing the target area image.
[0058] The above-mentioned current event information acquisition method is only exemplary, and the current event information acquisition method can be selected according to actual conditions.
[0059] Step S202: determining a target recognition algorithm according to current event information.
[0060] The current event information may be a target area image.
[0061] The target recognition algorithm here may include a wandering person algorithm, a vehicle detention algorithm, a lane congestion algorithm, a traffic statistics algorithm, a vehicle type recognition algorithm, a vehicle following algorithm, a non-motor vehicle recognition algorithm, a gate status algorithm, etc. The target recognition algorithm can be adjusted according to actual conditions.
[0062] It is understandable that one or more target recognition algorithms may be set corresponding to each event information. After the current event information is determined, the corresponding target recognition algorithm may be matched.
[0063] Step S203, identifying the current event information through a target recognition algorithm to determine whether there is an abnormal event in the target area.
[0064] The target recognition algorithm here can be a preset image recognition algorithm and / or recognition model. After obtaining the current event information, the target recognition algorithm corresponding to the current event information is matched, and the current event information is recognized based on the target recognition algorithm, so as to determine whether there is an abnormal event in the target area.
[0065] Optionally, the abnormal event may include: wandering persons, turning back vehicles, stranded vehicles, non-motor vehicles, lane congestion, following vehicles, etc. The abnormal event may be selected according to actual conditions.
[0066] Step S204: when an abnormal event occurs in the target area, a corresponding management action is performed according to the abnormal event.
[0067] The management action configuration here is to remind on-site personnel to perform corresponding actions, or to notify management personnel to give instructions, or to mark event information of abnormal events, etc.
[0068] Optionally, the abnormal event may include issuing an alarm, on-site command, push notification, abnormal marking, and data storage, etc. The management action may be selected according to actual conditions.
[0069] In the above implementation process, by determining the corresponding target recognition algorithm based on the current event information obtained, and identifying the current event information based on the corresponding target recognition algorithm to determine whether there is an abnormal event in the target area, and then performing the corresponding management action, it is possible to automatically manage the target area according to the real-time event information of the target area, which can reduce the workload of parking lot staff, realize intelligent management of parking lots, and improve management efficiency and accuracy.
[0070] In a possible implementation, step S202 includes: determining the object type of the target object in the current event information; determining the number of objects corresponding to each object type in the current event; and determining a target recognition algorithm based on the object type and / or the object number.
[0071] The current event information includes one or more target objects, and the one or more target objects correspond to one or more object types.
[0072] The object type here can be a person, a car, a tricycle, a gate, a truck, cargo, etc. The object type can be selected according to actual conditions.
[0073] It should be understood that, when the current event information includes multiple object types, the object types may be classified, and the number of objects corresponding to each object type may be determined.
[0074] Optionally, the target recognition algorithm may be determined comprehensively based on the object type and the number of objects, or may be determined based only on the object type, or may be determined based only on the object type. The target recognition algorithm may be determined in a manner that is selected based on actual conditions.
[0075] For example, when the object type is a barrier gate, the target recognition algorithm corresponding to the barrier gate can be directly determined according to the object type.
[0076] When the object type is a vehicle, a corresponding target recognition algorithm may be determined according to the object type and the number of objects (ie, the number of vehicles).
[0077] When there are many types of objects, the corresponding target recognition algorithm can be determined by directly combining the total number of objects of multiple object types.
[0078] The above-mentioned determination method of the target recognition algorithm is only exemplary, and the determination method of the target recognition algorithm can be selected according to actual conditions.
[0079] In the above implementation process, by determining the object type and / or the number of objects according to actual conditions and then determining the corresponding target recognition algorithm, the efficiency and accuracy of the target recognition algorithm can be improved.
[0080] In a possible implementation, determining a target recognition algorithm according to the object type includes: if the object type is a person, determining the target recognition algorithm to be a person wandering algorithm.
[0081] The personnel wandering algorithm here is configured to determine whether there are people who stay abnormally in the target area, such as people wandering back and forth in the target area, people gathering in the target area, people staying in the lane, etc.
[0082] Understandably, the lanes, entrances and exits of parking lots usually only allow vehicles to stay or pass through. For these areas with special needs, people are usually not allowed to pass or stay. If people are identified in these areas, the target recognition algorithm can be determined as a wandering person algorithm, and the wandering person algorithm is used to determine whether there are people wandering, people staying, people gathering, etc. in the area.
[0083] The personnel wandering algorithm here can be implemented through algorithms such as density-based clustering algorithm, hidden Markov model, spatial association rule mining algorithm, pedestrian trajectory prediction algorithm, and trajectory generation algorithm based on location technology.
[0084] Among them, the density-based clustering algorithm refers to an algorithm that determines the gathering area of people based on the density of their trajectory points. For example, the hidden Markov model is a statistical model that can model the behavior of people as a series of hidden states and infer the real state sequence through the observed trajectory points. The spatial association rule mining algorithm is used to find the spatial association relationship in the trajectory of people. The pedestrian trajectory prediction algorithm refers to predicting a person's future trajectory by learning the implicit scene rules. The trajectory generation algorithm based on location technology refers to the deployment of technologies such as the global positioning system (GPS), wireless sensor network (WSN), and radio frequency identification (RFID) to obtain the location information of people in real time and generate mobile trajectories.
[0085] In one embodiment, step S203 includes: obtaining a first residence time of a person in a target area; and determining that an abnormal event exists in the target area when the first residence time exceeds a first set time threshold.
[0086] The first dwell time here refers to the time interval from when a person enters the target area to when he exits the target area, or from when he enters the target area to the current moment.
[0087] The first set time threshold is the preset maximum time that a person is allowed to stay in the target area. For example, if the target area is a passage, and under normal circumstances, the maximum time for a person to pass through the passage is A minutes, then A minutes can be determined as the first set time threshold. For another example, if the target area is the gate before and after the entrance and exit of a parking lot, and people are not allowed to pass through the area, then the first set time threshold can be determined to be 0. The first set time threshold can be set according to actual conditions, and this application does not make specific restrictions.
[0088] It should be understood that when the first stay time of a person in the target area exceeds the first set time threshold, it means that the person stays abnormally in the target area, and it can be determined that an abnormal event exists in the target area.
[0089] In one embodiment, when the first residence time exceeds the first set time threshold, the characteristics of the people in the target area may be further acquired, and the specific abnormal type of the people in the target area may be further determined based on the characteristics of the people.
[0090] Among them, personnel characteristics may include characteristic data such as the number of personnel, the movement trajectory of personnel, and the surrounding environment of personnel.
[0091] Exemplarily, when the first residence time exceeds the first set time threshold, the actual number of people in the target area can be further obtained, and whether there is a gathering of people in the target area can be determined based on the actual number of people and the preset number of people.
[0092] When the first stay time exceeds the first set time threshold, the movement trajectory of the person in the target area can be further obtained, and whether there is a person wandering in the target area can be determined according to the movement trajectory of the person in the target area.
[0093] When the first stay time exceeds the first set time threshold, the surrounding environment of the person can be further acquired. If the person stays within the set range distance of the vehicle, it can be determined that there is no abnormal event in the target area.
[0094] The above-mentioned method of determining the specific abnormal type of a person in the target area is only exemplary, and the method of determining the specific abnormal type of a person in the target area can be selected according to actual conditions.
[0095] In the above implementation process, when the object type is identified as a person, the first residence time of the person in the target area is obtained, and then whether there is any abnormality in the person's stay in the target area is determined based on the person's residence time in the target area, thereby realizing automatic monitoring of abnormal conditions of people and improving the safety of people in the target area.
[0096] In one possible implementation, a target recognition algorithm is determined based on the type and number of objects, including: if the object type is a vehicle and there is only one vehicle, then the target recognition algorithm is determined to be one or more of a return algorithm, a vehicle retention algorithm, a vehicle type recognition algorithm, or a non-motor vehicle recognition algorithm; if the object type is a vehicle and there are multiple vehicles, then the target recognition algorithm is determined to be one or more of a vehicle retention algorithm, a lane congestion algorithm, a traffic statistics algorithm, a vehicle type recognition algorithm, a vehicle following algorithm, and a non-motor vehicle recognition algorithm.
[0097] In one embodiment, when the object type is determined to be a vehicle, the number of vehicles in the target area may be further determined, and then a corresponding target recognition algorithm may be determined according to the number of vehicles.
[0098] Among them, the turnaround algorithm is used to determine whether the vehicle has turned back in the target area. The vehicle retention algorithm is used to determine whether the vehicle has stopped abnormally in the target area. The vehicle type recognition algorithm is used to determine the vehicle type (such as new energy vehicles, oil vehicles, etc.). The non-motor vehicle recognition algorithm is used to determine whether the vehicle is a non-motor vehicle (such as a tricycle, electric vehicle, etc.). The lane congestion algorithm is used to determine whether the lane is congested. The traffic flow statistics algorithm is used to determine the traffic flow in the target area. The vehicle following algorithm is used to determine whether the distance between two adjacent vehicles is too close.
[0099] It is understandable that in some parking lot charging systems, if the time interval between two vehicle information collections is within the set time interval, it is determined that the vehicle may have entered the parking lot by mistake, and measures such as not charging parking fees or only charging basic parking fees are taken. Therefore, in some paid parking lots, in order to evade parking fees, some vehicle owners often use the following means: after the vehicle passes through the gate and enters the parking lot area, the vehicle returns to the gate again, and after the information collection device at the gate collects the vehicle information, it returns to the parking lot to park. In this way, the information collection device at the gate collects vehicle information twice in a relatively short period of time, and the owner does not need to pay the parking fee or only needs to pay the basic parking fee, thereby achieving the purpose of evading fees. Through the folding algorithm, it can be determined whether the vehicle has evaded fees by turning back and forth.
[0100] When a vehicle stays in a target area such as a lane or around a gate for too long, it may affect normal traffic. The detention algorithm can determine whether the vehicle is in an abnormal detention state.
[0101] In some special parking lots, such as new energy parking lots (where each parking space may have a corresponding charging pile), in order to avoid gasoline vehicles occupying the charging pile position, the new energy parking lot usually does not allow gasoline vehicles to enter. In this scenario, the vehicle type recognition algorithm can be used to determine whether the vehicle in the target area is a new energy vehicle.
[0102] When the parking lot is a motor vehicle parking lot, in order to prevent non-motor vehicles from illegally occupying motor vehicle spaces, a non-motor vehicle recognition algorithm may be used to identify whether an object in the target area is a non-motor vehicle.
[0103] When the target area is a lane, around a gate, or other area where vehicles need to flow, if the target area is congested, it may cause a traffic jam, thereby affecting traffic. Therefore, when the target area includes multiple vehicles, the lane congestion algorithm can be used to determine whether the target area is congested.
[0104] For some smart parking lots, it may be necessary to count the parking space occupancy in real time and provide timely feedback on the parking space occupancy in the parking lot to avoid continuing to allow vehicles to enter the parking lot when there are no parking spaces, causing congestion in the parking lot and wasting the time of car owners. In this case, the parking space occupancy in the parking lot can be determined in real time through the traffic statistics algorithm.
[0105] In some paid parking lots, when a vehicle passes through a barrier, the information collection device around the barrier may collect the vehicle information for billing. However, in order to evade parking fees, some vehicle owners often use the following means: when the front vehicle drives around the barrier, the rear vehicle closely follows the front vehicle to avoid the information collection device from collecting vehicle information. Since the information collection device does not collect the vehicle information, it cannot be charged, and the owner achieves the purpose of evading fees. The vehicle following algorithm can be used to determine whether the vehicle is following the vehicle to evade fees.
[0106] In the above implementation process, when the target objects include vehicles, the corresponding target recognition algorithm is determined according to the vehicles and their corresponding quantities, thereby improving the accuracy of the target recognition algorithm.
[0107] In a possible implementation, the specific implementation methods of step S203 corresponding to different target recognition algorithms may be different. The following lists the specific implementation methods of step S203 corresponding to several target recognition algorithms: Method 1: If the target recognition algorithm is a return algorithm, then when the vehicle departure data is obtained, it is determined whether there is a vehicle leaving the target area based on the current event information; if there is a vehicle leaving the target area, it is determined that there is an abnormal event in the target area.
[0108] The departure data here can be obtained through the information collection device at the gate.
[0109] It should be understood that after the information collection device obtains the departure data of a vehicle, the departure data is fed back to the image collection device, and the image collection device obtains the target area image under the trigger of the departure data to determine whether there is a vehicle leaving the target area through the current event information in the target area image. If there is still a vehicle leaving the target area at this time, it means that the vehicle has not left the target area, and the vehicle has returned to the target area after the departure data is collected by the information collection device.
[0110] Among them, the current event information in the target area image is used to determine whether there is a vehicle leaving the target area, and the license plate number of the vehicle in the target area image can be identified to determine the vehicle leaving the target area based on the license plate number. Of course, the vehicle's individual characteristics in the target area image can also be identified (such as a personalized doll hanging in front of the vehicle, a personalized sticker attached to the vehicle body, etc.), and the vehicle leaving the target area can be determined based on the individual characteristics. The method for determining the vehicle leaving the target area can be selected according to the actual situation.
[0111] In one embodiment, the reentry algorithm can be implemented by a classification algorithm, a regression algorithm, etc.
[0112] The specific calculation formula of the regression algorithm here can be shown as follows: ;in, is the Euclidean distance, is the coordinate point at the next moment, is the coordinate point at the previous moment.
[0113] The specific calculation formula of the above regression algorithm can be shown as follows: ; in, is the real coordinate value of the vehicle, is the predicted coordinate value of the vehicle, is the number of observations, λ is the regularization parameter, is the regression coefficient of the independent variable. Here, λ can be set in advance based on experience or experiment. Determined through model training.
[0114] Method 2: If the target recognition algorithm is a vehicle retention algorithm, a second residence time of the vehicle in the target area is obtained; when the second residence time exceeds a second set time threshold, it is determined that an abnormal event exists in the target area.
[0115] The second dwell time here refers to the time interval from when the vehicle enters the target area to when it exits the target area, or from when it enters the target area to the current moment.
[0116] It can be understood that if the target area includes multiple vehicles, the second stay time of each vehicle in the target area can be obtained separately, and whether the second stay time of each vehicle exceeds the second set time threshold can be calculated separately. When the second stay time of any vehicle exceeds the second set time threshold, it is determined that an abnormal event exists in the target area.
[0117] Of course, it may also be determined that an abnormal event exists in the target area when the second stay time of a set number of vehicles exceeds the second set time threshold.
[0118] In one embodiment, the vehicle retention algorithm can be implemented by frame difference method, background difference method, data fusion algorithm, classification algorithm, clustering algorithm, etc.
[0119] Among them, the specific calculation formula of the inter-frame difference method can be shown as follows: ; in, for The pixel position of the vehicle in the image at the moment, for The pixel position of the vehicle in the image at the moment, is the pixel distance.
[0120] Method three: If the target recognition algorithm is a vehicle type recognition algorithm, the vehicle features of the vehicles in the target area image are extracted; when the vehicle features do not meet the target features, it is determined that an abnormal event exists in the target area.
[0121] The vehicle features here may include license plate, vehicle model, vehicle type, vehicle logo, vehicle shape and other features, and the vehicle features can be selected according to actual conditions.
[0122] The target feature may be a pre-set feature of vehicles allowed to enter the target area. For example, if the target area is a new energy parking lot, the target feature may be a feature of a new energy vehicle (eg, a green license plate, a new energy vehicle dealer's logo, etc.).
[0123] It should be understood that after the vehicle features of the vehicle in the target area image are extracted, the vehicle features are matched with the target features. If the vehicle features do not match the target features, the vehicle is determined to be a vehicle that is not allowed to enter the target area, and an abnormal event is determined to exist in the target area.
[0124] In one embodiment, the vehicle type recognition algorithm can be implemented by edge detection algorithms such as Canny and Sobel.
[0125] Among them, the specific implementation of the Canny edge detection algorithm is as follows: First, calculate the gradient of the target area image. The specific formula is as follows: ; ; in, is the gradient amplitude of the target area image, is the direction of the target area image, For icon area images The gradient of the direction, For icon area images Direction gradient.
[0126] Secondly, perform maximum value suppression. That is, for each pixel, compare the gradient values of the two adjacent pixels in the gradient direction, and only retain the pixel with the largest gradient value in the gradient direction, which is helpful to eliminate the blur effect on the edge. The specific processing formula of maximum value suppression is as follows: ; in, represents the gradient amplitude after non-maximum suppression, is the low threshold of gradient strength, for The offset value in the y direction is δy, and δy is the offset value in the y direction.
[0127] Furthermore, the pixels are subjected to dual threshold processing. That is, the pixels are divided into: strong edge, weak edge and non-edge. The formula for the dual threshold processing can be: ; in, is the gradient amplitude, Indicates the high threshold, Indicates the low threshold, Indicates a strong edge, Indicates a weak edge, Represents a non-edge.
[0128] Finally, the strong edges are connected with the surrounding weak edges to form a complete edge and obtain the vehicle feature.
[0129] Method 4: If the target recognition algorithm is a non-motor vehicle recognition algorithm, then one or more of the vehicle model information, vehicle make information, license plate information, vehicle logo information and vehicle shape information in the target area image is extracted; when one or more of the vehicle model information, vehicle make information, license plate information, vehicle logo information and vehicle shape information is non-motor vehicle information, it is determined that there is an abnormal event in the target area.
[0130] The non-motorized information here can be set in advance and stored in the memory. After extracting one or more of the vehicle model information, vehicle model information, license plate information, vehicle logo information and vehicle shape information of the vehicle in the target area image, one or more of the vehicle model information, vehicle model information, license plate information, vehicle logo information and vehicle shape information is matched with the non-motorized vehicle information. If the match is successful, the vehicle is determined to be a motor vehicle. If the match fails, the vehicle is determined to be a non-motorized vehicle.
[0131] In one embodiment, the non-motor vehicle recognition algorithm can be implemented by edge detection algorithms such as Canny, Sobel, etc. The non-motor vehicle recognition algorithm can be the same algorithm as the vehicle type recognition algorithm, or can be a different algorithm.
[0132] Method 5: If the target recognition algorithm is a lane congestion algorithm, the number of vehicles in the target area is obtained, and when the number of vehicles exceeds a number threshold, it is determined that an abnormal event exists in the target area.
[0133] The quantity threshold here may be a maximum number of vehicles allowed to be accommodated under normal traffic conditions in the target area, which is set in advance.
[0134] Optionally, the lane congestion algorithm may be an average speed determination method, a lane time occupancy algorithm, a congestion index method, etc. The lane congestion algorithm may be selected according to actual conditions.
[0135] Among them, the specific formula of the congestion index method is: congestion index = (actual vehicle driving time / ideal vehicle driving time) × 100%.
[0136] The average speed judgment method here refers to judging the traffic congestion by measuring the average speed of vehicles on the lane and comparing it with the congestion threshold. If the average speed is less than the congestion threshold, it is considered that there may be congestion in the lane.
[0137] The lane time occupancy algorithm mentioned above refers to the ratio of the time that there are cars on the road in a cycle to the cycle duration. By measuring the lane time occupancy and combining it with the average speed judgment, the traffic congestion situation can be evaluated more accurately. Specifically, the judgment criteria for "congestion" are: the average speed is less than the congestion threshold, and the lane time occupancy is greater than 50%.
[0138] Method 6: If the target recognition algorithm is a traffic flow counting algorithm, the number of vehicles in the target area within a period of time is obtained, and when the number of vehicles exceeds a traffic flow threshold, it is determined that an abnormal event exists in the target area.
[0139] Optionally, the period of time can be from the opening of the parking lot to the current time, from the early morning to the current time every day, from the last free parking space statistics time to the current time, etc. The period of time can be selected according to actual conditions.
[0140] The number of vehicles here can be determined by the number of vehicles entering the target area and the number of vehicles leaving the target area within a period of time.
[0141] The above-mentioned traffic threshold refers to the set number of vehicles allowed to enter the target area.
[0142] It should be understood that the number of parking spaces in each parking lot is limited, and the number of parking spaces in the parking lot limits the number of vehicles allowed to be parked in the parking lot. In this case, the number of vehicles in the target area can be monitored in real time. When the number of vehicles in the target area exceeds the flow threshold, it is determined that the number of vehicles parked in the target area has reached the maximum range allowed by the target area, and at this time, it is determined that there is an abnormality in the target area.
[0143] In one embodiment, the specific calculation formula of the traffic flow statistics algorithm may be: traffic flow = unit time * vehicle speed / (vehicle distance + vehicle body length).
[0144] Method 7: If the target recognition algorithm is a vehicle following algorithm, the vehicle distance between adjacent vehicles is obtained, and when the vehicle distance exceeds the distance threshold, it is determined that an abnormal event exists in the target area.
[0145] The vehicle distance here can be determined based on the distance between the identified rear end of the preceding vehicle and the identified front end of the following vehicle.
[0146] The above-mentioned distance threshold may be a pre-set minimum distance allowed between two vehicles in front and behind. The distance threshold may be a minimum safety distance, or may be a minimum distance allowed between two vehicles when the information collection device is collecting information normally. The distance threshold may be adjusted according to actual conditions.
[0147] It should be understood that when the vehicle distance between two adjacent vehicles exceeds the distance threshold, it can be determined that the distance between the two vehicles is less than the safe distance. If an emergency occurs, the two vehicles may be at risk of rear-end collision.
[0148] Or, when the distance between two adjacent vehicles exceeds the distance threshold, it can be determined that the distance between the two vehicles is less than the minimum distance when the information collection device performs normal information collection. In this case, the information collection device may not be able to collect vehicle information of the following vehicle, and thus cannot record the toll of the following vehicle. At this point, it can be determined that the following vehicle is following the vehicle to evade the toll.
[0149] In one embodiment, the specific calculation formula of the safety distance may be: safety distance = vehicle speed (km / h) ÷ A. Wherein A is a preset multiple. The preset multiple may be customized according to actual needs.
[0150] In the above implementation process, when different target recognition algorithms are matched according to the object type and object type, the current actual situation of the vehicle is determined based on the different target recognition algorithms, automatic monitoring of various conditions of the vehicle is achieved, and the application scenarios of the target recognition algorithm are increased.
[0151] In a possible implementation, determining a target recognition algorithm according to the object type includes: if the object type is a barrier, determining the target recognition algorithm to be a barrier state algorithm.
[0152] The gate status algorithm here is used to determine the current status of the gate.
[0153] Optionally, the current state of the barrier may include a barrier open, a barrier closed, a barrier not falling, a barrier not opening, a barrier damaged, etc. The current state of the barrier may be selected according to actual conditions.
[0154] The above-mentioned gate opening means that when a vehicle passes by, the gate is perpendicular to the ground; the gate closing means that when a vehicle leaves, the gate is parallel to the ground; the gate not falling means that after the vehicle leaves, the gate remains perpendicular to the ground for a certain period of time without falling; the gate not opening means that when a vehicle arrives around the gate, the gate remains parallel to the ground for a period of time; the gate being damaged means that the gate structure is damaged such as the gate being broken or deformed.
[0155] Among them, the above-mentioned gate opening and gate closing are the normal working states of the gate, and the gate not falling and the gate not opening are the abnormal working states of the gate.
[0156] In one embodiment, step S203 includes: if the target recognition algorithm is a barrier state algorithm, obtaining the position information and shape information of the barrier; if there are no passing objects around the barrier, and the duration of the barrier in a direction perpendicular to the ground exceeds a first time threshold, determining that the barrier is in a non-falling state; if there are passing objects around the barrier, and the duration of the barrier in a direction parallel to the ground exceeds a second time threshold, determining that the barrier is in a closed state; if the difference between the shape information of the barrier and the target shape exceeds a difference range, determining that the barrier is in a damaged state.
[0157] Optionally, the position information can be determined by obtaining the positions of the two end points of the barrier, or by obtaining the position of the overall structure of the barrier in the target area image. The method for obtaining the position information can be selected according to actual conditions.
[0158] The shape information here can be determined by the gate shape extracted from the target area image.
[0159] The above-mentioned target shape is the shape of the gate in an undamaged state.
[0160] It should be understood that after acquiring the target area image, if the target area image contains a barrier, the location information and shape information of the barrier can be extracted. At the same time, the barrier surrounding environment information (such as whether there are vehicles around the barrier) can also be extracted. The current state of the barrier can be determined based on the barrier surrounding environment information, the barrier location information and the barrier shape information, and then whether the barrier is abnormal can be determined based on the current state of the barrier.
[0161] It is understandable that if there are no passing objects around the barrier, the barrier should be in a closed state, that is, the barrier should be parallel to the ground. If it is detected that there are no objects around the barrier and the barrier is perpendicular to the ground, the barrier may be closed because the objects have just left and have not yet fallen, or the barrier may be in a non-falling state. At this time, the barrier can be continuously monitored to determine the duration of the barrier being perpendicular to the ground. If the duration of the barrier being perpendicular to the ground exceeds the first time threshold, it is determined that the barrier is in a non-falling state.
[0162] If there are passing objects around the barrier, the barrier should be in an open state, that is, the barrier should be perpendicular to the ground. If it is detected that there are passing objects around the barrier and the barrier is parallel to the ground, the barrier may not have been opened in time because the objects have just arrived, or the barrier may be closed. At this time, the barrier can be continuously monitored to determine the duration of the barrier being parallel to the ground. If the duration of the barrier being parallel to the ground exceeds the first time threshold, it is determined that the barrier is closed.
[0163] If there is a barrier gate in the target area image, the shape information of the barrier gate can be further obtained and compared with the target shape. If the difference between the shape information of the barrier gate and the target shape exceeds the difference range, it is determined that the barrier gate is in a damaged state.
[0164] Among them, abnormal events include the non-falling state, the non-opening state and the damaged state.
[0165] In one embodiment, the barrier gate status algorithm can be implemented by algorithms such as cosine similarity, Pearson correlation coefficient, NCC algorithm, and Hamming distance.
[0166] In the above implementation process, when there is a barrier in the target area, the corresponding barrier status algorithm can also be matched. The current status of the barrier is determined by the barrier status algorithm, and then it is determined whether there is an abnormal situation in the target area. The objects monitored by the parking lot management method are increased, the application scenarios of the parking lot management method can be improved, and intelligent management of the parking lot can be realized.
[0167] In a possible implementation, step S201 includes: monitoring a target area; when a target object appears in the target area, acquiring a width of the target object; and when the width of the target object exceeds a width threshold range, acquiring an image of the target area.
[0168] The target area here can be monitored in real time through monitoring devices.
[0169] It is understandable that, since the position of the monitoring device is fixed, the farther the target object is from the monitoring device, the smaller the target object in the acquired target area image. If the target object in the target area image is too small, on the one hand, it is difficult to obtain the current event information of the target object through the target area image, and on the other hand, there may be a problem that the target object has not yet moved into the target area. By setting the width threshold range, when it is determined that the width of the target object exceeds the width threshold range, the target area image is acquired again, so that a target area image of better quality can be acquired. In addition, it can also be ensured as much as possible that the target object in the target area image has moved into the target area.
[0170] The width of the target object mentioned above can be determined by the pixels in the target area image.
[0171] Optionally, the width threshold range may be the width range of the target object in the target area image when the target object is in the target area. It may also be the width range of the target object in the target image when the current event information of the target object can be accurately extracted from the target area image. The determination method of the width threshold range may be adjusted according to actual conditions.
[0172] The width threshold ranges corresponding to different target objects may be the same or different. The width threshold ranges of various target objects may be set independently. For example, the width threshold range corresponding to a person may be the first width range. The width threshold range corresponding to a car may be the second width range.
[0173] In the above implementation process, by setting the target area image to be acquired only when the width of the target object exceeds the width threshold range, distant objects can be filtered out, the acquisition of invalid target area images can be reduced, and the accuracy of target area image acquisition can be improved.
[0174] In a possible implementation, before step S201, the method further includes: dividing the target area into one or more identification areas and one or more shielding areas through detection lines.
[0175] The detection line here can be marked manually or automatically formed by the boundaries between the sub-regions in the target region. The determination of the detection line can be selected according to the actual situation.
[0176] The above-mentioned identification area refers to the area where the current event information needs to be identified. The shielding area refers to the area where the current event information does not need to be identified.
[0177] It is understandable that the viewing angle of the image acquisition device is usually fixed. When the image is acquired under the viewing angle of the image acquisition device, the acquired target area image may contain both areas that need to be monitored for abnormalities (such as lanes, entrances and exits, parking spaces) and areas that do not need to be monitored for abnormalities (such as corners and green belts). By further dividing the target area into a recognition area and a shielding area, when identifying the current event information, the current event information can be extracted only from the recognition area, reducing the scope of the current event information extraction.
[0178] Among them, each recognition area can be configured with one or more target recognition algorithms.
[0179] The target area may include one or more parking spaces. The detection line is configured to mark one or more parking spaces as one or more parking space areas.
[0180] The identification area may include one or more parking space areas.
[0181] In one embodiment, step S201 includes: obtaining current event information in the recognition area; and determining a target recognition algorithm according to the current event information.
[0182] In the above implementation process, by dividing the target area into the recognition area and the shielding area, when extracting the current event information, only the recognition area needs to be extracted, thereby reducing the amount of current event information obtained and improving processing efficiency.
[0183] In a possible implementation, determining a target recognition algorithm according to the object type includes: if the object type is a vehicle, and the vehicle is parked in a parking space area, determining the target recognition algorithm to be a cross-space parking algorithm.
[0184] The cross-parking algorithm here is used to detect whether the vehicle is parked in a cross-parking space.
[0185] In one embodiment, step S203 includes: determining the positional relationship between the detection line and the vehicle; if at least one detection line passes through the orthographic projection area of the vehicle on the ground, determining that an abnormal event exists in the target area.
[0186] The detection lines here are formed by the marking lines around the parking space.
[0187] It can be understood that if the vehicle is parked inside a parking space, the detection line corresponding to the parking space should be outside the vehicle and not intersecting with the vehicle. If the vehicle is parked between two parking spaces, the detection line between the two adjacent parking spaces intersects with the vehicle, which is reflected in the target area image, that is, the detection line passes through the vehicle's orthographic projection area on the ground. Therefore, by judging whether the vehicle is parked across spaces based on the positional relationship between the detection line and the vehicle, it can be determined whether the vehicle is parked across spaces.
[0188] In the above implementation process, when the vehicle is located in the parking area, the straddle parking algorithm is used to determine whether the vehicle is straddling parking, so as to monitor irregular parking, realize automatic monitoring of the vehicle's parking situation, and increase the application scenarios of the target recognition algorithm.
[0189] In a possible implementation, step S204 includes: determining a corresponding management action according to the abnormal type of the abnormal event; and notifying a target person or handling the abnormal event through the management action.
[0190] The management actions include one or more of issuing alarms, pushing notifications, on-site command, abnormal marking, and data storage.
[0191] The abnormal types here may include wandering people, stranded people, gathering of people, vehicle turning back, stranded vehicles, lane congestion, full parking spaces, non-permitted vehicles in the target area, gate abnormalities, cross-space parking, etc. The abnormal type can be adjusted according to actual conditions.
[0192] Each exception type may be set with a corresponding management action, and the management actions for different exception types may be the same or different.
[0193] For example, if the abnormality type is that the parking space is full, non-permitted vehicles appear in the target area, people are wandering, people are gathering, etc., the corresponding management action can be set to issue an alarm through an alarm device to disperse the wandering and gathering people, or drive away the vehicle.
[0194] If the abnormality type is stranded personnel, lane congestion, wandering personnel, stranded vehicles, etc., the corresponding management action can be set to on-site command to direct the movement of corresponding personnel or vehicles to maintain smooth traffic.
[0195] If the abnormality type is gate abnormality, cross-space parking, etc., the corresponding management action can be set as a push notification to notify the corresponding car owner or parking lot manager of the abnormality, and then notify the car owner to move the car, or notify the parking lot manager to repair the gate.
[0196] Optionally, multiple management actions can be set for each exception type.
[0197] For example, if the abnormal type is vehicle turning back, vehicle following, etc., the corresponding management actions can be set to issue an alarm, abnormal marking, and data storage. Among them, issuing an alarm is used to remind the owner to stop the irregular behavior in time. The abnormal marking is used to mark the vehicle as abnormal when the owner still maintains the irregular behavior after the alarm is issued, and then the vehicle is charged normally. In addition, the vehicle data of the vehicle with irregular behavior can also be stored to formulate corresponding management measures for the vehicle.
[0198] The above management actions are only exemplary and can be adjusted according to actual conditions.
[0199] In the above implementation process, by determining the corresponding management action according to the abnormal type of the abnormal event, and then notifying the target personnel or handling the abnormal event through the management action, the parking lot can be managed automatically, the workload of the parking lot staff can be reduced, and the parking lot management efficiency can be improved.
[0200] Based on the same application concept, a parking lot management device corresponding to the parking lot management method is also provided in the embodiment of the present application. Since the principle of solving the problem by the device in the embodiment of the present application is similar to that in the aforementioned parking lot management method embodiment, the implementation of the device in this embodiment can refer to the description in the embodiment of the above method, and the repeated parts will not be repeated.
[0201] See also Figure 4 , is a functional module diagram of a parking lot management device provided in an embodiment of the present application. Each module in the parking lot management device in this embodiment is used to execute each step in the above method embodiment. The parking lot management device includes an acquisition module 301, a first determination module 302, a second determination module 303, and an execution module 304; wherein, The acquisition module 301 is used to acquire current event information in a target area; wherein the target area is one or more monitoring areas in a parking lot.
[0202] The first determination module 302 is used to determine a target recognition algorithm according to the current event information.
[0203] The second determination module 303 is used to identify the current event information through the target recognition algorithm to determine whether there is an abnormal event in the target area.
[0204] The execution module 304 is used to execute corresponding management actions according to the abnormal event when the abnormal event exists in the target area.
[0205] In a possible implementation, the first determination module 302 is further used to: determine the object type of the target object in the current event information; wherein the current event information includes one or more target objects; the target objects correspond to one or more object types; determine the number of objects corresponding to each object type in the current event; and determine the target recognition algorithm based on the object type and / or the object number.
[0206] In a possible implementation manner, the first determination module 302 is specifically configured to: if the object type is a person, determine that the target recognition algorithm is a person wandering algorithm.
[0207] In a possible implementation manner, the second determination module 303 is specifically configured to: obtain a first residence time of the person in the target area; and determine that an abnormal event exists in the target area when the first residence time exceeds a first set time threshold.
[0208] In a possible implementation, the first determination module 302 is specifically used to: if the object type is a vehicle and there is only one vehicle, determine that the target recognition algorithm is one or more of a return algorithm, a vehicle retention algorithm, a vehicle type recognition algorithm or a non-motor vehicle recognition algorithm; if the object type is a vehicle and there are multiple vehicles, determine that the target recognition algorithm is one or more of a vehicle retention algorithm, a lane congestion algorithm, a traffic statistics algorithm, a vehicle type recognition algorithm, a vehicle following algorithm and a non-motor vehicle recognition algorithm.
[0209] In one possible implementation manner, the second determination module 303 is specifically used to: if the target recognition algorithm is a reentry algorithm, then when the departure data of the vehicle is obtained, determine whether there is a departing vehicle in the target area according to the current event information; when the departing vehicle exists in the target area, determine that there is an abnormal event in the target area; and / or if the target recognition algorithm is a vehicle retention algorithm, obtain a second residence time of the vehicle in the target area; when the second residence time exceeds a second set time threshold, determine that there is an abnormal event in the target area; and / or if the target recognition algorithm is a vehicle type recognition algorithm, extract vehicle features of the vehicle in the target area image; when the vehicle features do not meet the target features, determine that there is an abnormal event in the target area; and / or if the target recognition algorithm is a non-motor vehicle recognition algorithm, extract the vehicle type information of the vehicle in the target area image. one or more of the vehicle information, vehicle model information, license plate information, vehicle logo information and vehicle shape information; when one or more of the vehicle model information, vehicle model information, license plate information, vehicle logo information and vehicle shape information is non-motor vehicle information, determining that there is an abnormal event in the target area; and / or if the target recognition algorithm is a lane congestion algorithm, obtaining the number of vehicles in the target area, and when the number of vehicles exceeds a number threshold, determining that there is an abnormal event in the target area; and / or if the target recognition algorithm is a traffic flow statistics algorithm, obtaining the number of vehicles in the target area within a period of time, and when the number of vehicles exceeds a traffic flow threshold, determining that there is an abnormal event in the target area; and / or if the target recognition algorithm is a vehicle following algorithm, obtaining the vehicle distance between adjacent vehicles, and when the vehicle distance exceeds a distance threshold, determining that there is an abnormal event in the target area.
[0210] In a possible implementation manner, the first determination module 302 is specifically configured to: if the object type is a barrier gate, determine that the target recognition algorithm is a barrier gate state algorithm.
[0211] In a possible implementation, the second determination module 303 is specifically used to: if the target recognition algorithm is a barrier state algorithm, obtain the position information and shape information of the barrier; if there are no passing objects around the barrier and the duration of the barrier in a direction perpendicular to the ground exceeds a first time threshold, determine that the barrier is in a non-falling state; if there are passing objects around the barrier and the duration of the barrier in a direction parallel to the ground exceeds a second time threshold, determine that the barrier is in a non-opening state; if the difference between the shape information of the barrier and the target shape exceeds a difference range, determine that the barrier is in a damaged state; wherein the abnormal events include the non-falling state, the non-opening state and the damaged state.
[0212] In a possible implementation, the acquisition module 301 is further used to: monitor the target area; when the target object appears in the target area, acquire the width of the target object; when the width of the target object exceeds a width threshold range, acquire an image of the target area.
[0213] In a possible implementation, the parking lot management device further includes: a division module, configured to divide the target area into one or more identification areas and one or more shielding areas through detection lines; wherein the identification areas are configured with one or more target recognition algorithms.
[0214] In a possible implementation manner, the acquisition module 301 is specifically configured to: acquire current event information in the recognition area; and determine the target recognition algorithm according to the current event information.
[0215] In a possible implementation manner, the first determination module 302 is specifically configured to: if the object type is a vehicle, and the vehicle is parked in the parking area, determine that the target recognition algorithm is a cross-parking algorithm.
[0216] In a possible implementation, the second determination module 303 is specifically used to: determine the positional relationship between the detection line and the vehicle; if at least one of the detection lines passes through the vehicle's orthographic projection area on the ground, determine that an abnormal event exists in the target area.
[0217] In one possible implementation, the execution module 304 is specifically used to: determine a corresponding management action based on the abnormal type of the abnormal event; notify a target person or handle the abnormal event through the management action; wherein the management action includes one or more of issuing an alarm, push notification, on-site command, abnormal marking, and data storage.
[0218] In addition, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the parking lot management method described in the above method embodiment are executed.
[0219] The computer program product of the parking lot management method provided in the embodiment of the present application includes a computer-readable storage medium storing program code, and the instructions included in the program code can be used to execute the steps of the parking lot management method described in the above method embodiment. For details, please refer to the above method embodiment, which will not be repeated here.
[0220] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.
[0221] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
[0222] If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes. It should be noted that in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variation thereof are intended to cover non-exclusive inclusion, so that a process, method, article or apparatus that includes a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or apparatus. In the absence of more restrictions, the elements defined by the sentence "includes..." do not exclude the presence of other identical elements in the process, method, article or apparatus that includes the elements. The above description is only a preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application. It should be noted that similar numbers and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings.
[0223] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A parking lot management method, characterized in that: include: Acquire current event information in a target area; wherein the target area is one or more monitoring areas in a parking lot; Determining a target recognition algorithm according to the current event information; By identifying the current event information through the target recognition algorithm, it is determined whether there is an abnormal event in the target area; In the case where the abnormal event occurs in the target area, a corresponding management action is performed according to the abnormal event.
2. The method according to claim 1, characterized in that Determining a target recognition algorithm according to the current event information includes: Determine an object type of a target object in the current event information; wherein the current event information includes one or more target objects; and the target objects correspond to one or more object types; Determine the number of objects corresponding to each object type in the current event; The target recognition algorithm is determined according to the object type and / or the number of objects.
3. The method according to claim 2, characterized in that in, The target items include people; Determining the target recognition algorithm according to the object type includes: If the object type is a person, determining the target recognition algorithm to be a wandering person algorithm; The identifying of the current event information by the target recognition algorithm to determine whether there is an abnormal event in the target area includes: Obtaining a first stay time of the person in the target area; When the first residence time exceeds a first set time threshold, it is determined that an abnormal event exists in the target area.
4. The method according to claim 2, characterized in that: in, The target object includes a vehicle; Determining the target recognition algorithm according to the object type and the object quantity includes: If the object type is a vehicle, and there is one vehicle, determining that the target recognition algorithm is one or more algorithms selected from a return algorithm, a vehicle retention algorithm, a vehicle type recognition algorithm, or a non-motor vehicle recognition algorithm; If the object type is a vehicle, and there are multiple vehicles, the target recognition algorithm is determined to be one or more algorithms selected from the group consisting of a vehicle retention algorithm, a lane congestion algorithm, a traffic flow statistics algorithm, a vehicle type recognition algorithm, a vehicle following algorithm, and a non-motor vehicle recognition algorithm.
5. The method according to claim 4, characterized in that The identifying of the current event information by the target recognition algorithm to determine whether there is an abnormal event in the target area includes: If the target recognition algorithm is a return algorithm, then when the departure data of the vehicle is obtained, it is determined whether there is a departing vehicle in the target area according to the current event information; if there is the departing vehicle in the target area, it is determined that there is an abnormal event in the target area; and / or If the target recognition algorithm is a vehicle retention algorithm, a second retention time of the vehicle in the target area is obtained; when the second retention time exceeds a second set time threshold, it is determined that an abnormal event exists in the target area; and / or If the target recognition algorithm is a vehicle type recognition algorithm, extracting vehicle features of the vehicle in the target area image; if the vehicle features do not meet the target features, determining that an abnormal event exists in the target area; and / or If the target recognition algorithm is a non-motor vehicle recognition algorithm, one or more of the vehicle model information, vehicle type information, license plate information, vehicle logo information and vehicle shape information of the vehicle in the target area image is extracted; if one or more of the vehicle model information, vehicle type information, license plate information, vehicle logo information and vehicle shape information is non-motor vehicle information, it is determined that an abnormal event exists in the target area; and / or If the target recognition algorithm is a lane congestion algorithm, obtaining the number of vehicles in the target area, and determining that an abnormal event exists in the target area when the number of vehicles exceeds a number threshold; and / or If the target recognition algorithm is a traffic flow statistics algorithm, the number of vehicles in the target area within a period of time is obtained, and when the number of vehicles exceeds a traffic flow threshold, it is determined that an abnormal event exists in the target area; and / or If the target recognition algorithm is a vehicle following algorithm, the vehicle distance between adjacent vehicles is obtained, and when the vehicle distance exceeds a distance threshold, it is determined that an abnormal event exists in the target area.
6. The method according to claim 2, characterized in that in, The target object includes a barrier gate; Determining the target recognition algorithm according to the object type includes: If the object type is a barrier gate, determining that the target recognition algorithm is a barrier gate state algorithm; The identifying of the current event information by the target recognition algorithm to determine whether there is an abnormal event in the target area includes: If the target recognition algorithm is a barrier state algorithm, obtaining the position information and shape information of the barrier; If there is no passing object around the barrier, and the duration of the barrier in the direction perpendicular to the ground exceeds a first time threshold, it is determined that the barrier is in a non-falling state; If there are passing objects around the barrier, and the duration of the barrier in a direction parallel to the ground exceeds a second time threshold, it is determined that the barrier is in a closed state; If the difference between the shape information of the barrier and the target shape exceeds the difference range, it is determined that the barrier is in a damaged state; The abnormal events include the non-falling state, the non-opening state and the damaged state.
7. The method according to claim 2, characterized in that in, The current event information is a target area image; The obtaining of current event information in the target area includes: monitoring the target area; When the target object appears in the target area, obtaining the width of the target object; When the width of the target object exceeds a width threshold range, a target area image is acquired.
8. The method according to claim 1, characterized in that: Before acquiring the current event information in the target area, the method further includes: Dividing the target area into one or more identification areas and one or more shielding areas through detection lines; wherein the identification areas are configured with one or more target identification algorithms; The obtaining of current event information in the target area includes: Acquire current event information in the identified area; The target recognition algorithm is determined according to the current event information.
9. The method according to claim 8, characterized in that in, The target area includes one or more parking spaces, and the detection line is configured to mark one or more of the parking spaces as one or more parking space areas; The identification area includes one or more parking space areas; Determining the target recognition algorithm according to the object type includes: If the object type is a vehicle, and the vehicle is parked in the parking area, determining that the target recognition algorithm is a cross-parking algorithm; The identifying of the current event information by the target recognition algorithm to determine whether there is an abnormal event in the target area includes: Determining a positional relationship between the detection line and the vehicle; If at least one of the detection lines passes through the orthographic projection area of the vehicle on the ground, it is determined that an abnormal event exists in the target area.
10. The method according to claim 1, characterized in that When the abnormal event occurs in the target area, performing corresponding management actions according to the abnormal event includes: Determine a corresponding management action according to the abnormal type of the abnormal event; Notify the target person or handle the abnormal event through the management action; The management action includes one or more of issuing an alarm, pushing a notification, on-site command, abnormal marking, and data storage.
11. A parking lot management system, characterized in that: include: Image acquisition device and electronic device; the image acquisition device is connected to the electronic device; The image acquisition device is used to acquire a target area image and send the target area image to the electronic device; The electronic device is used to execute the parking lot management method described in any one of claims 1 to 10.
12. 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 steps of the method according to any one of claims 1 to 10 are executed.
Citation Information
Patent Citations
Image-based parking lot order management method and system
CN108717521A
Parking space management method and device and computer storage medium
CN111784857A
Vehicle following behavior judgment method and system based on parking lot barrier gate state
CN116168339A
Parking lot management method and device, computer equipment and storage medium
CN117315564A
Method and device for scheduling parking space in parking lot
WO2018232950A1
Cited By
Early warning method and system for abnormal state of bus stop parking area and medium
CN120319031A
A method, system and medium for warning abnormal status of bus stop parking area
CN120319031B
Target management method, management system, electronic device and computer program product
CN121505856A
