A method, apparatus, device and storage medium for determining a parking point

Abstract

The present disclosure provides a method, device and equipment for determining a parking point, and a storage medium, relates to the technical field of artificial intelligence, and particularly relates to the fields of automatic driving, autonomous parking, intelligent traffic and the like. The specific implementation scheme is as follows: a suspected parking point is obtained; a parking point meeting a preset road network requirement is selected from the suspected parking point, so as to obtain a screened parking point; a road environment image corresponding to the screened parking point is acquired, wherein the road environment image represents an image of a road where the screened parking point is located; road data of the road environment image is identified; and in response to the road data meeting a preset road requirement, the screened parking point is determined as a target parking point. The present disclosure can verify the suspected parking point, and improve the accuracy of the determined parking point.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of artificial intelligence, in particular to the fields of automatic driving, autonomous parking, intelligent transportation, etc. BACKGROUND

[0002] A traditional map vendor collects road form data by collecting vehicles, and the collected road form data can include road length, road width, road direction, road curvature, road grade, whether it is a separated road, etc. Then, the collected road form data is objectively presented in an electronic map system and acts on people's daily travel navigation. With the progress of collection and production technology, the map field has developed from traditional maps to lane-level, high-precision, and other more detailed and comprehensive information maps. In the application layer, products have gradually been applied to many traditional industries. Therefore, it is particularly important to provide fine data mining. SUMMARY

[0003] The present disclosure provides a method, device, equipment and storage medium for determining a parking point.

[0004] According to a first aspect of the present disclosure, a method for determining a parking point is provided, comprising:

[0005] obtaining a suspected parking point;

[0006] screening a parking point meeting a preset road network requirement from the suspected parking point to obtain a screened parking point;

[0007] obtaining a road environment image corresponding to the screened parking point, wherein the road environment image represents an image of a road where the screened parking point is located;

[0008] identifying road data of the road environment image;

[0009] in response to the road data meeting a preset road requirement, determining the screened parking point as a target parking point.

[0010] According to a second aspect of the present disclosure, a device for determining a parking point is provided, comprising:

[0011] an obtaining module configured to obtain a suspected parking point;

[0012] a screening module configured to screen a parking point meeting a preset road network requirement from the suspected parking point to obtain a screened parking point;

[0013] an obtaining module configured to obtain a road environment image corresponding to the screened parking point, wherein the road environment image represents an image of a road where the screened parking point is located;

[0014] an identifying module configured to identify road data of the road environment image;

[0015] determining, in response to the road data satisfying the preset road requirement, that the screened parking point is a target parking point.

[0016] According to a third aspect of the present disclosure, an electronic device is provided, comprising:

[0017] at least one processor; and

[0018] a memory connected with the at least one processor in communication; wherein,

[0019] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of the first aspect.

[0020] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to perform the method according to the first aspect.

[0021] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the method according to the first aspect.

[0022] The present disclosure can verify the suspected parking point by screening the parking point meeting the preset road network requirement from the suspected parking point, obtaining the screened parking point, and obtaining the road environment image corresponding to the screened parking point, identifying the road data of the road environment image, and determining the screened parking point as the target parking point in response to the road data satisfying the preset road requirement, thereby improving the accuracy of the determined parking point.

[0023] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the present disclosure, nor to limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0024] The accompanying drawings are used to better understand the present scheme and do not limit the present disclosure. Among them:

[0025] Figure 1 is a flowchart of a method for determining a parking point provided by an embodiment of the present disclosure;

[0026] Figure 2 is a flowchart of obtaining a suspected parking point in an embodiment of the present disclosure;

[0027] Figure 3 is a flowchart of obtaining a road environment image corresponding to a screened parking point in an embodiment of the present disclosure;

[0028] Figure 4A is a schematic diagram of recognition by using a semantic segmentation model in an embodiment of the present disclosure;

[0029] Figure 4B is a schematic diagram of a conical ice cream cone recognized in an embodiment of the present disclosure;

[0030] Figure 4C is a schematic diagram of a road surface state recognized in an embodiment of the present disclosure;

[0031] Figure 5 is a schematic diagram of applying a method for determining a parking point provided by an embodiment of the present disclosure;

[0032] Figure 6 is a schematic diagram of using image recognition to verify a parking point in an embodiment of the present disclosure;

[0033] Figure 7 is a structural schematic diagram of a device for determining a parking point provided by an embodiment of the present disclosure;

[0034] Figure 8 is a block diagram of an electronic device for implementing a method for determining a parking point according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0035] Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, which include various details of the embodiments of the present disclosure to assist in understanding, and should be considered as merely exemplary. Thus, those of ordinary skill in the art will recognize various changes and modifications to the embodiments described herein, without departing from the scope and spirit of the present disclosure. Also, descriptions of known functions and constructions are omitted in the following description for clarity and conciseness.

[0036] A parking point can represent a position where a vehicle can be parked. The mined parking points can be combined with road network data and the like for user guidance, guiding the user to park the vehicle at the parking point, facilitating user travel, reducing the risk of freight drivers violating regulations due to incomplete information, and improving travel experience.

[0037] The method for determining a parking point provided by an embodiment of the present disclosure can be applied to an electronic device, specifically, the electronic device can be a server, a terminal, and the like.

[0038] An embodiment of the present disclosure provides a method for determining a parking point, which can include:

[0039] Obtaining a suspected parking point;

[0040] Selecting a parking point meeting a preset road network requirement from the suspected parking point to obtain a screened parking point;

[0041] obtain a road environment image corresponding to the screened parking point, where the road environment image represents an image of a road where the screened parking point is located;

[0042] identify road data of the road environment image;

[0043] In response to the road data satisfying a preset road requirement, determine that the screened parking point is a target parking point.

[0044] In the embodiments of the present disclosure, by screening the parking points that meet the preset road network requirements from the suspected parking points, obtaining the screened parking points, obtaining the road environment image corresponding to the screened parking points, identifying the road data of the road environment image, and in response to the road data satisfying the preset road requirement, determining that the screened parking point is the target parking point, the suspected parking points can be verified, and the accuracy of the determined parking points can be improved.

[0045] Figure 1 is a flowchart of the method for determining a parking point provided by the embodiments of the present disclosure. Referring to Figure 1 , the method for determining a parking point provided by the embodiments of the present disclosure can include:

[0046] S101, obtaining a suspected parking point.

[0047] The suspected parking point can be understood as a parking point obtained through preliminary screening.

[0048] In one implementation manner, the suspected parking point can be manually annotated in the road network data according to experience, and the electronic device can obtain the suspected parking point annotated in the road network data from the road network data.

[0049] In another implementation manner, the suspected parking point can be obtained by mining historical vehicle driving trajectories. As Figure 2 shown, S101 can include:

[0050] S201, obtaining at least one vehicle driving trajectory.

[0051] The vehicle driving trajectory is a driving trajectory of a vehicle driving on a road.

[0052] Multiple user driving trajectories of vehicles driving on a road can be obtained. A user driving trajectory of a vehicle driving on a road at different times, etc. can also be obtained. A vehicle automatic driving trajectory on a road, etc. can also be obtained. The vehicle can be any type of vehicle, such as a truck, a small car, etc.

[0053] S202, determining multiple parking positions based on the at least one vehicle driving trajectory.

[0054] One vehicle driving trajectory can correspond to one parking position, or can correspond to multiple parking positions.

[0055] In an implementation manner, for each vehicle driving track, a deceleration behavior track in the vehicle driving track is detected, the deceleration behavior track being a track in which the speed decreases from a preset speed to 0; a position at which the speed is kept at 0 for more than a preset time in the deceleration behavior track is extracted, and the position is taken as a parking position corresponding to the vehicle driving track.

[0056] The preset speed can be determined according to actual needs or experience, for example, 30 kilometers per hour, 20 kilometers per hour, etc.

[0057] The preset time can also be determined according to actual needs or experience, for example, 9 minutes, 10 minutes, 15 minutes, etc.

[0058] For example, the vehicle driving track can be a truck driving track. A suspected parking deceleration behavior is found based on the time sequence speed feature of the truck driving track, such as decreasing from a speed of 20 kilometers per hour or more to 0, and the stopping time is longer than 10 minutes. The parking position with the stopping time longer than 10 minutes is taken as the position of this parking, that is, as the parking position of the truck driving track.

[0059] The suspected parking position can be conveniently obtained by detecting the deceleration behavior of the vehicle driving track, thereby providing a basis for the subsequent verification process of the parking position.

[0060] S203, clustering the plurality of parking positions to obtain a suspected parking position.

[0061] TrajDBSCAN based on trajectory clustering algorithm can be used for clustering. dbscan is a density-based clustering algorithm, and TrajDBSCAN can be understood as a modified algorithm of dbscan.

[0062] The distance between each parking position and other parking positions in the plurality of parking positions can be calculated; two parking positions with a distance less than a preset distance are taken as a position pair; for each position pair, the suspected parking position is selected as the parking position with a smaller position in the position pair, wherein and represent the sum of the grid points corresponding to the parking positions.

[0063] For example, the coordinate points, that is, the parking positions, can be grid processed, and all positions in each grid are subjected to spatial geometric calculation, and all position pairs within 5 meters are retained: id1, id2. Then all position pairs are cycled, positions with the same id are defined as a position group, position groups containing more than 10 different users are screened out, and finally the position point with the smallest sum of all points in each position group is selected and output.

[0064] Based on the vehicle driving track, only the parking position needs to be obtained, and the final parking point is not directly determined based on the vehicle driving track, without limiting the quality of the vehicle driving track, reducing the requirement for the accuracy and quality of the required vehicle driving track in the parking point determination process, and simplifying the calculation process. Moreover, the suspected parking points are clustered to obtain suspected parking points, which can reduce redundancy and reduce the calculation amount of subsequent verification. The clustering is based on multiple suspected parking points, which can improve the accuracy of the determined suspected parking points.

[0065] S102, screening parking points meeting the preset road network requirement from the suspected parking points to obtain screened parking points.

[0066] In simple terms, the obtained suspected parking points are verified by the road network.

[0067] For each suspected parking point, it is judged whether the suspected parking point meets the preset road network requirement. Specifically, it is judged whether the suspected parking point is a position that cannot be parked according to the preset road network requirement. In response to the suspected parking point meeting the preset road network requirement, the suspected parking point is retained, i.e. the suspected parking point is taken as a screened parking point. In response to the suspected parking point not meeting the preset road network requirement, the suspected parking point is deleted.

[0068] In one implementation manner, S102 can include:

[0069] It is judged whether the suspected parking point is located at a position marked in the road network data and unable to be parked according to the preset road network requirement.

[0070] In response to the suspected parking point being located at the position marked in the road network data and unable to be parked according to the preset road network requirement, the suspected parking point is taken as a parking point not meeting the preset road network requirement, and the parking point not meeting the preset road network requirement is deleted.

[0071] In response to the suspected parking point not being located at the position marked in the road network data and unable to be parked according to the preset road network requirement, the suspected parking point is taken as a parking point meeting the preset road network requirement, and the parking point meeting the preset road network requirement is taken as a screened parking point.

[0072] The suspected parking point can be projected to the coordinate system in which the road network data is located first, so that the suspected parking point and the road network data are in the same coordinate system, facilitating calculation. Then, it is judged whether the projection point corresponding to the suspected parking point is located at the position marked in the road network data and unable to be parked according to the preset road network requirement. The projection point represents the position of the suspected parking point projected to the coordinate system in which the road network data is located.

[0073] Among them, the position unable to be parked according to the preset road network requirement can be marked in the road network data in advance.

[0074] Pre-defined road network requirements may include laws and regulations, such as the Road Traffic Safety Law / Regulations for the Implementation of the Road Traffic Safety Law, etc.

[0075] Preset road network requirements for prohibited parking locations may include: locations within 3 meters of closed roads such as highways and urban expressways; road sections within 50 meters of intersections, railway crossings, bridges, and tunnels; road sections within 30 meters of bus stops, emergency stations, gas stations, fire hydrants, or fire station entrances; no-parking zones / yellow solid line sections, etc.

[0076] Road network verification can remove stop points that do not meet legal and regulatory requirements. This improves the compliance of initial stop point screening based on road network data and avoids violations. Verifying suspected stop points based on road network data can improve the accuracy of the final identified stop points.

[0077] S103, Obtain the road environment image corresponding to the selected stopping point.

[0078] Among them, the road environment image represents the image of the road where the stopping point is located.

[0079] It can collect and filter road environment images corresponding to stopping points.

[0080] In one possible implementation, such as Figure 3 As shown, S103 may include:

[0081] S301, crowdsourced collection of road environment images from multiple road locations.

[0082] For a given road location, the road environment image at that location represents the image of the road where that location is situated. A road location is simply a position within the road.

[0083] Road environment images of multiple road locations can be obtained from multiple datasets.

[0084] S302, compare the selected stopping point with multiple road locations to obtain road locations whose distance from the selected stopping point is within a preset distance range.

[0085] The coordinates of the selected stopping point are compared with the coordinates of each road location. Specifically, the distance between the selected stopping point and each road location can be calculated using methods such as cosine distance or Euclidean distance. By comparing the distances between the selected stopping point and multiple road locations, the coordinates within a preset distance range are obtained. The road locations corresponding to these preset distance ranges are the road locations within the preset distance range of the selected stopping point. The preset distance range can be determined based on actual needs or experience.

[0086] S303, acquire an image of a road position as a road environment image corresponding to the screening parking spot.

[0087] A plurality of road environment images of road positions can be obtained through crowdsourcing recycling, and the image of the road position within a preset distance range from the screening parking spot is provided as the road environment image corresponding to the screening parking spot, thereby providing a way to obtain a road environment image of a parking spot. For example, when the electronic device has no image acquisition function, the road environment image of the parking spot can also be obtained, and the parking spot is further verified based on the road environment image, thereby expanding the application range.

[0088] S104, identify road data of the road environment image.

[0089] S105, in response to the road data meeting a preset road requirement, determine the screening parking spot as a target parking spot.

[0090] The road data can include objects contained on the road where the screening parking spot is located.

[0091] The objects contained on the road can include at least one of the following: painted lines such as road boundary lines, road center lines, obstacles, moving targets such as motor vehicles, road surface conditions, and the like. The obstacles include barriers, fences, railings, isolation columns, water-filled, sand-filled isolation, conical ice cream tubes, and the like.

[0092] The road data contained in the road environment image can be identified through image recognition. In an implementation manner, the objects contained on the road where the screening parking spot is located corresponding to the road environment image can be identified through a semantic segmentation model. For example, a residual network (Residual Network, ResNet) such as ResNet50+Pyramid Scene Parsing Network (PSP) semantic segmentation model can be used for identification as shown in FIG. Figure 4A

[0093] S105 can include:

[0094] In response to the road data containing a parking area and the road data containing a target object of a preset category, the screening parking spot is determined as a target parking spot.

[0095] Whether the road data contains a parking area can be determined through painted lines such as road boundary lines, road center lines, and the like in the road data. For example, the road data in the road environment image corresponding to the screening parking spot is identified, and if the road data contains a landmark line indicating a parking area, it can be understood that the road data contains a parking area.

[0096] ​If the road data contains obstacles such as barriers, fences, railings, bollards, water injection, sand injection barriers such as Figure 4B As shown, and the like, the road data indicates that the road where the screening parking point is located is not a parking area. If the road data contains road conditions such as mud Figure 4C As shown, the road data indicates that the road where the screening parking point is located is not a parking area.

[0097] The target object of the preset category can be a vehicle, such as a truck, etc.

[0098] Whether the parking point in the field can be parked alone cannot be fully trusted by relying on the driver's behavior and road characteristics. By recycling images recycled from the crowd source, it is identified whether there are painted lines or trucks parked in the field, etc. Specifically, by using an image semantic segmentation model, whether there are motor vehicles, the type of motor vehicles, and whether there are motor vehicles such as trucks are identified. If there are motor vehicles such as trucks, it indicates that the parking point in the field can be parked. Whether it is identified by painted lines, the relationship with the road boundary, and the like to determine whether it is a parking point that actually allows parking.

[0099] Through image recognition, the screening parking point obtained by screening from the suspected parking point is further verified, and the accuracy of the determined parking point is further improved.

[0100] In the related art, truck detection data trajectory points are used to judge whether they meet the dynamics in terms of approach speed, angle or acceleration to find parking points. Then, the parking points in congestion and temporary parking are removed. The parking points are clustered in time and space, and the parking time is calculated. The parking point type is determined according to the trajectory characteristics. In the related art, the determination of the parking point is based on the trajectory characteristics. The accuracy of the trajectory and the degree of conformity with the actual road directly determine the accuracy of the determination of the parking point. The requirements for the pre-processing of the trajectory and the data accuracy are high. If the obtained trajectory is not accurate or does not meet the actual road requirements, the accuracy of the parking point obtained by processing the trajectory cannot be guaranteed, that is, it is difficult to verify the mined parking point.

[0101] The method for determining a parking point provided by the embodiments of the present disclosure utilizes the Geographic Information System (GIS) feature to integrate image recognition and trajectory data mining methods, which not only reduces the requirements for the accuracy and quality of the trajectory, but also identifies whether it is a real and compliant parking point through image recognition, thereby improving the accuracy of data output.

[0102] Figure 5 is a schematic diagram of applying the method for determining a parking point provided by the embodiments of the present disclosure. Specifically, it can include vehicle driving trajectory, clustering output, road network verification, image verification, and generation of a parking point layer.

[0103] Vehicle driving trajectory:

[0104] Obtaining at least one vehicle driving track, the vehicle driving track being a driving track of a vehicle driving on a road, and specifically being a truck driving track.

[0105] Clustering the output:

[0106] For each vehicle driving track, detecting a deceleration behavior track in the vehicle driving track, the deceleration behavior track being a track in which the speed decreases from a preset speed to 0; extracting a position at which the speed remains 0 for more than a preset time in the deceleration behavior track, and taking the position as a parking position corresponding to the vehicle driving track. That is, a plurality of parking positions are determined based on the at least one vehicle driving track.

[0107] Clustering the plurality of parking positions to obtain suspected parking points.

[0108] The distance between each parking position and other parking positions in the plurality of parking positions can be calculated; two parking positions with a distance less than a preset distance are taken as a position pair; for each position pair, the smaller parking position in the position pair is taken as a suspected parking point.

[0109] TrajDBSCAN can be used for clustering to obtain suspected parking points.

[0110] Road network verification:

[0111] Filtering parking points that meet preset road network requirements from the suspected parking points to obtain filtered parking points.

[0112] For each suspected parking point, it is determined whether the suspected parking point meets the preset road network requirements. Specifically, it is determined whether the suspected parking point is a position that cannot be parked according to the preset road network requirements. In response to the suspected parking point meeting the preset road network requirements, the suspected parking point is retained, i.e., the suspected parking point is taken as a filtered parking point. In response to the suspected parking point not meeting the preset road network requirements, the suspected parking point is deleted.

[0113] After obtaining suspected parking points by clustering based on TrajDBSCAN, the road network verification of the suspected parking points can improve the compliance of the road network data preliminary screening parking points and avoid illegal behavior.

[0114] Image verification:

[0115] Obtaining a road environment image corresponding to the filtered parking point; identifying road data of the road environment image; and in response to the road data meeting preset road requirements, determining the filtered parking point as a target parking point.

[0116] The road data contained in the road environment image can be identified through image recognition. In an implementation manner, the objects contained on the road where the screening stop point corresponding to the road environment image is located can be identified through a semantic segmentation model.

[0117] In response to the road data indicating that the road where the screening stop point is located is a parking area, and the road data containing a target object of a preset category, the screening stop point is determined as a target stop point.

[0118] As shown in Figure 6 , for the screening stop point, the road environment image corresponding to the screening stop point is obtained, the road environment image is subjected to semantic segmentation, and the road data is obtained, such as the objects contained on the road where the screening stop point corresponding to the road environment image is located. It is identified whether there is a parking area. If there is a parking area, it is continued to judge whether the road environment image corresponding to the screening stop point contains a target object; if there is no parking area, the screening stop point cannot be verified, that is, the screening stop point is removed, and the screening stop point is not taken as a target stop point.

[0119] After it is judged that there is a parking area, it is judged that the road environment image corresponding to the screening stop point contains a target object, and the screening stop point is a target stop point, that is, a verified stop point, that is, a verified stop point. After it is judged that there is no parking area, it is judged that the road environment image corresponding to the screening stop point does not contain a target object, and the screening stop point cannot be verified, that is, the screening stop point is removed, and the screening stop point is not taken as a target stop point.

[0120] The image is collected through image recognition, and the road data is identified. The road data can represent the road type, such as whether the road has a road surface, that is, the road state and obstacles, to prevent false fusion of special scenes such as construction roads, ferries, and cableways, and to improve the accuracy of stop point determination.

[0121] A stop point layer is generated.

[0122] A final stop point, that is, a target stop point, is generated.

[0123] The system including a road network compilation cross-section module, a user trajectory data acquisition module, a road trajectory center line calculation module, and a road form change mining module can be used to determine the stop point.

[0124] The road network data can be acquired by the road network compilation cross-section module; the vehicle driving trajectory can be acquired by the user trajectory data acquisition module; and the road trajectory center line calculation module and the road form change mining module can further verify the screening stop point after the road network verification based on the road environment data.

[0125] This embodiment first identifies potential stopping points by detecting the deceleration behavior of vehicle trajectories, such as those of trucks. Then, these potential stopping points are aggregated. Next, road network verification is used to remove stopping points that do not comply with legal regulations. For example, stopping points are projected onto the road network data to ensure they are in the same coordinate system, thus determining whether a stopping point is a location where stopping is prohibited by law. If so, it is removed; otherwise, it is retained. Multiple road environment images of various road locations are collected from multiple sources, and the stopping points are compared with these locations to identify those closest to the stopping point (within a preset distance). The image of this road location is the road environment image corresponding to the stopping point. Finally, image recognition is used to verify whether there are road markings or actual trucks parked in the road environment image corresponding to the stopping point, thus verifying the stopping point. This approach reduces the requirements for trajectory accuracy and quality while improving the accuracy of identified stopping points.

[0126] This disclosure also provides an apparatus for determining a stopping point, such as... Figure 7 As shown, it may include:

[0127] Module 701 is used to obtain suspected stopping points;

[0128] The filtering module 702 is used to filter out stopping points that meet the preset road network requirements from the suspected stopping points to obtain the filtered stopping points;

[0129] The acquisition module 703 is used to acquire the road environment image corresponding to the selected stopping point, wherein the road environment image represents the image of the road where the selected stopping point is located;

[0130] Recognition module 704 is used to recognize road data in road environment images;

[0131] The determination module 705 is used to determine the selected stopping point as the target stopping point in response to the road data meeting the preset road requirements.

[0132] Optionally, module 701 is specifically used to acquire at least one vehicle driving trajectory; determine multiple parking locations based on the at least one vehicle driving trajectory; and cluster the multiple parking locations to obtain suspected parking points.

[0133] Optionally, module 701 is specifically used to detect deceleration behavior trajectories in the driving trajectories of each vehicle, where the deceleration behavior trajectory is the trajectory where the speed decelerates from a preset speed to 0; extract the positions in the deceleration behavior trajectory where the speed remains at 0 for more than a preset time, and use the positions as the stopping positions corresponding to the vehicle driving trajectory.

[0134] Optionally, the screening module 702 is specifically configured to determine whether the suspected parking point is located at a position marked in the road network data and unable to park according to the preset road network requirement; in response to the suspected parking point being located at the position marked in the road network data and unable to park according to the preset road network requirement, the suspected parking point is determined as a parking point not meeting the preset road network requirement, and the parking point not meeting the preset road network requirement is deleted; in response to the suspected parking point not being located at the position marked in the road network data and unable to park according to the preset road network requirement, the suspected parking point is determined as a parking point meeting the preset road network requirement, and the parking point meeting the preset road network requirement is determined as a screening parking point.

[0135] Optionally, the acquisition module 703 is specifically configured to acquire road environment images of a plurality of road positions by crowdsourcing; compare the screening parking point with the plurality of road positions to obtain road positions having a distance from the screening parking point within a preset distance range; and acquire an image of the road position as a road environment image corresponding to the screening parking point.

[0136] Optionally, the determination module 705 is specifically configured to determine the screening parking point as a target parking point in response to the road data containing a parking area and the road data containing a target object of a preset category.

[0137] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in the technical solution comply with relevant laws and regulations and do not violate public order and good customs.

[0138] According to the embodiments of the present disclosure, the present disclosure further provides an electronic device, a readable storage medium and a computer program product.

[0139] Figure 8 A schematic block diagram of an example electronic device 800 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit the implementations of the present disclosure described and / or claimed in this document.

[0140] As Figure 8As shown, the electronic device 800 includes a computing unit 801 that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the device 800 can also be stored. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other through a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0141] A plurality of components in the device 800 are connected to the I / O interface 805, including: an input unit 806, such as a keyboard, a mouse, and the like; an output unit 807, such as various types of displays, speakers, and the like; a storage unit 808, such as a magnetic disk, an optical disk, and the like; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, and the like. The communication unit 809 allows the device 800 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0142] The computing unit 801 can be various general-purpose and / or special-purpose processing components having processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, and the like. The computing unit 801 performs various methods and processes described above, such as the method of determining a stopping point. For example, in some embodiments, the method of determining a stopping point can be implemented as a computer software program that is tangibly embodied in a machine-readable medium, such as the storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed on the device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded into the RAM 803 and executed by the computing unit 801, one or more steps of the method of determining a stopping point described above can be performed. Alternatively, in other embodiments, the computing unit 801 can be configured to perform the method of determining a stopping point by any other appropriate means, such as by means of firmware.

[0143] The various embodiments of the systems and techniques described above can be implemented in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a complex programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0144] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces the functions / operations specified in the flowcharts and / or the block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a stand-alone software package, or entirely on a remote machine or server.

[0145] In the context of the present disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium will include one or more lines of electrical connections, portable computer disks, hard disk drives, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fibers, portable compact disc read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0146] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0147] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0148] The computer system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server is generally established by computer programs running on the respective computers and having a client-server relationship to each other. The servers can be cloud servers, servers of a distributed system, or servers combined with a blockchain.

[0149] It should be understood that various forms of flow shown above can be used, with steps reordered, added, or removed. For example, the steps recited in the present disclosure can be performed in parallel, in series, or in a different order, without limitation herein, so long as the desired results of the technology disclosed in the present disclosure are achieved.

[0150] The specific embodiments described above are not intended to limit the scope of the present disclosure. Those skilled in the art will understand that various modifications, combinations, sub-combinations, and alternatives can be made to the specific embodiments without departing from the spirit and principles of the present disclosure. Any further modifications, equivalent substitutions, improvements, and the like, either presently known or later developed, that do not depart from the spirit and principles of the present disclosure are to be encompassed within the scope of the present disclosure.

Claims

1. A method for determining a parking point, comprising: obtaining at least one vehicle driving trajectory; determining a plurality of parking positions based on the at least one vehicle driving trajectory; clustering the plurality of parking positions to obtain suspected parking points, including: calculating the distance between each parking position and other parking positions in the plurality of parking positions; regarding two parking positions with a distance less than a preset distance as a position pair; for each position pair, selecting the suspected parking point as the parking position with a smaller position in the position pair and the sum of the grid points corresponding to the parking position pair; wherein, and represent the sum of the grid points; screening parking points from the suspected parking points that meet a preset road network requirement to obtain screened parking points; obtaining a road environment image corresponding to the screened parking point, wherein the road environment image represents an image of a road where the screened parking point is located; recognizing road data of the road environment image through image recognition, the road data including objects contained on the road where the screened parking point is located, and the recognizing road data of the road environment image through image recognition includes: identifying the objects contained on the road where the screened parking point is located through a semantic segmentation model; determining whether the road data contains a parking area through marking lines in the road data; in response to the road data containing a parking area and the road data containing a target object of a preset category, determining that the screened parking point is a target parking point, and the target object of the preset category includes a vehicle.

2. The method of claim 1, wherein, The determining a plurality of parking positions based on the at least one vehicle driving trajectory includes: for each vehicle driving trajectory, detecting a deceleration behavior trajectory in the vehicle driving trajectory, the deceleration behavior trajectory being a trajectory in which the speed decreases from a preset speed to 0; extracting a position in the deceleration behavior trajectory where the speed remains 0 for more than a preset time, and regarding the position as a parking position corresponding to the vehicle driving trajectory.

3. The method of claim 1 or 2, wherein, The screening parking points from the suspected parking points that meet a preset road network requirement to obtain screened parking points includes: determining whether the suspected parking point is located at a position marked in the road network data and at which the preset road network requirement cannot be met; in response to the suspected parking point being located at the position marked in the road network data and at which the preset road network requirement cannot be met, regarding the suspected parking point as a parking point that does not meet the preset road network requirement, and deleting the parking point that does not meet the preset road network requirement; in response to the suspected parking point not being located at the position marked in the road network data and at which the preset road network requirement cannot be met, regarding the suspected parking point as a parking point that meets the preset road network requirement, and regarding the parking point that meets the preset road network requirement as a screened parking point.

4. The method of claim 1 or 2, wherein, The obtaining a road environment image corresponding to the screened parking point includes: crowdsourcing road environment images of a plurality of road positions; comparing the screened parking point with the plurality of road positions to obtain road positions with a distance from the screened parking point within a preset distance range; obtaining an image of the road position as the road environment image corresponding to the screened parking point.

5. An apparatus for determining a parking point, comprising: an obtaining module configured to obtain at least one vehicle driving trajectory; determine a plurality of parking positions based on the at least one vehicle driving track; cluster the plurality of parking positions to obtain suspected parking points, including: calculating the distance between each parking position and other parking positions in the plurality of parking positions; regarding two parking positions with a distance less than a preset distance as a position pair; for each position pair, selecting the suspected parking point as the position pair midpoint and the parking position with a smaller position, wherein, and represent the sum of the grid points corresponding to the parking position pair; a screening module configured to screen the suspected parking points to obtain screened parking points that meet a preset road network requirement; an acquisition module configured to acquire a road environment image corresponding to the screened parking point, wherein the road environment image represents an image of a road where the screened parking point is located; an identification module configured to identify road data of the road environment image through image recognition, wherein the road data includes objects contained on the road where the screened parking point is located, and the identification of the road data of the road environment image through image recognition includes identifying the objects contained on the road where the screened parking point is located through a semantic segmentation model; a determination module configured to determine whether the road data contains a parking area through marking lines in the road data, and determine the screened parking point as a target parking point in response to the road data containing a parking area and the road data containing a target object of a preset category, wherein the target object of the preset category includes a vehicle.

6. The apparatus of claim 5, wherein, The acquisition module is specifically configured to detect a deceleration behavior track in the vehicle driving track for each vehicle driving track, the deceleration behavior track being a track in which the speed decreases from a preset speed to 0; extract a position where the speed remains 0 for more than a preset time in the deceleration behavior track, and take the position as a parking position corresponding to the vehicle driving track.

7. The apparatus of claim 5 or 6, wherein, The screening module is specifically configured to determine whether the suspected parking point is located at a position marked in the road network data and at which the preset road network requirement cannot be met, and take the suspected parking point as a parking point that does not meet the preset road network requirement and delete the parking point that does not meet the preset road network requirement in response to the suspected parking point being located at the position. The acquisition module is specifically configured to acquire road environment images of a plurality of road positions from a crowd source, compare the screened parking point with the plurality of road positions to obtain road positions with a distance from the screened parking point within a preset distance range, and acquire an image of the road position as the road environment image corresponding to the screened parking point.

8. The apparatus of claim 5 or 6, wherein, 9. An electronic device, comprising: at least one processor; and a memory connected to the at least one processor in communication; wherein, ​ The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-4.

10. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are for causing the computer to perform the method of any one of claims 1-4.

11. A computer program product comprising a computer program which, when executed by a processor, implements the method of any one of claims 1-4.

Images