Lane detection method, device, equipment and storage medium

Abstract

The present disclosure provides a lane line detection method, apparatus, device, and storage medium, relating to the field of artificial intelligence technology, specifically the field of intelligent transportation and autonomous driving technology. The method comprises: obtaining an initial image set collected by a target vehicle, the initial image set containing at least one road image; detecting the initial image set using a lightweight target detection model and a classification model to obtain attribute information of lane lines in each road image; deduplicating the detected initial image set based on a ray intersection algorithm and a density clustering algorithm to obtain a deduplicated target image set; comparing the attribute information of the lane lines in the target image with the information in a pre-constructed database, and determining the type of lane line change based on the comparison results. The lane line detection method provided by the present disclosure improves the detection accuracy and efficiency of lane lines.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, specifically to the field of intelligent transportation and autonomous driving technology, and in particular to lane line detection methods, devices, equipment, and storage media. Background Art

[0002] With the continuous advancement of urbanization, the resulting massive traffic demand has put pressure on modern urban transportation. Relying solely on human resources is no longer sufficient to meet the demands of urban traffic management. The increasing pressure on urban traffic has necessitated the development of smart transportation. Promoting a comprehensive smart transportation strategy can significantly alleviate urban congestion, improve traffic regulation, and enhance public travel efficiency. Smart transportation, in particular, places a higher demand on efficient and accurate road network data. Map navigation, especially mobile navigation apps (applications), has become a common practice and a necessity for many people. One of the most common scenarios in map navigation route planning is prompting users to stay in the correct lane and to indicate when to turn at intersections. This is also a cornerstone of future autonomous driving. If these prompts are not timely or are incorrect, the vehicle may end up in an unintended turning lane, significantly increasing travel time and disrupting travel plans. Furthermore, for many novice drivers, being directed into the wrong lane can cause anxiety and increase driving risks, leading to frequent accidents. Summary of the Invention

[0003] The present disclosure provides a lane line detection method, apparatus, device, and storage medium.

[0004] According to a first aspect of the present disclosure, a lane line detection method is provided, comprising: obtaining an initial image set collected from a target vehicle, the initial image set comprising at least one road image; detecting the initial image set using a lightweight target detection model and a classification model to obtain attribute information of lane lines in each road image; deduplicating the detected initial image set based on a ray intersection algorithm and a density clustering algorithm to obtain a deduplicated target image set; comparing the attribute information of the lane lines in the target image with information in a pre-constructed database, and determining the type of lane line change based on the comparison result.

[0005] According to a second aspect of the present disclosure, a lane line detection device is provided, comprising: an acquisition module, configured to acquire an initial image set collected from a target vehicle, the initial image set comprising at least one road image; a detection module, configured to detect the initial image set using a lightweight target detection model and a classification model, and obtain attribute information of lane lines in each road image; a deduplication module, configured to dedupe the detected initial image set based on a ray intersection algorithm and a density clustering algorithm, and obtain a deduplicated target image set; and a comparison module, configured to compare the attribute information of lane lines in the target image with information in a pre-constructed database, and determine the type of change in the lane lines based on the comparison result.

[0006] According to a third aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; 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 execute the method described in any implementation manner in the first aspect.

[0007] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, where the computer instructions are used to cause a computer to execute the method described in any implementation manner of the first aspect.

[0008] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program, which implements the method described in any implementation manner of the first aspect when executed by a processor.

[0009] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The accompanying drawings are provided to facilitate a better understanding of the present invention and do not constitute a limitation of the present disclosure.

[0011] Figure 1 is an exemplary system architecture diagram in which the present disclosure may be applied;

[0012] Figure 2 is a flow chart of an embodiment of a lane detection method according to the present disclosure;

[0013] Figure 3 is a flow chart of another embodiment of a lane detection method according to the present disclosure;

[0014] Figure 4 This is an application scenario diagram of the lane line detection method disclosed in the present invention;

[0015] Figure 5 is a schematic structural diagram of an embodiment of a lane detection device according to the present disclosure;

[0016] Figure 6 3 is a block diagram of an electronic device used to implement the lane line detection method of the embodiment of the present disclosure. DETAILED DESCRIPTION

[0017] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0018] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in the present disclosure may be combined with each other. The present disclosure will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0019] like Figure 1 As shown, system architecture 100 may include a server 101, a network 102, and a target vehicle 103. Network 102 is used to provide a medium for a communication link between server 101 and target vehicle 103. Network 102 may include various connection types, such as wired, wireless communication links, or fiber optic cables.

[0020] The server 101 can provide various services. For example, the server 101 can analyze and process the initial image set obtained from the target vehicle 103 and generate processing results (such as the change type of the lane line).

[0021] The target vehicle 103 can interact with the server 101 through the network 102 to receive or send messages, etc. For example, the target vehicle 103 can collect road images in real time and send the collected road images to the server 101.

[0022] It should be noted that the lane line detection method provided in the embodiment of the present disclosure is generally executed by the server 101 , and accordingly, the lane line detection device is generally provided in the server 101 .

[0023] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is merely illustrative. Any number of terminal devices, networks and servers may be provided as required.

[0024] Continue to refer Figure 2, which shows a process 200 of an embodiment of a lane line detection method according to the present disclosure. The lane line detection method includes the following steps:

[0025] Step 201: Acquire an initial image set collected of a target vehicle.

[0026] In this embodiment, the execution body of the lane line detection method (for example Figure 1 The server 101 shown in the figure will obtain an initial image set collected by the target vehicle, and the initial image set includes at least one road image. The target vehicle in this embodiment can refer to any socially operated vehicle. The on-board equipment of the target vehicle can collect a sequence of images associated with the road in real time, thereby obtaining an initial image set, and the initial image set includes at least one road image. The on-board equipment can be installed at any position on the target vehicle, as long as it can achieve the purpose of photographing the road. After collecting the initial image set, the acquisition device of the target vehicle will send it to the above-mentioned execution entity, and the above-mentioned execution entity will obtain the initial image set.

[0027] In step 202 , the initial image set is detected using a lightweight object detection model and a classification model to obtain attribute information of lane lines in each road image.

[0028] In this embodiment, the execution entity uses a lightweight object detection model and a classification model to detect the road images in the initial image set, thereby obtaining lane line attribute information for each road image. Lane lines are directional lane markings that indicate the direction vehicles should travel at the entrance to an intersection. These markings are often used at intersections with heavy traffic to clarify driving directions, ensure lane separation, and alleviate traffic pressure. Lane line attribute information generally includes the lane line's position in the road image and its type, such as left turn, right turn, or straight ahead.

[0029] Specifically, the execution entity can first use a lightweight object detection model (also known as a lane line object detection model) to detect the location information of lane lines in each initial road image in the initial image set, thereby obtaining a detection result. This detection result provides a target for subsequent lane line classification. In other words, the object detection model can determine whether the road image contains lane lines. If so, it detects the location information of the lane lines in the road image and annotates the lane lines in the form of annotation boxes.

[0030] The execution entity can then input the image set generated by the object detection model into a classification model (also known as a lane classification model), which outputs information about the lane types in the road image, such as left turn, right turn, and straight ahead. Because the object detection model can determine the presence of lanes in the road image and their positions within the image, and the classification model can then be used to determine the specific lane types, the lanes detected in the previous step can be accurately classified.

[0031] It should be noted that in order to ensure real-time recognition, a lightweight model can be configured on the vehicle-mounted device, where both the target detection model and the classification model can be implemented using models such as SSD MobileNet, EfficientDet, etc. This embodiment does not make specific restrictions on this.

[0032] Step 203 : Deduplication is performed on the detected initial image set based on a ray intersection algorithm and a density clustering algorithm to obtain a deduplication target image set.

[0033] In this embodiment, the above-mentioned execution subject will deduplicate the initial image set based on the ray intersection algorithm and the density clustering algorithm to obtain the deduplicated target image set. Specifically, the above-mentioned execution subject will first use the ray intersection algorithm to calculate the actual position information in each road image. Among them, the ray triangle intersection algorithm is a method for quickly calculating the intersection of rays and triangles in three dimensions. The intersection and center of gravity coordinates can be quickly obtained through vector and matrix calculations without the need to pre-calculate the plane equation containing the triangle. Here, the ray intersection algorithm can be used to calculate the depth information of the lane line in the road image, that is, the distance and angle information of the lane line from the target vehicle in the road image can be obtained through the ray intersection algorithm. Since the actual position information of the target vehicle when collecting the road image can be obtained, based on this, the above-mentioned execution subject can calculate the actual position information of the lane line based on the actual position information of the target vehicle and the depth information of the lane line. The actual position information here refers to the position information of the lane line in the real-world coordinate system.

[0034] Since the target vehicle captures a sequence of images, the lane lines contained in multiple consecutive images may be the same, and the initial image set needs to be deduplicated. Specifically, the above-mentioned execution entity will cluster the detected initial image set based on the calculated actual position information of the lane lines and the density clustering algorithm, thereby obtaining multiple clustered image groups, and the lane lines contained in the images of each image group are the same. Finally, the target image is selected from the multiple images in each image group, thereby obtaining a target image set containing multiple target images. Here, the target image can be selected according to the number of lane lines contained in the image, the clarity of the lane lines, and the distance between the lane lines and the target vehicle. For example, the image containing the largest number of lane lines is used as the target image, the image containing the clearest lane lines is used as the target image, and the image with the lane lines closest to the target vehicle is used as the target image.

[0035] Step 204 : Compare the attribute information of the lane line in the target image with the information in the pre-built database, and determine the change type of the lane line based on the comparison result.

[0036] In this embodiment, the execution entity compares the lane line attribute information in the target image with information in a pre-built database and determines the lane line change type based on the comparison results. Since the execution entity can obtain the actual location of the target vehicle in real time, it can obtain a local map within a certain range of that location from the cloud database. This local map includes the stored lane line attribute information. The execution entity then compares the lane line attribute information in the target image with the lane line attribute information in the local map. This comparison includes essential navigation data such as lane line location, lane line arrow type, and lane line combination to determine whether the current lane line information is consistent with the lane line information in the map. If a change occurs, the change type is classified as addition, modification, or removal based on the actual situation. The relevant image data is then retrieved for subsequent verification, thereby updating the lane line information in the cloud database.

[0037] The lane line detection method provided in the disclosed embodiment first obtains an initial image set collected by the target vehicle, the initial image set containing at least one road image; then uses a lightweight target detection model and a classification model to detect the initial image set to obtain the attribute information of the lane lines in each road image; then deduplication is performed on the initial image set based on a ray intersection algorithm and a density clustering algorithm to obtain a deduplicated target image set; finally, the attribute information of the lane lines in the target image is compared with the information in a pre-constructed database, and the type of lane line change is determined based on the comparison result. The lane line detection method in this embodiment uses a lightweight target detection model to locate the lane line position in the image, extracts the lane line area, and then uses a classification model to classify the lane line arrows, which can accurately identify the type of lane line arrow; by combining the position of the target vehicle and the lane line arrow with the real-time differential strategy of the cloud map, it can quickly determine whether the lane line has changed, thereby improving the detection accuracy and efficiency of the lane line.

[0038] In the technical solutions disclosed herein, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0039] Continue to refer Figure 3 , Figure 3 A process 300 of another embodiment of a lane line detection method according to the present disclosure is shown. The lane line detection method includes the following steps:

[0040] Step 301: Acquire an initial image set collected from a target vehicle.

[0041] In this embodiment, the execution body of the lane line detection method (for example Figure 1 The server 101 shown in FIG. 101 obtains an initial image set collected by the target vehicle, which includes at least one road image. Step 301 is substantially identical to step 201 in the aforementioned embodiment. The specific implementation method can be found in the aforementioned description of step 201 and will not be repeated here.

[0042] Step 302: For each road image in the initial image, the road image is input into a lightweight object detection model, and a road image containing image position information of lane lines is output.

[0043] In this embodiment, for each road image in the initial image, the execution entity uses a lightweight object detection model (also referred to as a lane line object detection model) to detect the position information of the lane lines in each initial image in the initial image set, thereby obtaining a detection result, which can provide a target for subsequent lane line classification. In other words, the road image is input into the lightweight object detection model, and a road image containing the image position information of the lane lines can be output. The object detection model can determine whether the road image contains lane lines, and if so, detect the position information of the lane lines in the road image and annotate the lane lines in the form of annotation boxes.

[0044] In step 303 , the road image including the image position information of the lane line is input into the classification model, and the road image including the type information of the lane line is output.

[0045] In this embodiment, the execution entity can input the image set obtained based on the object detection model into a classification model (also known as a lane classification model), thereby outputting information about the types of lane lines in the road image, such as left turn, right turn, and straight ahead. Because the object detection model can determine that the road image contains lane lines and their positions within the image, and then the classification model is used to determine the specific types of lane lines, accurate classification of the lane lines detected in the previous step is achieved.

[0046] Through the above steps, accurate detection of the position and type information of the lane lines is achieved.

[0047] Step 304 : Calculate the actual position information of the lane lines in each road image using a ray intersection algorithm.

[0048] In this embodiment, the execution entity may use a ray intersection algorithm to calculate the actual position information of the lane lines in each road image. For example, the execution entity may first determine the position information of the lane lines in the road image using the ray intersection algorithm, and then combine the actual position information of the target vehicle in the road image to calculate the actual position information of the lane lines in the road image, that is, the position information of the lane lines in the real-world coordinate system.

[0049] In some optional implementations of this embodiment, step 304 further includes: obtaining the position information of the target vehicle; calculating the depth information of the lane lines in each road image based on the ray intersection algorithm; and calculating the actual position information of the lane lines in the road image based on the position information and depth information of the target vehicle.

[0050] In this implementation, the execution entity can use a ray intersection algorithm to calculate the depth information of lane lines in a road image. Specifically, the ray intersection algorithm can be used to determine the distance and angle between the lane lines in the road image and the target vehicle. Because the target vehicle's actual position information when the road image was captured is available, the execution entity can calculate the lane lines' actual position information based on the target vehicle's actual position information and the lane lines' depth information. This actual position information refers to the lane lines' position information in a real-world coordinate system. This allows for more accurate determination of the lane lines' actual position information.

[0051] Step 305 : clustering the detected initial image set based on the actual position information and a density clustering algorithm to obtain at least one image group.

[0052] In this embodiment, since the target vehicle captures a sequence of images, the lane markings contained in multiple consecutive images may be identical, necessitating deduplication of the initial image set. Specifically, the execution entity clusters the detected initial image set based on the calculated actual lane marking positions and a density clustering algorithm, thereby generating multiple clustered image groups, each containing identical lane markings.

[0053] Step 306 : Determine a target image from each image group based on a preset rule to obtain a target image set.

[0054] In this embodiment, the execution entity selects a target image from multiple images in each image group based on preset rules, thereby obtaining a target image set comprising multiple target images. The preset rules may include selecting target images based on the number of lane lines contained in the image, the clarity of the lane lines, and the distance between the lane lines and the target vehicle. For example, the image containing the largest number of lane lines is selected as the target image, the image containing the clearest lane lines is selected as the target image, or the image with the lane lines closest to the target vehicle is selected as the target image. This embodiment does not specifically limit the method for selecting target images.

[0055] Through the above steps, the road images in the initial image set are filtered and deduplicated, thereby improving the efficiency of lane line detection.

[0056] Step 307 : Obtain a local map from a pre-built database based on the actual position information of the lane line.

[0057] In this embodiment, the above-mentioned execution entity can obtain the actual location information of the target vehicle in real time, and then the above-mentioned execution entity can obtain a local map within a certain range of the location from the cloud database, and the local map includes the attribute information of the stored lane lines.

[0058] Step 308 : Compare the attribute information of the lane lines in the target image with the attribute information of the lane lines in the local map.

[0059] In this embodiment, the above-mentioned execution entity will compare the attribute information of the lane line in the target image with the attribute information of the lane line in the local map. The comparison includes necessary navigation data such as lane line position, lane line arrow type, lane line combination, etc., so as to determine whether the current lane line information is consistent with the lane line information in the map.

[0060] Step 309: Determine the lane line change type based on the comparison result.

[0061] In this embodiment, the execution entity determines the lane change type based on the comparison results, which can be addition, modification, or removal. If a change occurs, the change type is classified as addition, modification, or removal based on the actual situation. The relevant image data is then collected for subsequent verification, thereby updating the lane information in the cloud database.

[0062] from Figure 3 It can be seen that Figure 2 Compared with the corresponding embodiments, the lane line detection method in this embodiment highlights the steps of detecting lane lines, deduplicating the initial image set, and comparing the change types of lane lines, so as to accurately identify the lane line arrow type; through the position of the target vehicle and the lane line arrow combined with the real-time differential strategy of the cloud map, it is possible to quickly determine whether the lane line has changed, thereby improving the lane line detection accuracy and efficiency.

[0063] Continue to refer Figure 4 , which illustrates an application scenario of the lane line detection method disclosed herein. In this application scenario, the target vehicle's onboard equipment can first collect a sequence of images associated with the road in real time, thereby obtaining an initial image set consisting of multiple road images. A lightweight object detection model is then used to detect lane lines (also known as vehicle signals) in each of the collected road images, outputting a road image containing image location information for the lane lines, with the lane lines already marked and framed. The road image containing the lane line location information is then input into a classification model to classify the specific lane line types, thereby obtaining a road image containing both lane line location information and type information (e.g., left turn, straight ahead, right turn). Finally, based on the target vehicle's actual location information, a local map is retrieved from a cloud database. The collected lane line information is then differentiated from the lane line information stored in the local map to determine whether the lane lines have changed. If so, the change type is classified based on the actual situation: addition, modification, or removal. The relevant image data is then retrieved for subsequent verification, and the lane line information in the map data is finally updated.

[0064] Further references Figure 5 As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of a lane line detection device. Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0065] like Figure 5 As shown, the lane line detection device 500 of this embodiment includes: an acquisition module 501, a detection module 502, a deduplication module 503, and a comparison module 504. The acquisition module 501 is configured to acquire an initial image set collected by a target vehicle, the initial image set including at least one road image; the detection module 502 is configured to detect the initial image set using a lightweight target detection model and a classification model to obtain attribute information of lane lines in each road image; the deduplication module 503 is configured to dedupe the detected initial image set based on a ray intersection algorithm and a density clustering algorithm to obtain a deduplicated target image set; and the comparison module 504 is configured to compare the attribute information of lane lines in the target image with information in a pre-built database and determine the type of lane line change based on the comparison results.

[0066] In this embodiment, the lane line detection device 500 includes the acquisition module 501, the detection module 502, the deduplication module 503, and the comparison module 504. The specific processing and technical effects thereof can be referred to in the respective Figure 2 The relevant descriptions of steps 201-204 in the corresponding embodiment are not repeated here.

[0067] In some optional implementations of this embodiment, the attribute information of the lane line includes the image position information of the lane line and the type information of the lane line; and the detection module is further configured to: for each road image in the initial image, input the road image into a lightweight target detection model, and output a road image containing the image position information of the lane line; input the road image containing the image position information of the lane line into a classification model, and output a road image containing the type information of the lane line.

[0068] In some optional implementations of this embodiment, the deduplication module includes: a calculation submodule, configured to use a ray intersection algorithm to calculate the actual position information of the lane lines in each road image; a clustering submodule, configured to cluster the initial image set after detection based on the actual position information and a density clustering algorithm to obtain at least one image group; a determination submodule, configured to determine the target image from each image group based on preset rules to obtain a target image set.

[0069] In some optional implementations of this embodiment, the computing submodule is further configured to: obtain the position information of the target vehicle; calculate the depth information of the lane lines in each road image based on the ray intersection algorithm; and calculate the actual position information of the lane lines in the road image based on the position information and depth information of the target vehicle.

[0070] In some optional implementations of this embodiment, the comparison module is further configured to: obtain a local map from a pre-built database based on the actual position information of the lane line; compare the attribute information of the lane line in the target image with the attribute information of the lane line in the local map; determine the change type of the lane line based on the comparison result, and the change type includes a new addition type, a modification type, or a removal type.

[0071] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0072] Figure 6 A schematic block diagram of an example electronic device 600 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 laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are provided as examples only and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0073] like Figure 6 As shown, the device 600 includes a computing unit 601, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 602 or a computer program loaded from a storage unit 608 into a random access memory (RAM) 603. Various programs and data required for the operation of the device 600 can also be stored in the RAM 603. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0074] Various components in device 600 are connected to I / O interface 605, including an input unit 606, such as a keyboard, mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a magnetic disk, optical disk, etc.; and a communication unit 609, such as a network card, modem, wireless communication transceiver, etc. The communication unit 609 allows device 600 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0075] The computing unit 601 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the computing unit 601 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 suitable processor, controller, microcontroller, etc. The computing unit 601 performs the various methods and processes described above, such as the lane detection method. For example, in some embodiments, the lane detection method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 600 via the ROM 602 and / or the communication unit 609. When the computer program is loaded into the RAM 603 and executed by the computing unit 601, one or more steps of the lane detection method described above can be performed. Alternatively, in other embodiments, the computing unit 601 can be configured to perform the lane detection method by any other suitable means (e.g., via firmware).

[0076] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0077] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0078] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A 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, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0079] To provide interaction with a user, the systems and techniques described herein 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 pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the 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 input, voice input, or tactile input).

[0080] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, 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.

[0081] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not a limitation herein.

[0082] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A lane line detection method, comprising: Acquire an initial image set collected by the target vehicle, wherein the initial image set includes at least one road image; Detecting the initial image set using a lightweight object detection model and a classification model to obtain attribute information of lane lines in each road image; The initial image set after detection is deduplicated based on the ray triangle intersection algorithm and the density clustering algorithm to obtain the deduplicated target image set; Comparing the attribute information of the lane line in the target image with the information in a pre-built database, and determining the change type of the lane line based on the comparison result; The deduplication of the detected initial image set based on the ray-triangle intersection algorithm and the density clustering algorithm to obtain the deduplication target image set includes: Calculating actual position information of lane lines in each road image using a ray triangle intersection algorithm, wherein the actual position information includes position information of the lane lines in a real-world coordinate system; Clustering the detected initial image set based on the actual position information and a density clustering algorithm to obtain at least one image group; A target image is determined from each image group based on a preset rule to obtain the target image set.

2. The method according to claim 1, wherein The lane line attribute information includes the image position information of the lane line and the type information of the lane line; as well as The lightweight object detection model and the classification model are used to detect the initial image set to obtain the attribute information of the lane lines in each road image, including: For each road image in the initial image, input the road image into a lightweight object detection model, and output a road image containing image position information of lane lines; The road image containing the image position information of the lane line is input into the classification model, and a road image containing the type information of the lane line is output.

3. The method according to claim 1, wherein The method of calculating the actual position information of lane lines in each road image using a ray-triangle intersection algorithm includes: Obtaining the location information of the target vehicle; Calculate the depth information of lane lines in each road image based on the ray triangle intersection algorithm; Based on the position information of the target vehicle and the depth information, actual position information of the lane line in the road image is calculated.

4. The method according to claim 3, wherein: The comparing the attribute information of the lane line in the target image with the information in the pre-built database, and determining the change type of the lane line based on the comparison result, includes: Obtaining a local map from a pre-built database based on the actual position information of the lane line; Comparing the attribute information of the lane line in the target image with the attribute information of the lane line in the local map; The change type of the lane line is determined based on the comparison result, and the change type includes a new type, a modified type, or a removed type.

5. A lane detection device, comprising: an acquisition module, configured to acquire an initial image set collected by a target vehicle, wherein the initial image set includes at least one road image; a detection module configured to detect the initial image set using a lightweight object detection model and a classification model to obtain attribute information of lane lines in each road image; A deduplication module is configured to perform deduplication on the initial image set after detection based on a ray-triangle intersection algorithm and a density clustering algorithm to obtain a deduplicated target image set; a comparison module configured to compare attribute information of lane lines in the target image with information in a pre-built database, and determine a change type of the lane lines based on the comparison result; Wherein, the deduplication module includes: a calculation submodule configured to calculate actual position information of lane lines in each road image using a ray-triangulation intersection algorithm, wherein the actual position information includes position information of the lane lines in a real-world coordinate system; a clustering submodule configured to cluster the detected initial image set based on the actual position information and a density clustering algorithm to obtain at least one image group; The determination submodule is configured to determine a target image from each image group based on a preset rule to obtain the target image set.

6. The device according to claim 5, wherein The lane line attribute information includes the image position information of the lane line and the type information of the lane line; and The detection module is further configured to: For each road image in the initial image, input the road image into a lightweight object detection model, and output a road image containing image position information of lane lines; The road image containing the image position information of the lane line is input into the classification model, and a road image containing the type information of the lane line is output.

7. The device according to claim 5, wherein The calculation submodule is further configured to: Obtaining the location information of the target vehicle; Calculate the depth information of lane lines in each road image based on the ray triangle intersection algorithm; Based on the position information of the target vehicle and the depth information, actual position information of the lane line in the road image is calculated.

8. The device according to claim 7, wherein The comparison module is further configured to: Obtaining a local map from a pre-built database based on the actual position information of the lane line; Comparing the attribute information of the lane line in the target image with the attribute information of the lane line in the local map; The change type of the lane line is determined based on the comparison result, and the change type includes a new type, a modified type, or a removed type.

9. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed 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 according to any one of claims 1 to 4.

10. A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 4.

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

Images