CN114926799A - Lane line detection method, device, equipment and readable storage medium - Google Patents

Abstract

The invention discloses a lane line detection method, a device, equipment and a readable storage medium, wherein the method comprises the following steps: acquiring a single-frame lane line, and performing lane line identification detection processing on the single-frame lane line to obtain a current lane line; matching the current lane line with a preset map to obtain a local map of the current vehicle; detecting whether the current lane line is accurate or not according to the local map and the current lane line; and when the current lane line is detected to be inaccurate, constructing a virtual lane line. According to the invention, the lane line is identified and detected, and is matched with the high-precision map, so that whether the current lane line is accurate or not is detected after the local map where the vehicle is located is obtained, therefore, the problem that the lane line detection is influenced by a large error generated by data due to hardware precision, environmental factors or sensor jitter is solved, the lane line detection efficiency is improved, meanwhile, the accuracy and reliability of the lane line detection are greatly improved, and the situations that the existing detection technology needs high detection conditions and has large detection errors are reduced.

Description

A lane line detection method, device, device and readable storage medium

technical field

The present invention relates to the technical field of artificial intelligence, and in particular, to a lane line detection method, device, device and readable storage medium.

Background technique

With the rise of intelligent assisted driving, lane line detection, as an important part of it, has also been vigorously developed in recent years. The lane line detection technology in ADAS (Advanced Driver Assistance Systems) assisted driving is mainly based on camera sensors. It detects the current lane line through image and video analysis, provides lane line information for subsequent lane departures, and effectively carries out departure warning.

In the process of conceiving and realizing this application, the inventors of this application found that the lane line detection based on image and video analysis in the prior art adopts the two-dimensional image collected by the camera sensor, which is greatly affected by the environment, especially in the case of poor imaging. Under the condition of , it is easy to be interfered by non-lane line points, can not achieve ideal results, and it is far difficult to meet the L3 and L4 level of automatic driving technical indicators. In addition, for lane line detection based on two-dimensional image information, there is no way to obtain a direct physical lane line model. It needs to be strictly calibrated according to the installation of the camera. On the basis of image semantic segmentation, it takes a lot of training to extract lane line pixels. The label data to solve the problem of multi-scenario application.

The preceding statements are intended to provide general background information and may not constitute prior art.

SUMMARY OF THE INVENTION

The technical problem to be solved by the embodiments of the present invention is to provide a lane line detection method, device, device, and readable storage medium, which can solve the problem of large errors in data caused by hardware accuracy, environmental factors or sensor jitter, which affect lane line detection. question.

In order to solve the above problem, a first aspect of the embodiments of the present application provides a lane line detection method, which at least includes the following steps:

Acquiring a single frame of lane line, and performing lane line recognition and detection processing on the single frame of lane line to obtain the current lane line;

Matching the current lane line with a preset map to obtain a local map where the current vehicle is located;

Detecting whether the current lane line is accurate according to the local map and the current lane line;

When it is detected that the current lane line is inaccurate, a corresponding virtual lane line is constructed.

In a possible implementation manner of the first aspect, after obtaining the current lane line, the method includes:

Acquire multiple frames of lane lines, and perform fusion processing on the multiple frames of lane lines to obtain a calibrated current lane line.

In a possible implementation manner of the first aspect, the acquiring multiple frames of lane lines, and performing fusion processing on the multiple frames of lane lines to obtain a calibrated current lane line, including:

Acquire multiple frames of lane lines, perform lane line recognition and detection processing on each frame of lane lines, and detect the initial lane lines corresponding to each frame of lane lines;

According to the target clustering algorithm, multiple groups of initial lane lines are clustered and fused successively to obtain calibrated lane lines.

In a possible implementation manner of the first aspect, after the obtaining the current lane line, the method further includes:

The current lane line and the historical lane line are compared to determine whether to deviate from the historical lane.

In a possible implementation manner of the first aspect, performing lane line recognition and detection processing on the single frame of lane lines includes:

The camera internal and external parameters are calibrated on the acquired single frame lane line, and the first lane line information is obtained;

Coordinate transformation and lane line fitting are successively performed on the first lane line to obtain the current lane line.

In a possible implementation manner of the first aspect, after obtaining the local map where the current vehicle is located, the method further includes:

Obtain vehicle visual odometer information, and detect whether the current lane line is accurate according to the vehicle visual odometer information.

In a possible implementation manner of the first aspect, the detecting whether the current lane line is accurate according to the local map and the current lane line includes:

obtaining a local map corresponding to the current lane line;

extracting location information of landmark objects in the local map corresponding to the current lane;

According to the position information of the landmark object, the current lane line is matched and identified to detect whether the current lane line is accurate.

Correspondingly, a second aspect of the embodiments of the present application provides a lane line detection device, including:

A single-frame lane line extraction module, used to obtain a single-frame lane line, and perform lane line identification and detection processing on the single-frame lane line to obtain the current lane line;

a map matching module for matching the current lane line with a preset map to obtain a local map where the current vehicle is located;

a lane line detection module, configured to detect whether the current lane line is accurate according to the local map and the current lane line;

The lane line building module is configured to construct a corresponding virtual lane line when it is detected that the current lane line is inaccurate.

A third aspect of the embodiments of the present application further provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the lane line described in any one of the above is implemented The steps of the detection method.

A fourth aspect of the embodiments of the present application further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, implements the steps of any one of the lane line detection methods described above.

Implementing the embodiment of the present invention has the following beneficial effects:

Embodiments of the present invention provide a lane line detection method, device, device, and readable storage medium. The method includes: acquiring a single frame of lane lines, performing lane line identification and detection processing on the single frame of lane lines, and obtaining a current lane match the current lane line with the preset map to obtain the local map where the current vehicle is located; according to the local map and the current lane line, detect whether the current lane line is accurate; when the current lane line is detected When inaccurate, construct the corresponding virtual lane line. In the embodiment of the present invention, the lane lines are identified and detected, matched with high-precision maps, and the local map where the vehicle is located is obtained, and then the current lane lines are detected to be accurate, so as to solve the problem of large errors in data caused by hardware accuracy, environmental factors or sensor jitter The problem that affects the detection of lane lines, while improving the efficiency of lane line detection, greatly improves the accuracy and reliability of lane line detection, effectively reduces the error of lane line detection, and reduces the existing detection technology that requires higher detection conditions and detection. Larger errors.

Description of drawings

FIG. 1 is a schematic flowchart of a lane line detection method according to an embodiment of the present application;

FIG. 2 is a schematic block diagram of the structure of a lane line detection device according to an embodiment of the application;

FIG. 3 is a schematic structural block diagram of a computer device according to an embodiment of the present application.

The realization, functional characteristics and advantages of the purpose of the present application will be further described with reference to the accompanying drawings in conjunction with the embodiments.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

In the description of the present application, it should be understood that the terms "first", "second", etc. are only used for description purposes, and cannot be interpreted as indicating or implying relative importance or implying the number of indicated technical features . Thus, a feature defined as "first", "second", etc., may expressly or implicitly include one or more of that feature. In the description of this application, unless stated otherwise, "plurality" means two or more.

The embodiments of the present application may be applied to a server, and the server may be an independent server, or may provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security Services, Content Delivery Network (CDN), and cloud servers for basic cloud computing services such as big data and artificial intelligence platforms.

First, the application scenarios that the present invention can provide are introduced, such as providing a lane line detection method, device, equipment and readable storage medium, which can realize lane line detection and improve the efficiency and accuracy of lane line detection.

The first embodiment of the present invention:

See Figure 1.

As shown in FIG. 1 , this embodiment provides a lane line detection method, which at least includes the following steps:

S1. Acquire a single frame of lane line, and perform lane line identification and detection processing on the single frame of lane line to obtain the current lane line;

S2, matching the current lane line with a preset map to obtain a local map where the current vehicle is located;

S3, according to the local map and the current lane line, detect whether the current lane line is accurate;

S4. When it is detected that the current lane line is inaccurate, construct a corresponding virtual lane line.

In the prior art, the lane line detection based on image and video analysis adopts the two-dimensional image collected by the camera sensor, which is greatly affected by the environment. To achieve ideal results, it is far from meeting the L3 and L4 level of autonomous driving technical indicators. In addition, for lane line detection based on two-dimensional image information, there is no way to obtain a direct physical lane line model. It needs to be strictly calibrated according to the installation of the camera. On the basis of image semantic segmentation, it takes a lot of training to extract lane line pixels. The label data to solve the problem of multi-scenario application.

In this embodiment, in order to solve the above technical problems, a lane line detection method is proposed. By identifying and detecting the lane lines and matching with a high-precision map, the local map where the vehicle is located is obtained and then the current lane line is detected. Hardware accuracy, environmental factors or sensor jitter cause large errors in data that affect lane line detection. While improving the efficiency of lane line detection, it greatly improves the accuracy and reliability of lane line detection, and effectively reduces lane line detection. error, reducing the situation that the existing detection technology requires higher detection conditions and the detection error is large.

Specifically, for step S1, a single frame of lane lines is acquired by an image acquisition device, and lane line recognition and detection processing is performed on the acquired single frame of lane lines, thereby obtaining the current lane lines.

For step S2, the extracted current lane line and the preset high-precision map are mainly matched to match the position of the high-precision map where the current lane line is located, and a local map near the position is obtained within the preset range of the position.

For step S3, according to the local map and the current lane line obtained on the map where the current lane line is located, and according to the position data and characteristics of the current lane line, match with the lane line in the local map to detect whether the current lane line is accurate.

For step S4, when it is detected that the current lane line does not match or is inaccurate, a corresponding virtual lane line is constructed to replace the current lane line.

In a preferred embodiment, after the obtaining the current lane line, it includes:

Acquire multiple frames of lane lines, and perform fusion processing on the multiple frames of lane lines to obtain a calibrated current lane line.

In a specific embodiment, after step S1, the current lane line can be calibrated by acquiring multiple frames of lane lines for fusion processing, and the calibrated current lane line can be obtained as the basis for subsequent lane line detection.

In a preferred embodiment, the acquiring multiple frames of lane lines, and performing fusion processing on the multiple frames of lane lines to obtain a calibrated current lane line, including:

Acquire multiple frames of lane lines, perform lane line recognition and detection processing on each frame of lane lines, and detect the initial lane lines corresponding to each frame of lane lines;

According to the target clustering algorithm, multiple groups of initial lane lines are clustered and fused successively to obtain calibrated lane lines.

In a specific embodiment, the specific process of calibrating the current lane line by acquiring multiple frames of lane lines and performing fusion processing includes: acquiring multiple frames of lane lines collected by an image acquisition device, and performing lane lines on each frame of lane lines respectively. Recognition and detection processing, so as to extract the initial lane lines corresponding to each frame of lane lines, and generate multiple groups of initial lane lines; according to the clustering algorithm according to the target, cluster and fuse multiple groups of initial lane lines successively to obtain calibrated lanes Wire.

Exemplarily, the target clustering algorithm may be a K-means clustering algorithm or a Gaussian mixture model clustering algorithm. This embodiment does not limit the target clustering algorithm, which can be determined by those skilled in the art as required.

Optionally, the multi-frame lane lines are fused to obtain a calibrated current lane line, which may specifically include:

According to the multiple sets of lane line positioning data of the target road, multiple lane line positioning data of any target position in the target road are obtained; according to the target clustering algorithm, the multiple lane line positioning data of any target position are clustered , obtain the lane line clustering center of any target position; repeat the multiple sets of lane line positioning data according to the target road to obtain multiple lane line positioning data of any target position in the target road to the The steps of clustering multiple lane line positioning data of any target position to obtain the lane line clustering center of any target position, and obtaining the lane line clustering centers of multiple target positions; The lane line clustering centers of the multiple target positions are fused to obtain the lane lines of the target road.

In addition, when the deviation between any lane line positioning data and the pre-stored map data is greater than a preset threshold, the lane line positioning data is deleted. Exemplarily, the preset threshold is twice the current road width, and when the difference between the longitude and latitude of the lane in any lane line positioning data and the longitude and latitude of the lane corresponding to the lane line in the pre-stored map exceeds the preset threshold, it is considered that the lane line The positioning data is abnormal data, and the abnormal data is deleted, thereby further improving the accuracy of lane line positioning and detection.

In a preferred embodiment, after obtaining the current lane line, the method further includes:

The current lane line and the historical lane line are compared to determine whether to deviate from the historical lane.

In a specific embodiment, after the current lane line is obtained, the current lane line and the historical lane line can also be compared to determine whether the deviation between the two is greater than a preset threshold, and if so, it is determined that the current lane line has deviated from the historical lane; If not, it is judged that the current lane line does not deviate from the historical lane.

Optionally, after obtaining the calibrated current lane line, it is also possible to compare the calibrated current lane line with the historical lane line to determine whether the deviation between the two is greater than a preset threshold, and if so, determine the calibrated current lane line. The line has deviated from the historical lane; if not, it is judged that the calibrated current lane line does not deviate from the historical lane.

In a preferred embodiment, performing lane line recognition and detection processing on the single-frame lane line includes:

The camera internal and external parameters are calibrated on the acquired single frame lane line, and the first lane line information is obtained;

Coordinate transformation and lane line fitting are successively performed on the first lane line to obtain the current lane line.

In a specific embodiment, the step S1 may specifically include: after acquiring a single frame of lane line, successively performing camera internal and external parameter calibration, coordinate transformation and fitting processing on the single frame of lane line to obtain the current lane line. The calibration of the internal and external parameters of the camera is an essential step in image processing, mainly to calculate the internal and external parameters of the camera and the distortion parameters; the distortion parameters are corrected for distortion, and the corrected image is generated; the internal and external parameters are generally used for image 3D scene reconstruction; while the camera The calibration involves four coordinate systems, including the world coordinate system, the camera coordinate system, the image physical coordinate system, and the image pixel coordinate system, so coordinate transformation is also required; after the coordinate transformation is completed, the lane line is fitted, and the current lane line is finally obtained. .

In a preferred embodiment, after obtaining the local map where the current vehicle is located, the method further includes:

Obtain vehicle visual odometer information, and detect whether the current lane line is accurate according to the vehicle visual odometer information.

In a specific embodiment, after step S2, the method may further include detecting whether the current lane line is accurate through the visual odometry technology, detecting the current lane line through the vehicle visual odometer information, first obtaining the visual odometer information of the current vehicle, and then Based on monocular vision lane line detection and monocular visual odometry positioning accuracy optimization, visual odometer (VO) uses the image information collected by the vehicle camera to restore the 6-DOF information of the car body itself, including the 3-DOF rotation. and translation with 3 degrees of freedom. Vision sensors can provide rich perception information for accurate lane line detection.

In a preferred embodiment, the detecting whether the current lane line is accurate according to the local map and the current lane line includes:

obtaining a local map corresponding to the current lane line;

extracting location information of landmark objects in the local map corresponding to the current lane;

According to the position information of the landmark object, the current lane line is matched and identified to detect whether the current lane line is accurate.

In a specific embodiment, for step S4, the main steps are to obtain a local map within the preset range of the current lane line; to extract the position information of the landmark objects in the local map corresponding to the current lane; according to the position information of the landmark objects, The current lane line is matched and identified to detect whether the current lane line is accurate.

Optionally, the specific process of step S4 is as follows: first obtain a plurality of local maps corresponding to the lane lines of the target road; exemplarily, the local map represents a map including the lane lines of the target road and other traffic elements, and other traffic elements may include road surfaces, Street lights, plants, sky, zebra crossings, buildings, etc. The local maps corresponding to multiple target road lane lines can be obtained from a database that pre-stores local maps corresponding to multiple target road lane lines, and the local maps corresponding to multiple target road lane lines can be obtained by cruising on the target road lane lines. The image data collected by the map collecting vehicle and the sensor data are obtained by data fusion. Second, the location information of landmark objects in the local map corresponding to the lane lines of multiple target roads is extracted; exemplarily, the landmark objects can be zebra crossings, traffic signs, and large billboards. For example, when multiple target roads are at an intersection , there are zebra crossings at the intersection of each target road and intersection, then the zebra crossing can be selected as a landmark object. When the map collecting vehicle collects the target road image, since the map collecting vehicle is equipped with a positioning system (such as GPS) and an image collecting device, image information of traffic elements and latitude and longitude information will be uploaded. Finally, after global fusion of the local maps corresponding to multiple target road lane lines according to the position information of the landmark object, the current lane line is matched and identified according to the position information of the landmark object, and whether the current lane line is detected is detected. precise.

A lane line detection method provided by this embodiment includes: acquiring a single frame of lane lines, performing lane line recognition and detection processing on the single frame of lane lines, and obtaining a current lane line; Match to obtain the local map where the current vehicle is located; according to the local map and the current lane line, detect whether the current lane line is accurate; when it is detected that the current lane line is inaccurate, construct a corresponding virtual lane line. In this embodiment, the lane lines are identified and detected and matched with the high-precision map to obtain the local map where the vehicle is located, and then check whether the current lane lines are accurate, so as to solve the problem of large errors in data caused by hardware accuracy, environmental factors or sensor jitter. The problem of lane line detection, while improving the efficiency of lane line detection, greatly improves the accuracy and reliability of lane line detection, effectively reduces the error of lane line detection, and reduces the existing detection technology that requires higher detection conditions and detection errors. larger case.

The second embodiment of the present invention:

See Figure 2.

As shown in FIG. 2, this embodiment provides a lane line detection device, including:

A single-frame lane line extraction module 100, configured to acquire a single-frame lane line, and perform lane line identification and detection processing on the single-frame lane line to obtain a current lane line;

a map matching module 200, configured to match the current lane line with a preset map to obtain a local map where the current vehicle is located;

a lane line detection module 300, configured to detect whether the current lane line is accurate according to the local map and the current lane line;

The lane line construction module 400 is configured to construct a corresponding virtual lane line when it is detected that the current lane line is inaccurate.

Optionally, the lane line detection device may further include:

The multi-frame lane line extraction module is used for acquiring multi-frame lane lines, and performing fusion processing on the multi-frame lane lines to obtain the calibrated current lane line.

Optionally, the multi-frame lane line extraction module may specifically include:

The multi-frame lane line extraction unit is used to obtain multi-frame lane lines, respectively perform lane line recognition and detection processing on each frame of lane lines, and detect the initial lane lines corresponding to each frame of lane lines;

The multi-frame lane line clustering unit is used to cluster and fuse multiple groups of initial lane lines successively according to the target clustering algorithm to obtain calibrated lane lines.

Optionally, the lane line detection device may further include:

The historical comparison module is used to compare the current lane line with the historical lane line to determine whether to deviate from the historical lane.

Optionally, the single-frame lane line extraction module 100 may specifically include:

a first processing unit, configured to perform camera internal and external parameter calibration on the acquired single frame lane line to obtain the first lane line information;

The second processing unit is used for successively performing coordinate transformation and lane line fitting processing on the first lane line to obtain the current lane line.

Optionally, the lane line detection device may further include:

The visual odometry module is used for acquiring vehicle visual odometer information, and detecting whether the current lane line is accurate according to the vehicle visual odometer information.

Optionally, the lane line detection module 300 may specifically include:

a map acquisition unit, configured to acquire a local map corresponding to the current lane line;

a location extraction unit, configured to extract location information of landmark objects in the local map corresponding to the current lane;

The detection unit is used for matching and identifying the current lane line according to the position information of the landmark object, and detecting whether the current lane line is accurate.

In this embodiment, the single-frame lane line extraction module first obtains a single-frame lane line, and performs lane line recognition and detection processing on the single-frame lane line to obtain the current lane line; Set the map to be matched to obtain the local map where the current vehicle is located; then use the lane line detection module to detect whether the current lane line is accurate according to the local map and the current lane line; When the current lane line is inaccurate, a corresponding virtual lane line is constructed. In this embodiment, the lane lines are identified and detected and matched with the high-precision map to obtain the local map where the vehicle is located, and then check whether the current lane lines are accurate, so as to solve the problem of large errors in data caused by hardware accuracy, environmental factors or sensor jitter. The problem of lane line detection, while improving the efficiency of lane line detection, greatly improves the accuracy and reliability of lane line detection, effectively reduces the error of lane line detection, and reduces the existing detection technology that requires higher detection conditions and detection errors. larger case.

Referring to FIG. 3 , an embodiment of the present application further provides a computer device. The computer device may be a server, and its internal structure may be as shown in FIG. 3 . The computer device includes a processor, memory, a network interface, and a database connected by a system bus. Among them, the processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium, an internal memory. The nonvolatile storage medium stores an operating system, a computer program, and a database. The memory provides an environment for the execution of the operating system and computer programs in the non-volatile storage medium. The database of the computer device is used to store data such as lane line detection methods. The network interface of the computer device is used to communicate with an external terminal through a network connection. The computer program implements a lane line detection method when executed by a processor. The lane line detection method includes: acquiring a single frame of lane lines, performing lane line recognition and detection processing on the single frame lane lines to obtain a current lane line; and matching the current lane line with a preset map to obtain a current vehicle The local map where it is located; according to the local map and the current lane line, it is detected whether the current lane line is accurate; when it is detected that the current lane line is inaccurate, a corresponding virtual lane line is constructed.

An embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, a method for detecting lane lines is implemented, including the steps of: acquiring a single frame of lane lines, Frame lane lines for lane line recognition and detection processing to obtain the current lane line; match the current lane line with a preset map to obtain a local map where the current vehicle is located; detect the current lane line according to the local map and the current lane line Whether the lane line is accurate; when it is detected that the current lane line is inaccurate, a corresponding virtual lane line is constructed.

The lane line detection method implemented above recognizes and detects lane lines, matches them with high-precision maps, obtains the local map where the vehicle is located, and then detects whether the current lane lines are accurate, so as to solve the problem of data caused by hardware accuracy, environmental factors or sensor jitter. The problem that a large error affects the detection of lane lines, while improving the efficiency of lane line detection, greatly improves the accuracy and reliability of lane line detection, effectively reduces the error of lane line detection, and reduces the need for existing detection technology. Detection conditions and detection errors are large.

In the above-mentioned embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are only illustrative, for example, the division of the modules may be a logical function division, and there may be other division methods in actual implementation, for example, multiple modules or components may be combined or Integration into another device, or some features may be omitted, or not implemented. On the other hand, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of units or modules, and may be in electrical or other forms.

The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, may be located in one place, or may be distributed to multiple modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically alone, or two or more modules may be integrated into one module. The above-mentioned integrated modules can be implemented in the form of hardware, and can also be implemented in the form of software function modules.

The above are the preferred embodiments of the present invention. It should be pointed out that for those skilled in the art, without departing from the principle of the present invention, several improvements and modifications can be made, and these improvements and modifications are also considered as It is the protection scope of the present invention.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage In the medium, when the computer program is executed, it may include the processes of the above-mentioned method embodiments. Wherein, any reference to memory, storage, database or other medium provided in this application and used in the embodiments may include non-volatile and/or volatile memory. Nonvolatile memory may include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

Claims (10)

1. A lane line detection method is characterized by at least comprising the following steps:

acquiring a single-frame lane line, and performing lane line identification detection processing on the single-frame lane line to obtain a current lane line;

matching the current lane line with a preset map to obtain a local map of the current vehicle;

detecting whether the current lane line is accurate or not according to the local map and the current lane line;

and when the current lane line is detected to be inaccurate, constructing a corresponding virtual lane line.

2. The lane line detection method according to claim 1, comprising, after the obtaining of the current lane line:

obtaining a plurality of frames of lane lines, and performing fusion processing on the plurality of frames of lane lines to obtain the calibrated current lane line.

3. The method according to claim 2, wherein the acquiring a plurality of frames of lane lines, and performing fusion processing on the plurality of frames of lane lines to obtain a calibrated current lane line includes:

acquiring multiple frames of lane lines, respectively carrying out lane line identification detection processing on each frame of lane line, and detecting to obtain an initial lane line corresponding to each frame of lane line;

and successively clustering and fusing the multiple groups of initial lane lines according to a target clustering algorithm to obtain the calibrated lane lines.

4. The lane line detection method according to claim 1, further comprising, after the obtaining of the current lane line:

and comparing the current lane line with the historical lane line, and judging whether the current lane line deviates from the historical lane line.

5. The lane line detection method according to claim 1, wherein the performing lane line identification detection processing on the single-frame lane line includes:

carrying out camera internal and external parameter calibration on the acquired single-frame lane line to obtain first lane line information;

and carrying out coordinate conversion and lane line fitting treatment on the first lane line in sequence to obtain the current lane line.

6. The lane line detection method according to claim 1, further comprising, after the obtaining of the local map of the current vehicle,:

and acquiring vehicle visual odometer information, and detecting whether the current lane line is accurate or not according to the vehicle visual odometer information.

7. The lane line detection method according to claim 1, wherein the detecting whether the current lane line is accurate according to the local map and the current lane line includes:

acquiring a local map corresponding to the current lane line;

extracting the position information of the landmark object in the local map corresponding to the current lane;

and matching and identifying the current lane line according to the position information of the landmark object, and detecting whether the current lane line is accurate or not.

8. A lane line detection apparatus, comprising:

the single-frame lane line extraction module is used for acquiring a single-frame lane line and carrying out lane line identification detection processing on the single-frame lane line to obtain a current lane line;

the map matching module is used for matching the current lane line with a preset map to obtain a local map of the current vehicle;

the lane line detection module is used for detecting whether the current lane line is accurate or not according to the local map and the current lane line;

and the lane line building module is used for building a corresponding virtual lane line when the current lane line is detected to be inaccurate.

9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the lane line detection method of any one of claims 1 to 7.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the lane line detection method according to any one of claims 1 to 7.

Images