A high-precision map intersection vectorization labeling method, device, equipment and medium

Abstract

The application discloses a high-precision map intersection vectorization labeling method and device, equipment and medium, relates to the field of artificial intelligence, and particularly relates to the field of intelligent transportation and automatic driving. The specific implementation scheme is as follows: a pre-acquired trajectory is processed to obtain a trajectory corresponding to each road; based on the trajectory corresponding to each road, a center point of each intersection is identified; according to the center point of each intersection and the trajectory corresponding to each road, a road trunk corresponding to each intersection is extracted; and based on the center point of each intersection and the road trunk corresponding to each intersection, each intersection is vectorized and labeled. The embodiment of the application can vectorize and label intersections, automatically complete the labeling task, reduce manual labeling investment, and thus can greatly improve the production efficiency of a precision map.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of artificial intelligence, further relates to the fields of intelligent transportation and autonomous driving, and in particular to a high-definition map intersection vectorization labeling method, device, equipment and medium. BACKGROUND

[0002] High-definition maps play an important role in autonomous vehicles. The current data production method of high-definition maps is mainly through professional collection vehicles to collect various sensor data of road information, and then generate basic data through automatic algorithm and manual labeling. The main sensor data currently available includes inertial measurement unit (IMU) trajectory data, laser point cloud data, high-speed camera image data, etc. Traditional labeling based on raw data has been widely used in production lines. By developing automatic labeling algorithms, the work time of labeling personnel can be greatly saved, labeling efficiency can be improved, and production costs can be reduced.

[0003] In the existing technical solution, raw data is processed to generate visual labeling data, including two-dimensional point cloud view or three-dimensional point cloud view, image view, trajectory data, etc. Labeling personnel refer to these data on customized labeling tools to label intersection vectorization data. The main elements of the labeling include lane line geometry, lane line topology, lane line attribute, stop line, pedestrian crossing, etc. The above solution mainly has the following two problems: 1) low labeling efficiency, labeling personnel need to complete the generation of the above elements manually from scratch; 2) high quality requirements for labeling personnel, high-definition maps have certain requirements for data accuracy, and therefore labeling personnel need to be trained for a long time to be competent for this work, increasing the personnel investment cost. SUMMARY

[0004] The present disclosure provides a high-definition map intersection vectorization labeling method, device, equipment and medium.

[0005] In a first aspect, the present disclosure provides a high-definition map intersection vectorization labeling method, which comprises:

[0006] processing the pre-acquired trajectories to obtain trajectories corresponding to each road;

[0007] identifying the center points of each intersection based on the trajectories corresponding to each road;

[0008] extracting the road trunks corresponding to each intersection according to the center points of each intersection and the trajectories corresponding to each road;

[0009] The center points of the intersections are vectorized and labeled based on the center points of the intersections and the road trunks corresponding to the intersections.

[0010] In a second aspect, the application provides a high-definition map intersection vectorization labeling device, which comprises a processing module, an identification module, an extraction module and a labeling module, wherein

[0011] The processing module is configured to process the pre-acquired trajectories to obtain trajectories corresponding to each road.

[0012] The identification module is configured to identify the center points of the intersections based on the trajectories corresponding to each road.

[0013] The extraction module is configured to extract road trunks corresponding to each intersection according to the center points of the intersections and the trajectories corresponding to each road.

[0014] The labeling module is configured to vectorize and label each intersection based on the center points of the intersections and the road trunks corresponding to the intersections.

[0015] In a third aspect, the application provides an electronic device, which comprises:

[0016] one or more processors;

[0017] a memory configured to store one or more programs,

[0018] When the one or more programs are executed by the one or more processors, the one or more processors implement the high-definition map intersection vectorization labeling method according to any of the embodiments of the application.

[0019] In a fourth aspect, the application provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the high-definition map intersection vectorization labeling method according to any of the embodiments of the application.

[0020] In a fifth aspect, a computer program product is provided, which, when executed by a computer device, implements the high-definition map intersection vectorization labeling method according to any of the embodiments of the application.

[0021] The technical solution of the application solves the technical problems of low labeling efficiency, the need for labeling personnel to manually generate each element from scratch, high quality requirements for labeling personnel, the need for long-term training of labeling personnel to perform the work, and increased personnel investment costs in the prior art. The technical solution provided by the application can vectorize and label intersections, automatically complete labeling tasks, reduce manual labeling investment, and thus greatly improve the production efficiency of high-definition maps.

[0022] 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

[0023] The accompanying drawings are used to better understand the present scheme and do not limit the present disclosure. Among them:

[0024] Figure 1 is the first flowchart of the high-precision map intersection vectorization labeling method provided by the embodiment of the present application;

[0025] Figure 2 is the second flowchart of the high-precision map intersection vectorization labeling method provided by the embodiment of the present application;

[0026] Figure 3 is the third flowchart of the high-precision map intersection vectorization labeling method provided by the embodiment of the present application;

[0027] Figure 4 is the structure diagram of the high-precision map intersection vectorization labeling device provided by the embodiment of the present application;

[0028] Figure 5 is the block diagram of the electronic device for implementing the high-precision map intersection vectorization labeling method of the embodiment of the present application. DETAILED DESCRIPTION

[0029] The exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to help understanding, which should be considered only as exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Also, in order to be clear and concise, the description below omits the description of well-known functions and structures.

[0030] Embodiment one

[0031] Figure 1 is the first flowchart of the high-precision map intersection vectorization labeling method provided by the embodiment of the present application, which can be executed by a high-precision map intersection vectorization labeling device or an electronic device. The device or the electronic device can be realized by software and / or hardware, and the device or the electronic device can be integrated in any smart device with network communication function. As shown in Figure 1 The high-precision map intersection vectorization labeling method can include the following steps:

[0032] S101, process the pre-acquired trajectory to obtain the trajectory corresponding to each road.

[0033] In this step, the electronic device can process the pre-acquired trajectories to obtain trajectories corresponding to each road. Specifically, the electronic device can first match each two trajectories in the pre-acquired trajectories to obtain a matching result of each two trajectories; and then determine the trajectories corresponding to each road based on the matching result of each two trajectories.

[0034] S102, identify the center points of each intersection based on the trajectories corresponding to each road.

[0035] In this step, the electronic device can identify the center points of each intersection based on the trajectories corresponding to each road. Specifically, the electronic device can first calculate the same direction trajectory matching pairs in each two trajectories based on the trajectories corresponding to each road according to a same direction trajectory point calculation formula; then calculate the crossing trajectory matching pairs in each two trajectories based on the trajectories corresponding to each road according to a crossing trajectory point calculation formula; and then identify the center points of each intersection based on the same direction trajectory matching pairs and the crossing trajectory matching pairs.

[0036] S103, extract the road trunks corresponding to each intersection according to the center points of each intersection and the trajectories corresponding to each road.

[0037] In this step, the electronic device can extract the road trunks corresponding to each intersection according to the center points of each intersection and the trajectories corresponding to each road. Specifically, the electronic device can search for trajectory points around each intersection within a predetermined radius with the center point of each intersection as the center; and if at least one trajectory point around each intersection is searched within the predetermined radius, the electronic device can generate the road trunk of each intersection based on each trajectory point in the at least one trajectory point around each intersection.

[0038] S104, vectorize and label each intersection based on the center points of each intersection and the road trunks corresponding to each intersection.

[0039] In this step, the electronic device can vectorize and label each intersection based on the center points of each intersection and the road trunks corresponding to each intersection; wherein the road trunk can include an out-intersection trunk and an in-intersection trunk. Specifically, the electronic device can draw a corresponding ray as the road trunk of each intersection with the center point of each intersection as the ray coordinate and the road trunk corresponding to each intersection as the ray direction.

[0040] The high-precision map intersection vectorization labeling method provided in the embodiments of the present application first processes the pre-acquired trajectories to obtain trajectories corresponding to each road; then identifies the center points of each intersection based on the trajectories corresponding to each road; then extracts the road trunks corresponding to each intersection according to the center points of each intersection and the trajectories corresponding to each road; and finally vectorizes and labels each intersection based on the center points of each intersection and the road trunks corresponding to each intersection. That is, the present application can realize vectorization labeling of intersections through trajectory processing, intersection identification, and road trunk generation. In the existing high-precision map intersection vectorization labeling method, the original data is first processed to generate visual labeling data; then the labeling personnel label the intersection vectorization data on the customized labeling tool by referring to these data. Because the present application adopts the technical means of trajectory processing, intersection identification, and road trunk generation, it overcomes the technical problems of low labeling efficiency, the labeling personnel needing to complete the generation operation of each element manually from scratch, high quality requirements for labeling personnel, certain requirements for data precision of high-precision maps, and the need for long-term training of labeling personnel to be competent for the work, which increases the personnel investment cost. The technical solution provided in the present application can vectorize and label intersections, automatically complete the labeling task, reduce manual labeling investment, and thus can greatly improve the production efficiency of the precision map. Moreover, the technical solution of the embodiments of the present application is simple and convenient to implement, easy to popularize, and has a wider application range.

[0041] Embodiment Two

[0042] Figure 2 FIG. 2 is a second flowchart of the high-precision map intersection vectorization labeling method provided in the embodiments of the present application. Based on the above technical solution, further optimization and expansion can be performed, and the technical solution can be combined with the above various optional embodiments. As shown in FIG. 2, the high-precision map intersection vectorization labeling method can include the following steps: Figure 2

[0043] S201, match each two trajectories in the pre-acquired trajectories to obtain a matching result of each two trajectories.

[0044] ​In this step, the electronic device can match each two trajectories in the pre-acquired trajectories to obtain a matching result of each two trajectories. Specifically, the electronic device can first randomly extract two trajectories in the pre-acquired trajectories as a first trajectory and a second trajectory respectively; then for each trajectory point in the first trajectory, search for other trajectory points in the first trajectory that have an angle within a threshold range with the each trajectory point; and then based on the each trajectory point and the other trajectory points, calculate a nearest point of the each trajectory point to the second trajectory; take the each trajectory point and the nearest point as the matching result of the first trajectory and the second trajectory; and repeat the above operations until the matching result of each two trajectories is obtained. In specific embodiments of the present application, the electronic device can calculate the nearest point according to the following formula:

[0045] d curr <d prev &&d curr <d next

[0046] In the above formula, d curr is the distance from the target point to the current trajectory point; d prev is the distance from the target point to the predecessor (prev) of the current trajectory point, and d next is the distance from the target point to the successor (next) of the current trajectory point. After filtering, the target point of MIN{d curr} is taken as the nearest point.

[0047] S202, determining the trajectory corresponding to each road based on the matching result of each two trajectories.

[0048] In this step, the electronic device can determine the trajectory corresponding to each road based on the matching result of each two trajectories. For example, if the distance between any one trajectory point on the first trajectory and the nearest point of the second trajectory in the two trajectories is less than a preset distance, the two trajectories can be determined as the trajectories corresponding to the same road; otherwise, the two trajectories can be determined as the trajectories corresponding to different roads.

[0049] S203, identifying the center point of each intersection based on the trajectory corresponding to each road.

[0050] In this step, the electronic device can identify the center point of each intersection based on the trajectory corresponding to each road. Specifically, the electronic device can first calculate a same-direction trajectory matching pair in each two trajectories according to a same-direction trajectory point calculation formula based on the trajectory corresponding to each road; then calculate a crossing trajectory matching pair in each two trajectories according to a crossing trajectory point calculation formula based on the trajectory corresponding to each road; and then identify the center point of each intersection based on the same-direction trajectory matching pair and the crossing trajectory matching pair.

[0051] S204, extracting road trunks corresponding to each intersection according to the center point of each intersection and the trajectory corresponding to each road.

[0052] In this step, the electronic device can extract road trunks corresponding to each intersection according to the center point of each intersection and the trajectory corresponding to each road. Specifically, the electronic device can take the center point of each intersection as the center of a circle, search for trajectory points around each intersection within a predetermined radius, and if at least one trajectory point around each intersection is searched within the predetermined radius, generate a road trunk of each intersection based on each trajectory point in the at least one trajectory point around each intersection. Further, the electronic device can extract one trajectory point from the at least one trajectory point around each intersection as a current trajectory point, determine whether the other trajectory point in the same direction trajectory matching pair containing the current trajectory point has been marked, and if the other trajectory point is not marked, connect the current trajectory point and the intersection corresponding thereto as the road trunk of each intersection relative to the current trajectory point. The above operations are repeated until the road trunk of each intersection is obtained.

[0053] S205, vectorizing and labeling each intersection based on the center point of each intersection and the road trunk corresponding to each intersection.

[0054] The high-definition map intersection vectorization labeling method provided in the embodiments of the present application first processes the pre-acquired trajectory to obtain the trajectory corresponding to each road, then identifies the center point of each intersection based on the trajectory corresponding to each road, extracts the road trunk corresponding to each intersection according to the center point of each intersection and the trajectory corresponding to each road, and finally vectorizes and labels each intersection based on the center point of each intersection and the road trunk corresponding to each intersection. That is, the present application can realize vectorization labeling of intersections through trajectory processing, intersection identification, and road trunk generation. In the existing high-definition map intersection vectorization labeling method, the original data is first processed to generate visual labeling data, and then the labeling personnel label the intersection vectorization data on the customized labeling tool according to the data. Because the present application adopts the technical means of trajectory processing, intersection identification, and road trunk generation, it overcomes the technical problems of low labeling efficiency, the need for labeling personnel to manually complete the generation of each element from scratch, high requirements for labeling personnel, the need for long-term training of labeling personnel to meet the data precision requirements of high-definition maps, and the increase in personnel investment costs. The technical solution provided in the present application can vectorize and label intersections, automatically complete labeling tasks, reduce manual labeling investment, and thus greatly improve the production efficiency of high-definition maps. Furthermore, the technical solution of the embodiments of the present application is simple and convenient to implement, easy to popularize, and has a wider application range.

[0055] Embodiment Three

[0056] Figure 3 FIG. 3 is a third flowchart illustrating a method for vectorizing intersection labeling of a high-definition map according to an embodiment of the present disclosure. The method for vectorizing intersection labeling of a high-definition map can include the following steps: Figure 3

[0057] S301, match each two trajectories in the pre-acquired trajectories to obtain a matching result of each two trajectories.

[0058] S302, determine a trajectory corresponding to each road based on the matching result of each two trajectories.

[0059] S303, calculate a same-direction trajectory matching pair in each two trajectories according to a same-direction trajectory point calculation formula based on the trajectory corresponding to each road.

[0060] In this step, the electronic device can calculate a same-direction trajectory matching pair in each two trajectories according to a same-direction trajectory point calculation formula based on the trajectory corresponding to each road. Specifically, the electronic device can calculate a same-direction trajectory matching pair in each two trajectories according to the following same-direction trajectory point calculation formula, and then based on the trajectory corresponding to each road:

[0061]

[0062] In the above formula, p a = 1 / (lambda1+d) represents a Euclidean distance coefficient, d is the Euclidean distance of the two trajectory points in the matching pair; p b = 1 / (lambda1+theta) represents an angle coefficient, theta is the included angle of the driving directions of the two trajectory points; p c = 1 / (lambda1+t) represents a height difference, t is the height difference of the two trajectory points; if the calculated Score same is greater than a given same-direction score threshold, the two trajectory points are marked as a same-direction trajectory matching pair.

[0063] S304, calculate a crossing trajectory matching pair in each two trajectories according to a crossing trajectory point calculation formula based on the trajectory corresponding to each road.

[0064] In this step, the electronic device can calculate a crossing trajectory matching pair in each two trajectories according to a crossing trajectory point calculation formula based on the trajectory corresponding to each road. Specifically, the electronic device can calculate a crossing trajectory matching pair in each two trajectories according to the following crossing trajectory point calculation formula, and then based on the trajectory corresponding to each road:

[0065] ​

[0066] In the above formula, p s = 1 / (lambda2+d), represents the Euclidean distance coefficient, d is the Euclidean distance of the two trajectory points in the matching pair; p e = 1 / (lambda2+theta), represents the angle coefficient, theta is the included angle of the driving directions of the two trajectory points; p f = 1 / (lambda2+t), represents the height difference, t is the height difference of the two trajectory points; if the calculated Score cross is greater than the given intersection score threshold, the two trajectory points are marked as an intersection trajectory matching pair.

[0067] S305, based on the same direction trajectory matching pair and the intersection trajectory matching pair, identifying the center point of each intersection.

[0068] In this step, the electronic device can identify the center point of each intersection based on the same direction trajectory matching pair and the intersection trajectory matching pair. Specifically, according to the intersection trajectory matching pair identified above, the intersection point is calculated, and the calculation method is as follows:

[0069] Position cross = get_cross(Pos1, Pos2, dir1, dir2)

[0070] In the above formula, get_cross is a function for calculating the intersection of rays, Pos1 and Pos2 are ray coordinates, and dir1 and dir2 are ray directions.

[0071] After the intersection point is identified, each valid intersection matching pair will generate a position point, and the position points are clustered by Euclidean distance to obtain intersection center clustering points, and each clustering point represents an intersection.

[0072] S306, taking the center point of each intersection as the center, searching for trajectory points around each intersection within a predetermined radius.

[0073] S307, if at least one trajectory point around each intersection is searched within a predetermined radius, generating a road trunk of each intersection based on each trajectory point in the at least one trajectory point around each intersection.

[0074] In the specific embodiments of the application, the electronic device can take the center point of each intersection as the center of a circle, search for trajectory points around each intersection within a predetermined radius, and if at least one trajectory point around each intersection is searched within the predetermined radius, generate the road trunk of each intersection based on each trajectory point of the at least one trajectory point around each intersection. Specifically, the electronic device can take the identified intersection center point as the center of a circle, give a radius d, obtain trajectory point data using spatial indexing, and generate the trunk road for each obtained trajectory point. Next, cross-intersection trunk recognition is performed, a threshold is given, each intersection center point is matched, the matched other intersection points are connected, and vectorization is completed. The generated trunk road connection and the cross-intersection connection are sampled at a given interval length d, and virtual trajectory points are generated. The final vectorization process is completed.

[0075] The intersection vectorization labeling method of the high-definition map provided in the embodiments of the application first processes the pre-acquired trajectories to obtain trajectories corresponding to each road, then identifies the center points of each intersection based on the trajectories corresponding to each road, extracts the road trunks corresponding to each intersection according to the center points of each intersection and the trajectories corresponding to each road, and finally labels each intersection based on the center points of each intersection and the road trunks corresponding to each intersection. That is, the application can realize the vectorization labeling of intersections through trajectory processing, intersection identification, and road trunk generation. In the existing intersection vectorization labeling method of high-definition maps, the original data is first processed to generate visual labeling data, and then the labeling personnel labels the intersection vectorization data on the customized labeling tool by referring to the data. Because the application adopts the technical means of trajectory processing, intersection identification, and road trunk generation, it overcomes the technical problems of low labeling efficiency, the need for labeling personnel to manually complete the generation of each element from scratch, high requirements for the quality of labeling personnel, the need for long-term training of labeling personnel to perform the work, and the increase in personnel investment costs. The technical solution provided in the application can vectorize and label intersections, automatically complete the labeling task, reduce manual labeling investment, and thus greatly improve the production efficiency of high-definition maps. Moreover, the technical solution of the embodiments of the application is simple and convenient to implement, easy to popularize, and suitable for a wider range of applications.

[0076] Embodiment Four

[0077] Figure 4 is a structural schematic diagram of the intersection vectorization labeling device of the high-definition map provided in the embodiments of the application. As shown in Figure 4 the device 400 includes a processing module 401, an identification module 402, an extraction module 403, and a labeling module 404. The processing module 401 is configured to process pre-acquired trajectories to obtain trajectories corresponding to each road.

[0078] The processing module 401 is configured to process the pre-acquired trajectories to obtain trajectories corresponding to each road.

[0079] The identification module 402 is configured to identify the center points of each intersection based on the trajectories corresponding to each road.

[0080] The extraction module 403 is configured to extract the road trunks corresponding to each intersection according to the center points of each intersection and the trajectories corresponding to each road.

[0081] The labeling module 404 is configured to vectorize label each intersection based on the center points of each intersection and the road trunks corresponding to each intersection.

[0082] Further, the processing module 401 is specifically configured to match each two trajectories in the pre-acquired trajectories to obtain matching results of each two trajectories, and determine the trajectories corresponding to each road based on the matching results of each two trajectories.

[0083] Further, the processing module 401 is specifically configured to randomly extract two trajectories in the pre-acquired trajectories as a first trajectory and a second trajectory, search for other trajectory points in the first trajectory that have an angle within a threshold range with each trajectory point in the first trajectory, calculate the nearest point of each trajectory point and the second trajectory based on each trajectory point and the other trajectory points, take each trajectory point and the nearest point as the matching results of the first trajectory and the second trajectory, and repeat the above operations until the matching results of each two trajectories are obtained.

[0084] Further, the identification module 402 is specifically configured to calculate same-direction trajectory matching pairs in each two trajectories according to a same-direction trajectory point calculation formula based on the trajectories corresponding to each road, calculate crossing trajectory matching pairs in each two trajectories according to a crossing trajectory point calculation formula based on the trajectories corresponding to each road, and identify the center points of each intersection based on the same-direction trajectory matching pairs and the crossing trajectory matching pairs.

[0085] Further, the extraction module 403 is specifically configured to search for trajectory points around each intersection within a predetermined radius with the center point of each intersection as the center, and generate the road trunk of each intersection based on each trajectory point in at least one trajectory point around each intersection if the at least one trajectory point around each intersection is searched within the predetermined radius.

[0086] Further, the extraction module 403 is specifically configured to extract one trajectory point from at least one trajectory point around each intersection as a current trajectory point; determine whether another trajectory point in a same-direction trajectory matching pair containing the current trajectory point has been marked; if the another trajectory point has not been marked, connect the current trajectory point and a corresponding intersection of the current trajectory point as a road trunk of the each intersection relative to the current trajectory point; and repeat the above operations until road trunks of each intersection are obtained.

[0087] The high-definition map intersection vectorization labeling device described above can execute the method provided by any embodiment of the application, has the function modules and beneficial effects corresponding to the execution method. Technical details not described in detail in the embodiment can be referred to the high-definition map intersection vectorization labeling method provided by any embodiment of the application.

[0088] Embodiment five

[0089] According to embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0090] Figure 5 A schematic block diagram of an example electronic device 500 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 assistants, 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 meant to limit implementations of the present disclosure described and / or claimed in this document.

[0091] As shown in Figure 5 The device 500 includes a computing unit 501 that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. Various programs and data required for operation of the device 500 can also be stored in the RAM 503. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0092] A plurality of components in the device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a magnetic disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the device 500 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0093] The computing unit 501 can be various general and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 501 performs various methods and processes described above, such as the high-definition map intersection vectorization labeling method. For example, in some embodiments, the high-definition map intersection vectorization labeling method can be implemented as a computer software program, which is tangibly embodied in a machine-readable medium, such as the storage unit 508. In some embodiments, part or all of the computer program can be loaded and / or installed on the device 500 via the ROM 502 and / or the communication unit 509. When the computer program is loaded into the RAM 503 and executed by the computing unit 501, one or more steps of the high-definition map intersection vectorization labeling method described above can be performed. Alternatively, in other embodiments, the computing unit 501 can be configured to perform the high-definition map intersection vectorization labeling method by any other appropriate means, such as by means of firmware.

[0094] Various implementations of the systems and techniques described above herein can be realized 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 programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various implementations 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.

[0095] 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 block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a standalone software package, or entirely on a remote machine or server.

[0096] 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.

[0097] 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.

[0098] The systems and techniques described herein 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 herein), 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), blockchain network, and the Internet.

[0099] The computer system can include clients and servers. The clients and the servers are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also known as cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of large management difficulty and weak business scalability in traditional physical host and VPS service.

[0100] It should be understood that the various forms of flow shown above can be re-ordered, added to, or deleted from without departing from the scope of the present disclosure. For example, the steps recited in the present disclosure can be executed in parallel, in series, or in a different order, as long as the desired results of the technical solutions of the present disclosure are achieved, and the present disclosure is not limited herein. In the technical solutions of the present disclosure, the acquisition, storage, and application of user personal information involved comply with relevant legal regulations and do not violate public order and good customs.

[0101] The above detailed description does not constitute a limitation on the protection scope of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A high-precision map intersection vectorization labeling method, the method comprising: processing pre-acquired trajectories to obtain trajectories corresponding to each road; calculating same-direction trajectory matching pairs in each two trajectories according to a same-direction trajectory point calculation formula based on the trajectories corresponding to each road; calculating crossing trajectory matching pairs in each two trajectories according to a crossing trajectory point calculation formula based on the trajectories corresponding to each road; identifying intersection points based on the same-direction trajectory matching pairs and the crossing trajectory matching pairs, and performing Euclidean distance clustering according to the identified intersection points to obtain center points of each intersection; extracting road trunks corresponding to each intersection according to the center points of each intersection and the trajectories corresponding to each road; vectorizing and labeling each intersection based on the center points of each intersection and the road trunks corresponding to each intersection.

2. The method of claim 1, wherein, The processing of the pre-acquired trajectories to obtain trajectories corresponding to each road comprises: matching each two trajectories in the pre-acquired trajectories to obtain matching results of each two trajectories; determining trajectories corresponding to each road based on the matching results of each two trajectories.

3. The method of claim 2, wherein, The matching of each two trajectories in the pre-acquired trajectories to obtain matching results of each two trajectories comprises: randomly extracting two trajectories in the pre-acquired trajectories as a first trajectory and a second trajectory; searching for other trajectory points in the first trajectory that have an angle within a threshold range with each trajectory point in the first trajectory; calculating a nearest point of the second trajectory for each trajectory point based on each trajectory point and the other trajectory points, and taking each trajectory point and the nearest point as matching results of the first trajectory and the second trajectory; and repeating the above operations until matching results of each two trajectories are obtained.

4. The method of claim 1, wherein, The extraction of road trunks corresponding to each intersection according to the center points of each intersection and the trajectories corresponding to each road comprises: searching for trajectory points around each intersection within a predetermined radius with the center point of each intersection as the center; if at least one trajectory point around each intersection is searched within the predetermined radius, generating a road trunk of each intersection based on each trajectory point in the at least one trajectory point around each intersection.

5. The method of claim 4, wherein, The generation of a road trunk of each intersection based on each trajectory point in the at least one trajectory point around each intersection comprises: extracting a trajectory point in the at least one trajectory point around each intersection as a current trajectory point; judging whether another trajectory point in a same-direction trajectory matching pair containing the current trajectory point has been marked; if the other trajectory point has not been marked, connecting the current trajectory point and the corresponding intersection as a road trunk of each intersection relative to the current trajectory point; and repeating the above operations until road trunks of each intersection are obtained.

6. A high-definition map intersection vectorization labeling device, the device comprising: a processing module, an identifying module, an extracting module, and a labeling module; wherein the processing module is configured to process pre-acquired trajectories to obtain trajectories corresponding to each road; The identification module is configured to calculate, based on the trajectories corresponding to each road, same-direction trajectory matching pairs in each two trajectories according to a same-direction trajectory point calculation formula; calculate, based on the trajectories corresponding to each road, cross trajectory matching pairs in each two trajectories according to a cross trajectory point calculation formula; identify a cross point based on the same-direction trajectory matching pairs and the cross trajectory matching pairs; and perform Euclidean distance clustering according to the identified cross point to obtain a center point of each intersection. The extraction module is configured to extract a road stem corresponding to each intersection according to the center point of each intersection and the trajectories corresponding to each road. The labeling module is configured to label each intersection in a vector manner based on the center point of each intersection and the road stem corresponding to each intersection.

7. The apparatus of claim 6, wherein the processing module is specifically configured to match each two trajectories in the pre-acquired trajectories to obtain matching results of each two trajectories; and determine the trajectory corresponding to each road based on the matching results of each two trajectories.

8. The apparatus of claim 7, wherein the processing module is specifically configured to randomly extract two trajectories in the pre-acquired trajectories as a first trajectory and a second trajectory; search, for each trajectory point in the first trajectory, other trajectory points in the first trajectory that have an angle within a threshold range with the trajectory point; calculate, based on each trajectory point and the other trajectory points, a nearest point of each trajectory point to the second trajectory; take each trajectory point and the nearest point as a matching result of the first trajectory and the second trajectory; and repeat the above operations until the matching results of each two trajectories are obtained.

9. The apparatus of claim 6, wherein the extraction module is specifically configured to take the center point of each intersection as a center of a circle, search for trajectory points around each intersection within a predetermined radius, and generate a road stem of each intersection based on each trajectory point in at least one trajectory point around each intersection if the at least one trajectory point around each intersection is searched within the predetermined radius.

10. The apparatus of claim 9, wherein the extraction module is specifically configured to extract one trajectory point in the at least one trajectory point around each intersection as a current trajectory point; determine whether another trajectory point in a same-direction trajectory matching pair containing the current trajectory point has been marked; if the another trajectory point has not been marked, connect the current trajectory point and a corresponding intersection as a road stem of the corresponding intersection relative to the current trajectory point; and repeat the above operations until the road stem of each intersection is obtained.

11. An electronic device, comprising: at least one processor; and a memory connected with the at least one processor in communication; wherein 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 any one of claims 1-5.

12. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are used to enable the computer to perform the method of any one of claims 1-5.

13. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1-5.

Images