CN116994227B

CN116994227B - Parking state detection method, device, equipment and medium

Info

Publication number: CN116994227B
Application number: CN202311265702.8A
Authority: CN
Inventors: 陈佳骏; 邵旭昂; 陈圣军; 黄顺利; 徐昊; 盛雪
Original assignee: Shenzhen Zhongzhi Chelian Science And Technology Co ltd
Current assignee: Shenzhen Zhongzhi Chelian Science And Technology Co ltd
Priority date: 2023-09-28
Filing date: 2023-09-28
Publication date: 2024-05-28
Anticipated expiration: 2043-09-28
Also published as: CN116994227A

Abstract

The application relates to the technical field of parking state identification, in particular to a parking state detection method, a device, equipment and a medium, wherein the method comprises the following steps: acquiring a vehicle type of a target vehicle and a parkable vehicle type identifier of a parking space of the target vehicle; judging whether the vehicle type is consistent with the vehicle type corresponding to the parkable vehicle type identifier; if yes, acquiring first point cloud data corresponding to each key edge part of the target vehicle and a plurality of second point cloud data corresponding to a parking space of the target vehicle; determining a parking state of the target vehicle based on the first point cloud data and the second point cloud data corresponding to each key edge part, wherein the parking state comprises: normal parking and non-normal parking. The method and the device effectively improve the accuracy of detecting the parking state of the automobile.

Description

Parking state detection method, device, equipment and medium

Technical Field

The present application relates to the field of parking status recognition, and in particular, to a method, apparatus, device, and medium for detecting a parking status.

Background

In order to ensure the passing of vehicles, the vehicle parking device better meets the increasing parking requirements of people and solves the problem of difficult parking.

Generally, vehicles with different sizes need to be parked in different parking spaces, so that in the related art, whether the target vehicle enters the parking space with the corresponding size is judged, when the target vehicle enters the parking space with the corresponding size, the target vehicle is determined to be in normal parking, and when the target vehicle occupies the parking space corresponding to the vehicle with other sizes, the target vehicle is determined to be in non-normal parking. However, even if the target vehicle enters a parking space of a corresponding size, there may still be a problem of irregular parking.

Disclosure of Invention

In order to accurately identify the parking state of a vehicle, the application provides a parking state detection method, a device, equipment and a medium.

In a first aspect, the present application provides a method for detecting a parking status, which adopts the following technical scheme:

a parking status detection method, comprising:

Acquiring a vehicle type of a target vehicle and a parkable vehicle type identifier of a parking space of the target vehicle;

Judging whether the vehicle type is consistent with the vehicle type corresponding to the parkable vehicle type identifier;

If yes, acquiring first point cloud data corresponding to each key edge part of the target vehicle and a plurality of second point cloud data corresponding to a parking space of the target vehicle;

Determining a parking state of the target vehicle based on the first point cloud data corresponding to each key edge part and each second point cloud data, wherein the parking state comprises: normal parking and non-normal parking.

In one possible implementation manner, before determining the parking state of the target vehicle based on the first point cloud data corresponding to each critical edge component and each second point cloud data, the method further includes:

acquiring an environment image in a preset range corresponding to a parking space of the target vehicle;

Identifying the environment image and obtaining an obstacle identification result in a preset range of a parking space of the target vehicle;

if the obstacle recognition result is that the obstacle exists, acquiring third point cloud data corresponding to the obstacle;

correspondingly, the determining the parking state of the target vehicle based on the first point cloud data corresponding to each key edge part and each second point cloud data includes:

And determining the parking state of the target vehicle based on the first point cloud data, the second point cloud data and the third point cloud data corresponding to the key edge components.

In one possible implementation, the critical edge components include at least: rearview mirrors and other edge components, wherein determining the parking state of the target vehicle based on the first point cloud data, the second point cloud data and the third point cloud data corresponding to the key edge components comprises:

determining a first distance between each rearview mirror and the obstacle based on the third point cloud data and the first point cloud data corresponding to each rearview mirror;

Determining a collision probability corresponding to a first target distance based on a preset corresponding relation between the distance and the collision probability and the first target distance, wherein the first target distance is the distance between a rearview mirror on the same side as the side where the obstacle is located and the obstacle;

Judging whether the collision probability is smaller than a preset collision probability threshold value or not, and obtaining a first judgment result;

Determining a second distance between the target rearview mirror and a target boundary line of a parking space where the target vehicle is located based on the first point cloud data and the second point cloud data of the target rearview mirror, wherein the target rearview mirror is a rearview mirror on the opposite side of the side where the obstacle is located, and the target boundary line is a parking space boundary line on the same side as the target rearview mirror;

judging whether the second distance is not larger than a preset second distance threshold value or not to obtain a second judgment result;

And determining the parking state of the target vehicle according to the first judging result and the second judging result.

In one possible implementation, the critical edge component further comprises: and determining, by the bumper and the tail light, a parking state of the target vehicle based on the first point cloud data corresponding to each of the key edge components and the second point cloud data if the obstacle recognition result indicates that no obstacle exists, including:

Determining a third distance between the rearview mirror and a first parking space line based on first point cloud data of the rearview mirror and second point cloud data corresponding to the first parking space line, wherein the first parking space line is a parking space line positioned on the side of the rearview mirror;

Determining a fourth distance between the bumper and a second parking space line based on the first point cloud data of the bumper and second point cloud data corresponding to the second parking space line, wherein the second parking space line is a parking space line positioned on the bumper side;

Determining a fifth distance between the tail lamp and a third parking space line based on the first point cloud data of the tail lamp and second point cloud data corresponding to the third parking space line, wherein the third parking space line is a parking space line positioned at the side of the tail lamp;

And determining the parking state of the target vehicle according to the third distance, the fourth distance and the fifth distance.

In one possible implementation manner, before the determining that the parking state of the target vehicle is the normal parking, the method further includes:

obtaining parking direction mark information of a parking space where the target vehicle is located;

acquiring the orientation information of the target vehicle;

Obtaining a third judgment result according to the orientation information and the parking direction indicator information, wherein the third judgment result represents whether the parking orientation of the target vehicle is compliant or not;

Correspondingly, the determining the parking state of the target vehicle according to the first judgment result and the second judgment result includes:

And determining the parking state of the target vehicle according to the first judging result, the second judging result and the third judging result.

In one possible implementation, obtaining a vehicle type of a target vehicle includes:

acquiring a vehicle image of the target vehicle;

And inputting the vehicle image into a preset vehicle type identification model to determine the vehicle type of the target vehicle.

In one possible implementation manner, after determining the parking state of the target vehicle based on the first point cloud data corresponding to each critical edge component and each second point cloud data, the method further includes:

if the parking state of the target vehicle is not standard parking, acquiring the number of times of non-standard parking of the target vehicle in a preset period;

Updating the non-standard parking grade information of the target vehicle based on the non-standard parking times to obtain the updated non-standard parking grade of the target vehicle;

Judging whether the updated non-standard parking level is larger than a preset non-standard parking level threshold;

If yes, license plate number information of the target vehicle is obtained, the license plate number information is added into a preset blacklist, and the blacklist is the vehicle which is forbidden to enter and the corresponding license plate number information.

In a second aspect, the present application provides a parking state detection apparatus, which adopts the following technical scheme:

a parking state detection device, comprising:

the system comprises an acquisition module, a parking module and a control module, wherein the acquisition module is used for acquiring the vehicle type of a target vehicle and the identification of the type of a parkable vehicle in a parking space of the target vehicle;

The judging module is used for judging whether the vehicle type is consistent with the vehicle type corresponding to the parkable vehicle type identifier; if yes, triggering a point cloud data acquisition module;

The point cloud data acquisition module is used for acquiring first point cloud data corresponding to each key edge part of the target vehicle and a plurality of second point cloud data corresponding to a parking space of the target vehicle;

The parking state determining module is configured to determine a parking state of the target vehicle based on the first point cloud data corresponding to each key edge component and each second point cloud data, where the parking state includes: normal parking and non-normal parking.

Third, the application provides an electronic device, which adopts the following technical scheme:

At least one processor;

A memory;

At least one application program, wherein the at least one application program is stored in the memory and configured to be executed by the at least one processor, the at least one application program configured to: the parking state detection method according to any one of the first aspects is performed.

In a fourth aspect, the present application provides a computer readable storage medium, which adopts the following technical scheme:

A computer-readable storage medium having stored thereon a computer program which, when executed in a computer, causes the computer to perform the parking state detection method according to any one of the first aspects.

In summary, the application at least comprises the following beneficial technical effects:

The method comprises the steps of obtaining a vehicle type of a target vehicle and a parkable vehicle type identifier of a parking space of the target vehicle, judging whether the vehicle type is consistent with a vehicle type corresponding to the parkable vehicle type identifier, so as to avoid the situation that the target vehicle occupies the parking space of other vehicle types, and the corresponding type of vehicle cannot be parked normally, and judging the parking state of the target vehicle from the angle of the parking space type; if the types of the vehicles are the same, the first point cloud data of the key edge part of the target vehicle and the second point cloud data of the parking space are obtained to determine the parking state of the target vehicle.

Drawings

Fig. 1 is a schematic diagram of a parking status detection scenario provided in an embodiment of the present application.

Fig. 2 is a flow chart of a parking status detection method according to an embodiment of the present application.

Fig. 3 is a schematic structural diagram of a parking status detection device according to an embodiment of the present application.

Fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

The application is described in further detail below in connection with fig. 1 to 4.

The present embodiment is merely illustrative of the present application and is not intended to limit the present application, and those skilled in the art, after having read the present specification, may make modifications to the present embodiment without creative contribution as necessary, but are protected by patent laws within the scope of the present application.

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

In addition, the term "and/or" herein is merely an association relationship describing an association object, and means that three relationships may exist, for example, a and/or B may mean: a exists alone, A and B exist together, and B exists alone. In this context, unless otherwise specified, the term "/" generally indicates that the associated object is an "or" relationship.

Fig. 1 is a schematic diagram of a parking state detection scene provided by the embodiment of the application, after a target vehicle enters a parking space, a fixed camera acquires a target vehicle image, an electronic device determines that the target vehicle is parked according to the target vehicle image, and then the electronic device controls an unmanned aerial vehicle to acquire first point cloud data and second point cloud data corresponding to the target vehicle and the parking space of the target vehicle, and upload the first point cloud data and the second point cloud data to the electronic device, so that the electronic device analyzes the first point cloud data and the second point cloud data to realize detection of the parking state of the target vehicle.

The embodiment of the application provides a parking state detection method which is executed by electronic equipment, wherein the electronic equipment can be a server, and the server can be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server for providing cloud computing service. The servers may be directly or indirectly connected via wired or wireless communication, and embodiments of the present application are not limited in this regard.

Referring to fig. 2, fig. 2 is a schematic flow chart of a parking status detection method according to an embodiment of the present application, where the method includes steps S101, S102, S103, and S104, where:

Step S101, acquiring a vehicle type of a target vehicle and a parkable vehicle type identifier of a parking space of the target vehicle.

The vehicle type can be a new energy vehicle, a fuel vehicle, a bus, a minibus and a van, the vehicle type of the target vehicle can be determined through a vehicle type identification model, and the vehicle type can be quickly and accurately identified through the vehicle type identification model; the type identifier of the parkable vehicle is a sign board which is placed beside a parking space or in a corresponding parking area by a related technician in advance, and the type identifier of the parkable vehicle corresponding to the new energy vehicle can also be a charging pile; the type of the parked vehicle corresponding to each vehicle position and the type of the vehicle corresponding to each parked vehicle type identifier are input into the electronic equipment in advance.

Step S102, judging whether the vehicle type is consistent with the vehicle type corresponding to the parkable vehicle type identifier.

The matching algorithm may determine, for example, that the type of the parked vehicle is text information of "minibus", and then the text information corresponding to the type of the vehicle of the target vehicle may be matched with the text information, for example, the type of the vehicle of the identified target vehicle is "van", and the text information of "van" and "minibus" may be matched to determine whether the vehicle types are consistent. If yes, the vehicle type is consistent with the vehicle corresponding to the parkable vehicle type identifier, and step S103 is executed to acquire point cloud data. Otherwise, the vehicle type is inconsistent with the vehicle type corresponding to the type of the parkable vehicle, and the target vehicle can be determined to be not parked normally. It can be understood that the parking spaces corresponding to different vehicle types are different, when the vehicle type corresponding to the parking space of a certain vehicle is different from the vehicle type of the vehicle, the vehicle is indicated to occupy other parking spaces, and the vehicle corresponding to the vehicle type of the parking space can not normally park, and finally the overall parking efficiency of the parking lot is influenced, so that the vehicle type needs to be judged first, so that the vehicle in the parking lot is parked in the corresponding parking space, the management efficiency can be effectively improved while the attractiveness is improved, and the occurrence probability of the random parking phenomenon of the parking lot is reduced.

And step S103, if yes, acquiring first point cloud data corresponding to each key edge part of the target vehicle and a plurality of second point cloud data corresponding to a parking space of the target vehicle.

The critical edge components may be the rear view mirror, bumper and tail light, it being understood that the rear view mirror is located outermost in the body structure of the target vehicle; on a parking space, the probability of collision of the rearview mirror is maximum; the bumper and the tail lamp are positioned at the foremost side and the rearmost side of the target vehicle, respectively, and the probability of being collided is maximized, so that the rear view mirror, the bumper and the tail lamp are used as key edge parts. The plurality of second point cloud data corresponding to the parking space of the target vehicle may be a plurality of second point cloud data corresponding to the parking line, and it may be understood that four boundary lines are drawn in the parking space, each boundary line corresponds to different second point cloud data, so the parking space corresponds to the plurality of second point cloud data, and point cloud coordinates corresponding to the first point cloud data and the second point cloud data are point cloud coordinates under the same origin of coordinates.

The first point cloud data and the second point cloud data are obtained by analyzing video data of the target vehicle after parking, wherein the video data are captured by the unmanned aerial vehicle in a cruising way. The unmanned aerial vehicle is small in size, flexible and portable, can carry the inclination camera device and carry out video acquisition in each appointed position in the parking area, and the range of acquisition video is wider, and the obtained point cloud data is more accurate. After the image collected by the fixed camera of the parking lot analyzes that the vehicle is parked, the position of the parked vehicle is determined, then a proper unmanned aerial vehicle is selected from a plurality of unmanned aerial vehicles, and then the unmanned aerial vehicle is controlled to fly to the position of a parking space where the target vehicle is located, and video collection is carried out. It can be understood that the placement positions of different unmanned aerial vehicles are different, in order to improve the judging efficiency, the target unmanned aerial vehicle closest to the parking space of the target vehicle can be selected first, whether the remaining endurance time of the target unmanned aerial vehicle can reach the parking space of the target vehicle or not is judged, if yes, the target unmanned aerial vehicle can be selected to execute the acquisition of the point cloud data; otherwise, selecting the unmanned aerial vehicle closest to the second unmanned aerial vehicle as a target unmanned aerial vehicle for acquiring the point cloud data, judging whether the residual duration of the target unmanned aerial vehicle can reach the parking space of the target vehicle, and if not, re-screening until the target unmanned aerial vehicle is determined.

Step S104, determining a parking state of the target vehicle based on the first point cloud data and the second point cloud data corresponding to each key edge component, where the parking state includes: normal parking and non-normal parking.

It can be understood that the critical edge components are all located at the outermost side of the target vehicle, and are represented in a certain way, that is, when the critical edge components located at the outermost side of the target vehicle are located in the parking area, the target vehicle can be indicated to be located in the parking area, so that the accurate detection of the parking state of the target vehicle can be realized based on the first point cloud data and the second point cloud data of the critical edge components. It can be understood that when each key edge part has a certain distance from the corresponding parking line, the vehicle body of the target vehicle is indicated to be positioned in the parking line, and the parking state of the target vehicle is determined to be standard parking; if any critical edge component of the target vehicle is not located in the parking space of the target vehicle, the parking state of the target vehicle can be determined to be non-standard parking.

Based on the embodiment, the vehicle type of the target vehicle and the parkable vehicle type identifier of the parking space of the target vehicle are obtained, and whether the vehicle type is consistent with the vehicle type corresponding to the parkable vehicle type identifier is judged, so that the situation that the vehicle of the corresponding type cannot normally park due to the fact that the target vehicle occupies the parking space of other vehicle types is avoided, and the judgment of the parking state of the target vehicle is realized from the angle of the parking space type; if the types of the vehicles are the same, the first point cloud data of the key edge part of the target vehicle and the second point cloud data of the parking space are obtained to determine the parking state of the target vehicle.

Further, in the embodiment of the present application, before determining the parking state of the target vehicle based on the first point cloud data and the second point cloud data corresponding to the respective critical edge components, steps SA1 to SA3 (not shown in the drawings) are further included, where:

And step SA1, acquiring an environment image in a preset range corresponding to a parking space of the target vehicle.

The preset range is input into the electronic equipment in advance by a related technician, and can be determined according to historical data, namely the probability of the vehicle being collided and the distance between the vehicle are recorded one by one, and the maximum preset range corresponding to the collision probability smaller than a preset collision probability threshold value is determined as the preset range; it can be understood that the preset range can be a preset range formed by circles with any point in the vehicle as a circle center and the preset distance as a radius, and each collision probability corresponds to a different radius. The environment image is a looking-around image acquired by the unmanned aerial vehicle and is uploaded to the electronic equipment. By setting the preset collision probability threshold and the preset distance, the collision probability of the target vehicle in the preset range can be accurately determined, and the probability of the target vehicle being collided is reduced.

And SA2, identifying an environment image, and obtaining an obstacle identification result within a preset range of a parking space of the target vehicle.

Specifically, the environment image can be identified through the image identification algorithm, the embodiment of the application does not limit the specific image identification algorithm, and the user can set the environment image by himself; the obstacle recognition result may include: an obstacle is present and an obstacle is not present, and when an obstacle is present, position information of the obstacle may be included; in the embodiment of the present application, when an object other than a vehicle exists within the preset range of the parking space of the target vehicle, the object other than the vehicle may be determined as an obstacle.

And step SA3, if the obstacle recognition result is that the obstacle exists, acquiring third point cloud data corresponding to the obstacle.

Correspondingly, determining the parking state of the target vehicle based on the first point cloud data and the second point cloud data corresponding to each key edge part comprises the following steps:

and determining the parking state of the target vehicle based on the first point cloud data, the second point cloud data and the third point cloud data corresponding to the key edge parts.

If the obstacle identification result indicates that an obstacle exists, the electronic device controls the unmanned aerial vehicle to perform laser scanning on the obstacle to acquire third point cloud data of the obstacle, and it can be understood that each point cloud data corresponds to a plurality of point cloud coordinates, and in order to improve calculation accuracy, the point cloud coordinates corresponding to the first point cloud data, the second point cloud data and the third point cloud data can be mapped to the same coordinate system, namely, the point cloud coordinates corresponding to the same origin of coordinates. Accordingly, the parking state of the target vehicle can be determined according to the first point cloud data and the collision probability corresponding to the rearview mirror, and it can be understood that when an obstacle exists in the preset range of the parking space of the target vehicle, the probability of the target vehicle being scratched and collided can be greatly increased, and in order to reduce the probability of the scratch and the collision, the target vehicle needs to be far away from the obstacle under the condition of not exceeding the parking line; further, whether the distance between the other side rearview mirror, namely the target rearview mirror and the parking line corresponding to the side of the target rearview mirror is not larger than a preset second distance threshold value is judged, if the parking line corresponding to the side of the target rearview mirror and the target rearview mirror is level, the situation that the target vehicle is far away from the obstacle as far as possible under the condition of being positioned in the parking line is indicated, and therefore the condition that the target vehicle is in a standard parking state can be determined. If the obstacle recognition result indicates that no obstacle exists, the parking state of the target vehicle can be determined directly based on the first point cloud data and the second point cloud data. The second distance threshold is not limited in the embodiment of the application.

Based on the above embodiment, the environmental image of the target vehicle is acquired, and the environmental image is identified to determine whether an obstacle exists, when the obstacle exists around the target vehicle, the target vehicle is required to be parked in the parking space and further away from the obstacle to avoid collision, so that when the obstacle exists, the parking state of the target vehicle is required to be comprehensively determined according to the obstacle, the first point cloud data and the second point cloud data, and the accuracy of parking state detection is effectively improved.

Further, in an embodiment of the present application, the critical edge components include at least: step S104 of determining the parking status of the target vehicle based on the first point cloud data, the second point cloud data, and the third point cloud data corresponding to the respective critical edge components includes steps SB 1-SB 6 (not shown in the drawings), wherein:

Step SB1, based on the third point cloud data and the first point cloud data corresponding to each rearview mirror, determining a first distance corresponding to each rearview mirror and the obstacle.

Specifically, the point cloud data includes a plurality of point cloud coordinates, so that any point cloud coordinate in the third point cloud data and any point cloud coordinate corresponding to each rearview mirror can be selected respectively; the outermost side of the rearview mirrors, namely the point cloud coordinate of the closest obstacle at the rearview mirrors, can be selected according to the point cloud data corresponding to the rearview mirrors, the selected points corresponding to each rearview mirror are the same, and further a first distance between the obstacle and each rearview mirror can be obtained based on the point cloud coordinate; the plurality of distances between the outermost point of the obstacle and each rearview mirror can be calculated by the plurality of point cloud coordinates of the outermost point of the obstacle and the corresponding rearview mirrors, and the minimum distance is selected from the plurality of distances to be used as the first distance. It will be appreciated that both the selection of the outermost point of the obstacle and the selection of the outermost point of the mirror may be achieved by taking a two-dimensional elevation of the obstacle and the vehicle and determining the outermost point from the two-dimensional elevation.

Step SB2, determining the collision probability corresponding to the first distance of the target, which is the distance between the rearview mirror on the same side as the obstacle and the obstacle, based on the preset corresponding relation between the distance and the collision probability and the first distance of the target.

The corresponding relation between the preset distance and the collision probability can be obtained by a Bayesian probability calculation formula, namelyWherein/>Representing the probability of collision that occurs with a first distance,/>Representing the probability of being the first distance in the event of a collision,/>Expressed as a probability that the vehicle is a first distance from the obstacle,/>Is the probability of a collision of the vehicle with an obstacle, wherein,And/>All can be obtained through a plurality of historical data,/>For example, five collisions occur, where the three collisions correspond to the first distance, and the probability of the collision being the first distance is obtained. It can be understood that two rearview mirrors are arranged on the target vehicle, when an obstacle exists in the preset range of the target vehicle, in order to accurately calculate the collision probability between the target vehicle and the obstacle, preferably, the rearview mirror close to the obstacle side is selected as the target rearview mirror, and the collision probability between the target rearview mirror and the obstacle is calculated; if at least one obstacle exists on each of different sides of the target vehicle, calculating the collision probability of each obstacle and the target vehicle, and selecting the maximum collision probability as the actual collision probability.

And step SB3, judging whether the collision probability is smaller than a preset collision probability threshold value, and obtaining a first judgment result.

The preset collision probability threshold may be preset by a related technician based on experience and stored in the electronic device, and the embodiment of the present application does not limit the preset collision probability threshold. The first judgment result comprises: the collision probability is less than a preset collision probability threshold, or the collision probability is not less than a preset collision probability threshold.

Further, if so, the probability of collision between the target vehicle and the obstacle is indicated to be high; otherwise, it indicates that the probability of collision of the target vehicle with the obstacle is small.

Step SB4, based on the first point cloud data and the second point cloud data of the target rearview mirror, determining a second distance between the target rearview mirror and a target boundary line of a parking space where the target vehicle is located, wherein the target rearview mirror is a rearview mirror on the opposite side of the side where the obstacle is located, and the target boundary line is a parking space boundary line on the same side as the target rearview mirror.

When the target rearview mirror is determined, second point cloud data corresponding to a parking line in a parking space on the side where the target rearview mirror is located, namely second point cloud data of a target boundary line, are obtained, and in order to accurately judge the parking state, the outermost point cloud coordinates corresponding to the target rearview mirror can be selected from the first point cloud data, second distances corresponding to each of the outermost point cloud coordinates and each of the second point cloud coordinates in the second point cloud data are calculated, and the minimum second distance is selected as the second distance between the target rearview mirror and the target boundary line.

And step SB5, judging whether the second distance is not larger than a preset second distance threshold value, and obtaining a second judgment result.

In the embodiment of the application, the preset second distance threshold is not limited, the user can set the preset second distance threshold by himself, and the preset second distance threshold can be input into the electronic device in advance by a related technician based on experience. The second judgment result comprises: the second distance is not greater than a preset second distance threshold, or the second distance is greater than a second preset distance threshold.

If so, the distance between the target vehicle and the target boundary line is indicated to be relatively short, namely the target vehicle is close to the target boundary line, and if the target vehicle continues to move, the influence on vehicles in other parking spaces can be generated; otherwise, the distance between the target vehicle and the target boundary line is far, the target vehicle can continue to move to the position where the target boundary line is located, and other vehicles cannot be affected in the parking space.

Step SB6, determining the parking state of the target vehicle according to the first judgment result and the second judgment result.

When the first judgment result is that the collision probability is not smaller than the preset collision probability threshold value, and the second judgment result is that the second distance is not larger than the preset second distance threshold value, the fact that the distance between the target vehicle and the target boundary line is relatively close is indicated, at the moment, the target vehicle is far away from the barrier under the condition that the target vehicle is already located in the parking space, if the target vehicle is far away from the barrier, the condition that the target vehicle exceeds the parking space and affects vehicles in other parking spaces possibly exists, and therefore the parking state of the target vehicle is determined to be standard parking at the moment; when the first judgment result is that the collision probability is not smaller than the preset collision probability threshold value, and the second judgment result is that the second distance is larger than the preset distance threshold value, the collision probability of the target vehicle and the obstacle is smaller, and the vehicles on surrounding parking spaces are not influenced, so that the parking state of the target vehicle is determined to be non-standard parking; when the first judgment result is that the collision probability is smaller than the preset collision probability, and the second judgment result is that the second distance is not larger than the preset second distance threshold value, the situation that the target vehicle is far away from the obstacle under the condition of being positioned in the parking space is indicated, so that the parking state of the target vehicle can be determined to be standard parking; when the first judgment result is that the collision probability is not smaller than the preset collision probability threshold value and the second judgment result is that the second distance is larger than the preset distance threshold value, the parking state of the target vehicle is not standard.

Based on the above embodiment, the first distance is determined according to the point cloud data of the obstacle and the first point cloud data corresponding to each rearview mirror, so as to accurately determine the probability of collision between the target vehicle and the obstacle, and obtain the first judgment result; obtaining a second distance between the target rearview mirror and a target boundary line according to the first point cloud data and the second point cloud data so as to determine whether the distance between the side corresponding to the side where the obstacle is located and the parking line is closer or not, and obtaining a second judgment result; the method and the device realize accurate detection of the parking state of the target vehicle from the angle of collision probability and the angle of the rearview mirror and the corresponding side boundary line, and effectively improve the detection efficiency of the parking state.

Further, in an embodiment of the present application, the critical edge component further comprises: if the bumper and the tail lamp have no obstacle as a result of obstacle recognition, determining a parking state of the target vehicle based on the first point cloud data and the second point cloud data corresponding to the key edge components, including steps SC1 to SC4 (not shown in the drawings), wherein:

step SC1, determining a third distance between the rearview mirror and a first parking space line based on first point cloud data of the rearview mirror and second point cloud data corresponding to the first parking space line, wherein the first parking space line is a parking space line positioned on the side of the rearview mirror.

Specifically, when the obstacle recognition result indicates that no obstacle exists, the third distance is determined, and it can be understood that two rearview mirrors exist in the target vehicle, so that the distance between each rearview mirror and the corresponding parking space line can be calculated respectively, and the minimum distance is selected as the third distance, and it can be understood that the minimum distance is selected to be more representative.

And selecting the outermost point cloud coordinate on the rearview mirror from the first point cloud data, selecting the point cloud coordinate closest to the rearview mirror from the second point cloud data corresponding to the first vehicle position line, and calculating the two point cloud coordinates to obtain a third distance between the rearview mirror and the first vehicle position line. In the embodiment of the present application, the first parking space line in step SB1 is the target boundary line in step SA 4.

And step SC2, determining a fourth distance between the bumper and a second vehicle position line based on the first point cloud data of the bumper and the second point cloud data corresponding to the second vehicle position line, wherein the second vehicle position line is a vehicle position line positioned at the side of the bumper.

The electronic device recognizes the image with the bumper and selects the forefront point, and it can be understood that the bumper is positioned at the forefront of the target vehicle, and when the bumper is positioned in the parking area, the vehicle head part does not exceed the parking line, so that whether the target vehicle exceeds the parking line can be accurately determined based on the bumper. The second parking space line is a parking space line corresponding to the side where the bumper is located. The calculation manner of the fourth distance is the same as that of the third distance in step SB1, and detailed description of the embodiment of the present application is omitted. The front-most point of the bumper may be selected by acquiring a two-dimensional side view of the target vehicle, and if the head portion of the target vehicle is located at the leftmost end of the two-dimensional side view in the two-dimensional side view, the leftmost point may be selected from the bumper region as the front-most point of the bumper.

And step SC3, determining a fifth distance between the tail lamp and a third parking space line based on the first point cloud data of the tail lamp and the second point cloud data corresponding to the third parking space line, wherein the third parking space line is a parking space line positioned at the side of the tail lamp.

The method comprises the steps of identifying a tail lamp image to select a point on the tail lamp, which is positioned at the rearmost side of a target vehicle, acquiring point cloud coordinates corresponding to the point from first point cloud data, selecting point cloud coordinates corresponding to a point closest to the tail lamp from second point cloud data of a parking space line corresponding to the side of the tail lamp, and determining a fifth distance between the tail lamp and a third parking space line based on the two point cloud coordinates. It can be understood that the tail lamp is the extreme structure in the target vehicle body structure, and when the tail lamp does not exceed the parking line, the tail lamp of the target parking space can be indicated not to exceed the parking line, so that the accurate detection of the parking state of the target vehicle can be realized based on the tail lamp. If the fifth distances corresponding to the two tail lamps of the target vehicle are the same, the two tail lamps can be selected randomly; and if the fifth distances corresponding to the two tail lamps of the target vehicle are different, selecting the tail lamp corresponding to the smallest fifth distance. The selection of the point on the rearmost side of the tail lamp may be to acquire a two-dimensional side view of the target vehicle, and if the head portion of the target vehicle is located at the leftmost end in the two-dimensional side view, the point on the rightmost side of the tail lamp may be selected as the point on the rearmost side of the tail lamp.

And step SC4, determining the parking state of the target vehicle according to the third distance, the fourth distance and the fifth distance.

Specifically, judging that the third distance is smaller than a preset distance threshold, judging that the third distance is larger than a preset distance threshold, judging that the fourth distance is smaller than a preset distance threshold, judging that the fifth distance is smaller than a preset distance threshold, judging that the distance between a rearview mirror, a bumper or a tail lamp of a target vehicle and a corresponding parking space line is smaller than the distance between the rearview mirror, the bumper or the tail lamp of the target vehicle and the corresponding parking space line, judging that the probability of exceeding the parking space line is higher, and judging that the occurrence and the parking of the vehicles on surrounding parking spaces are also greatly influenced, so that the parking state of the target vehicle is determined to be irregular parking, wherein any distance can be the third distance, the fourth distance and the fifth distance, and the corresponding preset distance threshold can be the preset third distance threshold, the preset fourth distance threshold and the preset fifth distance threshold; when the third distance is not smaller than the preset third distance threshold value, the fourth distance is not smaller than the preset fourth distance threshold value, and the fifth distance is not smaller than the preset fifth distance threshold value, the parking state of the target vehicle is determined to be normal parking, namely, a certain distance exists between the rearview mirror, the bumper and the tail lamp of the target vehicle and a corresponding parking line, generally, when the vehicle is in a state that the vehicle body is aligned and normal parking is performed, the rearview mirror, the bumper and the tail lamp are far away from the corresponding parking line, and therefore the parking state of the target vehicle can be determined to be normal parking. The embodiment of the application does not limit the preset third distance threshold, the preset fourth distance threshold and the preset fifth distance threshold, and the preset third distance threshold, the preset fourth distance threshold and the preset fifth distance threshold are all preset by related technicians and are input into the electronic equipment.

Based on the above embodiment, when the obstacle recognition result is that there is no obstacle, the target vehicle may be judged directly based on the critical edge component of the target vehicle; the rearview mirror, the bumper and the tail lamp are all outermost parts of the target vehicle, namely key edge parts, when the key edge parts of the target vehicle are far away from corresponding parking space lines, the target vehicle can be indicated to be in a standard parking state, and accurate judgment of the parking state of the target vehicle is achieved through the key edge parts.

Further, in the embodiment of the present application, before the step S104 determines that the parking state of the target vehicle is the normal parking, the method further includes steps SD1 to SD3 (not shown in the drawings), wherein:

and step SD1, obtaining parking direction mark information of a parking space where the target vehicle is located.

The parking pointer information may be input to the electronic device in advance, that is, the parking space number information and the corresponding parking pointer information are input to the electronic device at the same time. After the target vehicle finishes parking, the unmanned aerial vehicle can confirm the corresponding beacon information through obtaining the number of the parking space where the target vehicle is located; the parking direction indicator information can also be obtained by directly obtaining the image of the target vehicle parking space, and the direction indicator information in the parking space is identified in the process that the target vehicle enters the parking space so as to determine the parking direction indicator information. The direction corresponding to the parking direction indicator information may be: east, west, south and north.

Step SD2, acquiring the orientation information of the target vehicle.

The direction information of the parking lot can be determined according to the head part of the target vehicle, and the direction information of the target vehicle can be determined according to the direction of the head in the parking lot. In the embodiment of the application, in order to accurately determine the direction information of the target vehicle, the direction information is identical to the parking direction indicator information.

And step SD3, obtaining a third judgment result according to the direction information and the parking direction indicator information, wherein the third judgment result represents whether the parking direction of the target vehicle is compliant.

Correspondingly, determining the parking state of the target vehicle according to the first judgment result and the second judgment result comprises the following steps:

determining the parking state of the target vehicle according to the first, second and third judging results

The heading information is matched with the parking heading information to determine compliance. The third determination result may be that the orientation information is the same as the parking direction indicator information, or the orientation information is different from the parking direction indicator information. Correspondingly, when the orientation information is the same as the parking direction mark information, executing step SA1 to step SA3 to acquire first point cloud data of key edge parts of the target vehicle and second point cloud data of the parking space, judging whether an obstacle exists in a preset range of the target vehicle, and executing step SB1 to step SB6 to acquire third point cloud data of the obstacle if the obstacle exists, so as to determine the parking state of the target vehicle based on the third point cloud data, the first point cloud data and the second point cloud data; if no obstacle is present, steps SC1 to SC4 are performed to determine a vehicle state of the target vehicle based on the first point cloud data and the second point cloud data of the critical edge component.

Based on the above embodiment, the parking direction indicator information and the direction information of the parking space of the target vehicle are obtained, and according to the direction information and the parking direction indicator information, whether the head direction of the target vehicle is the same as the direction indicator is determined, when the direction indicator is arranged in the parking lot, the vehicle needs to park according to the direction information of the direction indicator, and a third judgment result is obtained; and judging whether the target vehicle is normally parked or not from multiple angles according to the first judging result, the second judging result and the third judging result, so that the judging accuracy is effectively improved.

Further, in an embodiment of the present application, acquiring a vehicle type of a target vehicle includes:

A vehicle image of a target vehicle is acquired.

Specifically, the vehicle image of the target vehicle may be acquired by the unmanned aerial vehicle when the target vehicle enters the parking lot, or may be acquired after the target vehicle is parked in the parking lot, so as to accurately identify the vehicle type of the target vehicle, where the vehicle image should at least include: a head portion of the target vehicle, a body portion of the target vehicle, or a tail portion of the target vehicle; the vehicle image may be one or more, and the embodiment of the application is not limited.

The vehicle image is input into a preset vehicle type recognition model to determine the vehicle type of the target vehicle.

The vehicle type recognition model can be obtained based on a plurality of vehicle training samples and convolutional neural network training, the specific training process of the vehicle type recognition model is not limited, and the user can set the vehicle type recognition model by himself. The vehicle image is input into a vehicle type recognition model, which outputs the vehicle image with the vehicle type information.

Based on the above-described embodiments, a vehicle image of a target vehicle is acquired, and the vehicle image is input into a preset vehicle class recognition model to determine the vehicle type, so that the accuracy of the vehicle type is improved by the vehicle class recognition model.

Further, in an embodiment of the present application, after determining the parking state of the target vehicle based on the first point cloud data and the second point cloud data corresponding to each key edge component, the method further includes:

and if the parking state of the target vehicle is the non-standard parking, acquiring the non-standard parking times of the target vehicle in a preset period.

Specifically, the preset period may be 15 days, one month or three months, and the embodiment of the present application does not limit the preset period. When the electronic equipment determines that the vehicle is not parked normally, the electronic equipment automatically records license plate information of the vehicle and corresponding non-normal parking times.

Updating the non-standard parking grade information of the target vehicle based on the non-standard parking times to obtain the updated non-standard parking grade of the target vehicle.

Updating the non-standard parking grade by recording the non-standard parking times, namely updating when the non-standard parking times of the vehicle reach the minimum times corresponding to the non-standard parking grade; in the embodiment of the present application, the unnormalized parking levels may be classified into a first level, a second level, and a third level. As the number of levels increases, the corresponding number of unnormalized stops increases. The number of parks corresponding to each non-standard parking level is not limited in the embodiment of the present application.

And judging whether the updated non-standard parking level is smaller than a preset non-standard parking level threshold.

Specifically, the preset non-standard parking level threshold may be input into the electronic device in advance by a related technician, and the embodiment of the application does not limit the non-standard parking level threshold, so that the user can set the non-standard parking level threshold by himself. By setting the non-standard parking grade, the reasonable management of the parking of the vehicles can be realized, and the number of the vehicles which are not normally parked in the parking lot is reduced. Further, if yes, the condition that the target vehicle is stopped for a plurality of times without standardization is indicated; otherwise, the condition that the target vehicle does not exist or has a small number of times of non-standard parking is indicated, and other vehicles in the parking lot and the parking lot are not affected.

The license plate number information of the target vehicle can be obtained by identifying the vehicle image of the target vehicle, and the license plate number information is added into a blacklist, wherein the blacklist is automatically generated and updated in real time by electronic equipment, and vehicles added into the blacklist cannot enter a parking lot.

Based on the above embodiment, when it is determined that the parking state of the target vehicle is not normal parking, the number of times of non-normal parking of the target vehicle in the preset period is obtained so as to update the non-normal parking level of the target vehicle, and the non-normal parking condition of the target vehicle can be accurately determined by updating the non-normal parking level; and judging the updated standard parking grade of the target vehicle so as to determine whether to add the target vehicle into the blacklist, and when the number of times of non-standard parking of the target vehicle is excessive, the parking of other vehicles and the management of a parking lot can be seriously influenced, so that the target vehicle needs to be added into the blacklist to limit the target vehicle from being forbidden to enter.

The above-described embodiments describe a parking state detection method from the viewpoint of a method flow, and the following embodiments describe a parking state detection device from the viewpoint of a virtual module or a virtual unit, and the following embodiments are described in detail.

An embodiment of the present application provides a parking state detection device, as shown in fig. 3, where the parking state detection device specifically may include:

an obtaining module 210, configured to obtain a vehicle type of the target vehicle and a parkable vehicle type identifier of a parking space in which the target vehicle is located;

a judging module 220, configured to judge whether the vehicle type is consistent with the vehicle type corresponding to the parkable vehicle type identifier; if yes, the point cloud data acquisition module 230 is triggered;

The point cloud data acquisition module 230 is configured to acquire first point cloud data corresponding to each key edge component of the target vehicle and a plurality of second point cloud data corresponding to a parking space of the target vehicle;

The parking state determining module 240 is configured to determine a parking state of the target vehicle based on each of the first point cloud data and each of the second point cloud data, where the parking state includes: normal parking and non-normal parking.

In one possible implementation manner of the embodiment of the present application, the parking state detection device further includes:

An obstacle recognition module for:

Acquiring an environment image in a preset range corresponding to a parking space of a target vehicle;

identifying an environment image, and obtaining an obstacle identification result in a preset range of a parking space of a target vehicle;

accordingly, the parking status determining module 240 is configured to, when executing the determination of the parking status of the target vehicle based on the first point cloud data and the second point cloud data corresponding to the respective critical edge components:

In one possible implementation manner of the embodiment of the present application, the parking status determining module 240 is configured to, when executing the determination of the parking status of the target vehicle based on the first point cloud data, the second point cloud data, and the third point cloud data corresponding to each key edge component:

Determining a collision probability corresponding to a first target distance based on a preset corresponding relation between the distance and the collision probability and the first target distance, wherein the first target distance is the distance between the rearview mirror on the same side as the obstacle and the obstacle;

Judging whether the collision probability is smaller than a preset collision probability threshold value or not to obtain a first judgment result;

In one possible implementation manner of the embodiment of the present application, if the obstacle recognition result indicates that no obstacle exists, the parking status determining module 240 is configured to, when executing determining the parking status of the target vehicle based on the first point cloud data and the second point cloud data corresponding to each key edge component:

Determining a fourth distance between the bumper and a second vehicle position line based on the first point cloud data of the bumper and the second point cloud data corresponding to the second vehicle position line, wherein the second vehicle position line is a vehicle position line positioned on the bumper side;

determining a fifth distance between the tail lamp and a third parking space line based on the first point cloud data of the tail lamp and second point cloud data corresponding to the third parking space line, wherein the third parking space line is a parking space line positioned at the tail lamp side;

A third judgment result determining module, configured to:

obtaining parking direction mark information of a parking space where a target vehicle is located;

acquiring the orientation information of a target vehicle;

obtaining a third judgment result according to the direction information and the parking direction standard information, wherein the third judgment result represents whether the parking direction of the target vehicle is compliant or not;

Accordingly, the parking state determining module 240 is configured to, when determining the parking state of the target vehicle according to the first determination result and the second determination result:

A vehicle type acquisition module for:

acquiring a vehicle image of a target vehicle;

An unnormalized parking grade determination module for:

Updating the standard non-parking grade information of the target vehicle based on the non-standard parking times to obtain an updated non-standard parking grade of the target vehicle;

Judging whether the updated non-standard parking level is larger than a preset non-standard parking level threshold value or not;

It will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process of the above-described parking state detection device may refer to the corresponding process in the foregoing method embodiment, and will not be described in detail herein.

An embodiment of the present application provides an electronic device, as shown in fig. 4, fig. 4 is a schematic structural diagram of the electronic device provided in the embodiment of the present application, and an electronic device 300 shown in fig. 4 includes: a processor 301 and a memory 303. Wherein the processor 301 is coupled to the memory 303, such as via a bus 302. Optionally, the electronic device 300 may also include a transceiver 304. It should be noted that, in practical applications, the transceiver 304 is not limited to one, and the structure of the electronic device 300 is not limited to the embodiment of the present application.

The Processor 301 may be a CPU (Central Processing Unit ), general purpose Processor, DSP (DIGITAL SIGNAL Processor, data signal Processor), ASIC (Application SPECIFIC INTEGRATED Circuit), FPGA (Field Programmabl EGATE ARRAY ) or other programmable logic device, transistor logic device, hardware component, or any combination thereof. Which may implement or perform the various exemplary logic blocks, modules and circuits described in connection with the disclosure of embodiments of the application. Processor 301 may also be a combination that implements computing functionality, e.g., comprising one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

Bus 302 may include a path to transfer information between the components. Bus 302 may be a PCI (PERIPHERAL COMPONENT INTERCONNECT, peripheral component interconnect standard) bus or an EISA (Extended Industry Standard Architecture ) bus, or the like. Bus 302 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in fig. 4, but not only one bus or one type of bus.

The Memory 303 may be, but is not limited to, a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory ) or other type of dynamic storage device that can store information and instructions, an EEPROM (ELECTRICALLY ERASABLE PROGRAMMABLE READ ONLY MEMORY ), a CD-ROM (Compact Disc Read Only Memory, compact disc Read Only Memory) or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer.

The memory 303 is used for storing application program codes for executing embodiments of the present application and is controlled to be executed by the processor 301. The processor 301 is configured to execute the application code stored in the memory 303 to implement what is shown in the foregoing method embodiments.

Among them, electronic devices include, but are not limited to: mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and the like, and stationary terminals such as digital TVs, desktop computers, and the like. The electronic device shown in fig. 4 is only an example and should not be construed as limiting the functionality and scope of use of the embodiments of the application.

Embodiments of the present application provide a computer-readable storage medium having a computer program stored thereon, which when run on a computer, causes the computer to perform the corresponding method embodiments described above.

It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited in order and may be performed in other orders, unless explicitly stated herein. Moreover, at least some of the steps in the flowcharts of the figures may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, the order of their execution not necessarily being sequential, but may be performed in turn or alternately with other steps or at least a portion of the other steps or stages.

The foregoing is only a partial embodiment of the present application, and it should be noted that it will be apparent to those skilled in the art that modifications and adaptations can be made without departing from the principles of the present application, and such modifications and adaptations should and are intended to be comprehended within the scope of the present application.

Claims

1. A parking state detection method, characterized by comprising:

If yes, acquiring first point cloud data corresponding to each key edge part of the target vehicle and a plurality of second point cloud data corresponding to a parking space of the target vehicle, wherein the first point cloud data and the second point cloud data are taken by an unmanned aerial vehicle in a cruising way, and the unmanned aerial vehicle is determined based on the placement position of the unmanned aerial vehicle and the position of the parking space of the target vehicle;

Determining a parking state of the target vehicle based on the first point cloud data corresponding to each key edge part and each second point cloud data, wherein the parking state comprises: normal parking and non-normal parking;

The method further includes, before determining the parking state of the target vehicle based on the first point cloud data corresponding to each critical edge component and each second point cloud data:

Determining a parking state of the target vehicle based on the first point cloud data, the second point cloud data and the third point cloud data corresponding to the key edge components;

the determining, based on the first point cloud data, the second point cloud data, and the third point cloud data corresponding to the key edge components, a parking state of the target vehicle includes:

determining a parking state of the target vehicle according to the first judging result and the second judging result;

Wherein the critical edge components include: rearview mirrors, bumpers and tail lamps, if the obstacle recognition result is that no obstacle exists, determining the parking state of the target vehicle based on the first point cloud data and the second point cloud data corresponding to the key edge components includes:

2. The parking state detection method according to claim 1, characterized by further comprising, before determining that the parking state of the target vehicle is a normal parking:

acquiring the orientation information of the target vehicle;

3. The parking state detection method according to claim 1, characterized in that acquiring a vehicle type of a target vehicle includes:

acquiring a vehicle image of the target vehicle;

4. The method according to claim 1, wherein after determining the parking state of the target vehicle based on the first point cloud data corresponding to the respective key edge members and the respective second point cloud data, further comprising:

5. A parking state detection device, characterized by comprising:

The system comprises a point cloud data acquisition module, a target vehicle parking space acquisition module and a target vehicle, wherein the point cloud data acquisition module is used for acquiring first point cloud data corresponding to each key edge part of the target vehicle and a plurality of second point cloud data corresponding to the target vehicle parking space, the first point cloud data and the second point cloud data are acquired by cruising and shooting of an unmanned aerial vehicle, and the unmanned aerial vehicle is determined based on the placement position of the unmanned aerial vehicle and the position of the target vehicle parking space;

The parking state determining module is configured to determine a parking state of the target vehicle based on the first point cloud data corresponding to each key edge component and each second point cloud data, where the parking state includes: normal parking and non-normal parking;

wherein, the parking state detection device further includes:

an obstacle detection module for:

The parking state determining module is configured to, when executing the determination of the parking state of the target vehicle based on the first point cloud data, the second point cloud data, and the third point cloud data corresponding to the respective key edge components:

Wherein the critical edge components include: the parking state determining module is used for determining the parking state of the target vehicle when executing the determination based on the first point cloud data and the second point cloud data corresponding to the key edge components if the obstacle recognition result is that no obstacle exists, wherein the first point cloud data and the second point cloud data correspond to the key edge components:

6. An electronic device, comprising:

At least one processor;

A memory;

At least one application program, wherein the at least one application program is stored in the memory and configured to be executed by the at least one processor, the at least one application program configured to: the parking state detection method according to any one of claims 1 to 4.

7. A computer-readable storage medium, having stored thereon a computer program which, when executed in a computer, causes the computer to perform the parking state detection method according to any one of claims 1 to 4.