CN113011285B - Lane line detection method and device, automatic driving vehicle and readable storage medium - Google Patents
Lane line detection method and device, automatic driving vehicle and readable storage medium Download PDFInfo
- Publication number
- CN113011285B CN113011285B CN202110229709.9A CN202110229709A CN113011285B CN 113011285 B CN113011285 B CN 113011285B CN 202110229709 A CN202110229709 A CN 202110229709A CN 113011285 B CN113011285 B CN 113011285B
- Authority
- CN
- China
- Prior art keywords
- lane line
- vanishing point
- point
- lane
- road image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Multimedia (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
The embodiment of the disclosure provides a lane line detection method, a lane line detection device, an automatic driving vehicle and a readable storage medium, wherein the method comprises the following steps: acquiring a road image, wherein the road image comprises at least one lane line; preprocessing the road image to obtain a lane line probability map corresponding to the road image; acquiring a lane line vanishing point in the road image; and performing curve fitting on each lane line in the road image according to the lane line vanishing point and the lane line probability graph to obtain a lane line detection result. The method can utilize the characteristic that the lane line inevitably passes through the lane line vanishing point, carry out geometric constraint on the fitting result through the lane line vanishing point, can overcome the problem that the deviation occurs in curve fitting due to the deviation of the sampling point obtained by the lane line probability map, and improves the accuracy of lane line detection.
Description
Technical Field
The embodiment of the disclosure relates to the field of unmanned driving, in particular to a lane line detection method and device, an automatic driving vehicle and a readable storage medium.
Background
The intelligent unmanned vehicle system generally comprises a sensing module, a decision-making module, a path planning module, a control module and the like. Vehicle positioning is the process of determining the current position and attitude of a vehicle. While visual image-based positioning plays a key role in vehicle positioning. A common visual positioning technology is that image acquisition is carried out based on an initial positioning, then lane line detection is carried out on the acquired image, the detected lane line is compared with a corresponding lane line in a high-precision map, and then the initial positioning is corrected, so that the vehicle positioning precision is improved. Therefore, the lane line detection technique plays an important role in stable vehicle positioning.
In the prior art, when lane line detection is performed, generally, an acquired image (as shown in fig. 1 a) is input into a trained deep learning network to obtain a lane line probability map (as shown in fig. 1 b), then the input image is sampled according to the lane line probability map to obtain sampling points corresponding to each lane line, and then curve fitting is performed on each lane line based on the sampling points of each lane line to obtain a final lane line detection result.
However, the deep learning network predicted lane probability map is susceptible to some deviation due to many factors (such as strong shading, lane wear, etc.), such as the block area in the lane probability map shown in fig. 1 b. Furthermore, the points sampled from this probability map and subsequently fitted to obtain a curve are often not accurate enough. Taking the rightmost line in fig. 1b as an example, the sampling point shown in fig. 2a is obtained after sampling, and it can be seen that there is already a sampling point deviating from the real lane. And then curve fitting is carried out on the sampling points in the figure 2a, so that the lane line detection result shown in the figure 2b is obtained, and it can be seen that obvious deviation exists between the lane line obtained by fitting and a real lane line.
Therefore, how to reduce the influence of the deviation of the sampling points obtained based on the lane line probability map on the lane line detection result and improve the accuracy of the lane line detection result becomes a problem which needs to be solved urgently.
Disclosure of Invention
The embodiment of the disclosure provides a lane line detection method and device, an automatic driving vehicle and a readable storage medium, which can perform geometric constraint on a fitting result through a lane line vanishing point, can overcome the problem that curve fitting is deviated due to deviation of sampling points obtained by a lane line probability map, and improve accuracy of lane line detection.
According to a first aspect of embodiments of the present disclosure, there is provided a lane line detection method, the method including:
acquiring a road image, wherein the road image comprises at least one lane line;
preprocessing the road image to obtain a lane line probability map corresponding to the road image;
acquiring a lane line vanishing point in the road image;
and performing curve fitting on each lane line in the road image according to the lane line vanishing point and the lane line probability graph to obtain a lane line detection result.
According to a second aspect of embodiments of the present disclosure, there is provided a lane line detection apparatus, the apparatus including:
the system comprises an image acquisition module, a data acquisition module and a data processing module, wherein the image acquisition module is used for acquiring a road image, and the road image comprises at least one lane line;
the image preprocessing module is used for preprocessing the road image to obtain a lane line probability map corresponding to the road image;
the lane vanishing point acquisition module is used for acquiring lane line vanishing points in the road image;
and the curve fitting module is used for performing curve fitting on each lane line in the road image according to the lane line vanishing point and the lane line probability graph to obtain a lane line detection result.
According to a third aspect of embodiments of the present disclosure, there is provided an autonomous vehicle comprising:
a processor, a memory and a computer program stored on the memory and executable on the processor, the processor implementing the aforementioned lane line detection method when executing the program.
According to a fourth aspect of embodiments of the present disclosure, there is provided a readable storage medium having instructions that, when executed by a processor of an autonomous vehicle, enable the autonomous vehicle to perform the aforementioned lane line detection method.
The embodiment of the disclosure provides a lane line detection method, a lane line detection device, an automatic driving vehicle and a readable storage medium, wherein the method comprises the following steps: acquiring a road image, wherein the road image comprises at least one lane line; preprocessing the road image to obtain a lane line probability map corresponding to the road image; acquiring a lane line vanishing point in the road image; and performing curve fitting on each lane line in the road image according to the lane line vanishing point and the lane line probability graph to obtain a lane line detection result. The embodiment of the disclosure adopts the lane line vanishing point and the lane line probability map to perform curve fitting on the lane line together, and compared with the prior art in which the curve fitting is performed on the sampling point obtained by only using the lane line probability map, the embodiment of the disclosure utilizes the property that the lane line inevitably passes through the lane line vanishing point, and performs geometric constraint on the fitting result through the lane line vanishing point, so that the problem that the deviation occurs in the curve fitting due to the deviation of the sampling point obtained by the lane line probability map can be overcome, and the accuracy of the lane line detection is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings needed to be used in the description of the embodiments of the present disclosure will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present disclosure, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive exercise.
FIG. 1a shows a road image captured in the prior art;
FIG. 1b is a prior art lane line probability map for a road image;
FIG. 2a shows a schematic representation of a prior art sampling of a road image;
FIG. 2b is a schematic diagram illustrating a lane line detection result corresponding to a road image in the prior art;
FIG. 3 shows a flow chart of method steps for lane line detection in one embodiment of the present disclosure;
fig. 4 shows a block diagram of a lane line detection apparatus in an embodiment of the present disclosure;
FIG. 5 illustrates a block diagram of an autonomous vehicle in one embodiment of the disclosure.
Detailed Description
Technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure, and it is apparent that the described embodiments are some, but not all, of the embodiments of the present disclosure. All other embodiments, which can be obtained by a person skilled in the art without making creative efforts based on the embodiments in the embodiments of the present disclosure, belong to the protection scope of the embodiments of the present disclosure.
Example one
Referring to fig. 3, a flowchart illustrating steps of a lane line detection method in an embodiment of the present disclosure is shown, which specifically includes the following steps:
And 102, preprocessing the road image to obtain a lane line probability map corresponding to the road image.
And 103, acquiring a lane line vanishing point in the road image.
And 104, performing curve fitting on each lane line in the road image according to the lane line vanishing point and the lane line probability graph to obtain a lane line detection result.
It should be noted that the lane line detection method provided by the embodiment of the disclosure is applied to an unmanned vehicle, and the lane line in the advancing direction of the vehicle can be detected by the lane line detection method of the disclosure, so that a decision basis is provided for a driving strategy. Unless otherwise specified, the term "vehicle" as used in this disclosure refers to an "unmanned vehicle," i.e., an "autonomous vehicle. The unmanned vehicles in the present disclosure include vehicles traveling on the ground, such as cars, trucks, buses, etc., but may also include unmanned devices traveling in the air, such as drones, airplanes, helicopters, etc., and unmanned devices traveling in water, such as boats, submarines, etc. Further, a "vehicle" in the present disclosure may or may not accommodate one or more passengers therein. The vehicle discussed in this disclosure may also be applied in the field of unmanned delivery, such as express delivery, take-away meals, etc.
The road image may be collected by a camera located at any position in the vehicle, for example, when the vehicle is running straight, the camera located in front of the vehicle is used to collect the road image, and when the vehicle turns, the camera located on the left side or the right side of the vehicle body is used to collect the road image. The road video can also be shot by a camera device on the vehicle in the driving process of the vehicle, and then the shot video is collected to the road image corresponding to the current frame. The acquired road image reflects the road condition of the environment where the vehicle is located and comprises at least one lane line. It should be noted that the lane lines in the embodiments of the present disclosure include a driving lane line of the vehicle, such as various indicator markings on a road surface, and a boundary line of a lane, such as a road edge of a road.
The lane line probability map can reflect the probability that each pixel point in the road image belongs to the lane line, and each coordinate point in the lane line probability map corresponds to a probability value. The lane line probability map and the collected road image have the same size and pixels, each pixel point in the road image has a corresponding coordinate point in the lane line probability map, and the probability value corresponding to the coordinate point is the probability that the pixel point belongs to the lane line. And marking the coordinate points in the lane line probability graph according to a preset threshold value to obtain the initial sampling points of the lane lines in the road image. For example, the coordinate points with the probability value larger than 90% in the lane line probability map are marked to obtain the initial sampling points of the lane lines in the road image. And further analyzing and processing the initial sampling point to obtain a detection result of the lane line.
The lane line vanishing point reflects a perspective structure of the road image, and the lane line vanishing point is inevitably passed no matter whether the lane line is a straight line or a curve. As long as a lane line is included in the road image, the road image necessarily includes a lane line vanishing point. For example, when the vehicle is moving straight, the lane line vanishing point in the present disclosure is a lane line vanishing point in a road image captured by a camera placed in front of the vehicle, that is, a lane line vanishing point in a forward direction of the vehicle; when the vehicle turns, the lane line vanishing point in the disclosure is a lane line vanishing point in a road image shot by a camera arranged on the left side or the right side of the vehicle, namely the lane line vanishing point of a curve where the vehicle is located. The lane line vanishing point can be generally determined from the intersection of a plurality of lane lines.
In the embodiment provided by the disclosure, after the road image is collected, the collected road image is preprocessed to obtain the lane line probability map corresponding to the road image. Specifically, the lane line detection model may be trained based on the deep learning network, and then the acquired road image may be input into the trained lane line detection model to obtain the lane line probability map. The process of obtaining the lane line probability map corresponding to the road image is essentially a process of classifying each pixel point on the road image. The lane line probability map of the road image may be obtained by any one of the existing processing schemes, and the embodiment of the present invention is not particularly limited.
After the lane line probability map is obtained, it is necessary to further determine a lane line vanishing point in the road image. Specifically, road images including lane lines in different scenes can be collected, the lane lines and vanishing points of the road images are labeled to form a data set, and the data set is divided into a training set, a verification set and a test set according to a certain proportion. The training set is used for training the deep convolutional network, the verification set is used for selecting the optimal training model, and the test set is used for testing the performance of the design model at the later stage. And then inputting the marked road image into a classification convolution neural network to obtain an optimal training model, and obtaining the coordinates of the vanishing point of the lane line through the training model. Or shooting and collecting continuous road images by using a camera device, carrying out image preprocessing on the collected continuous images to obtain a lane line probability map corresponding to each frame of image, marking coordinate points belonging to lane lines in the lane line probability map according to a preset threshold, respectively and randomly extracting two points on two lane lines in each lane line probability map to calculate intersection point coordinates of the image lane lines to obtain lane line intersection point coordinates of multi-frame images, counting intersection point coordinate histograms based on the lane line intersection point coordinates of the multi-frame images, calculating mean values, variances and center distances of the histograms, and determining lane line vanishing points in the road images according to a maximum likelihood estimation method.
And after the lane line vanishing point is determined, performing curve fitting on each lane line in the road image according to the lane line vanishing point and the lane line probability graph to obtain a lane line detection result. Specifically, the coordinate points in the lane line probability map may be marked according to a preset threshold to obtain an initial sampling point of each lane line in the road image, and then curve fitting may be performed on each lane line according to the lane vanishing point and the initial sampling point of each lane line to obtain a lane line detection result.
In an optional embodiment of the present disclosure, the step 104 of performing curve fitting on each lane line in the road image according to the lane line vanishing point and the lane line probability map to obtain a lane line detection result includes:
s11, respectively determining a first sampling point set corresponding to each lane line in the road image according to the lane line probability map;
step S12, performing curve fitting on each lane line according to the lane line vanishing point and at least one sampling point in the first sampling point set to obtain at least one fitting curve corresponding to each lane line;
step S13, determining the score value of each fitting curve;
and S14, determining the detection result of each lane line according to the score value of each fitting curve corresponding to each lane line.
In the embodiment provided by the present disclosure, after determining the lane line probability map corresponding to the road image, a first sampling point set corresponding to each lane line in the road image may be determined according to the lane line probability map. Specifically, coordinate points with probability values larger than a preset threshold value in the lane line probability are marked, the marked coordinate points in the lane line probability map are classified according to the position of each lane line in the road image, and a first sampling point set corresponding to each lane line in the road image is obtained and is marked as point _ set = { point1, point2, point3,. Once..
And then performing curve fitting on each lane line according to the lane vanishing point and at least one sampling point in the first sampling point set. Specifically, assuming that a first sampling point set corresponding to a lane line 1 in a road image is point _ set1= { point1, point2, point3,... And point N }, selecting at least one sampling point from the first sampling point set point _ set1, and obtaining a subset point _ sub1= { point i, point j,... And point m }, of the first sampling point. Let the lane vanishing point be point _ vp, the lane vanishing point and the subset point _ sub1 of the first sampling point together form a set subset _ vp = { point _ vp, point i, point j. The set subset _ vp is a union of the lane vanishing point _ vp and the subset point _ sub1 of the first sampling point. And performing curve fitting on the lane line 1 based on the set subset _ vp to obtain a fitted curve of the lane line 1. Since the subset point _ sub1 of the first sampling point has various combination conditions, the subset _ vp of the union of the subset point _ sub1 and the lane vanishing point _ vp also has various combination conditions, and the lane line 1 is subjected to curve fitting according to the subset _ vp to obtain at least one fitting curve.
When the subset point _ sub1 of the first sampling point set point _ set1 is determined, any sampling point can be randomly selected from the first sampling points to form the subset point _ sub1, and each sampling point in the first sampling point set point _ set1 can be orderly traversed to obtain the subset point _ sub1.
After obtaining at least one fitted curve corresponding to each lane line, it is necessary to finally determine one fitted curve from the at least one fitted curve as a detection result of the lane line. Specifically, each fitting curve corresponding to the lane line may be evaluated, the score value of each fitting curve may be determined, and then the fitting curve with the highest score value may be determined as the detection result of the lane line according to the score value of each fitting curve. It should be noted that, when determining the detection result of the lane line according to the score value of each fitted curve, specifically selecting which fitted curve corresponding to the score value is the final detection result, and determining according to a specific evaluation rule, determining the fitted curve with the highest score value as the detection result of the lane line is only an exemplary description, and does not constitute a limitation of the present disclosure. If the score is positive, selecting a fitting curve with the highest score as a detection result of the lane line; and if the score is negative, selecting the fitting curve with the lowest score as the detection result of the lane line.
In an optional embodiment of the present disclosure, the determining a score value of each fitted curve in step S13 includes:
substep S131, resampling each fitted curve corresponding to each lane line to obtain a second sampling point set of each fitted curve corresponding to each lane line;
substep S132, determining a probability value corresponding to each sampling point in the second sampling point set according to the lane line probability map;
and a substep S133 of calculating the sum of probability values corresponding to each sampling point in the second sampling point set of each fitting curve corresponding to each lane line to obtain a score value of each fitting curve corresponding to each lane line.
In the embodiment provided by the disclosure, when the score value of each fitted curve corresponding to each lane line is determined, each fitted curve corresponding to each lane line may be resampled to obtain a second sampling point set of each fitted curve, and then the score value of the fitted curve is calculated according to the probability value corresponding to each sampling point in the second sampling point set.
Specifically, taking the lane line 1 as an example, assuming that the lane line 1 corresponds to 3 fitting curves, when calculating the score value of the fitting curve a of the lane line, the fitting curve a is resampled according to a preset sampling rule, for example, the fitting curve a is mapped onto the lane line probability map according to a curve equation or a straight line equation of the fitting curve a to obtain a mapping curve a of the fitting curve a, then one sampling point is taken at intervals of m coordinate points on the mapping curve a and added into the second sampling point set of the fitting curve a, and after the sampling is finished, the second sampling point set of the fitting curve a can be obtained.
Because each coordinate point in the lane line probability map corresponds to a probability value, the probability value corresponding to each sampling point in the second sampling point set can be obtained according to the lane line probability map. And adding the probability values corresponding to the sampling points in the second sampling point set of the fitting curve A to obtain the score value of the fitting curve A. Or calculating the average probability value of each sampling point in the second sampling point set of the fitting curve A, and taking the obtained average probability value as the score value of the fitting curve A.
It should be noted that, in the embodiment provided in the present disclosure, when resampling is performed on each fitted curve corresponding to the same lane line, the rule needs to be calculated according to the same sampling rule and score value, so as to ensure the validity of the score result.
In an optional embodiment of the present disclosure, the determining, according to the lane line probability map, a probability value corresponding to each sampling point in the second set of sampling points in sub-step S132 includes:
a11, determining the coordinate value of each sampling point in the second sampling point set;
a12, searching a coordinate point which is the same as the coordinate value of each sampling point in the lane line probability map, and acquiring a probability value corresponding to the coordinate point, wherein the sampling points in the second sampling point set correspond to the coordinate points in the lane line probability map in a one-to-one manner;
and A13, determining the probability value corresponding to the coordinate point corresponding to the sampling point as the probability value of the sampling point.
No matter which curve fitting method is adopted, each sampling point in the fitting curve belongs to the road image, the lane line probability map and the collected road image have the same size and pixels, each pixel point in the road image has a corresponding coordinate point in the lane line probability map, and the probability value corresponding to the coordinate point is the probability that the pixel point belongs to the lane line. Each sampling point in the second sampling point set has a corresponding coordinate point in the lane line probability map, and the probability value of the corresponding coordinate point is the probability value of the sampling point.
In an optional embodiment of the present disclosure, after the step 103 of acquiring the lane line vanishing point in the road image, the method further includes:
s21, randomly disturbing the lane vanishing point in a preset range to obtain at least one new lane vanishing point;
step S22, generating a lane vanishing point set according to the lane vanishing point and the at least one new lane vanishing point, wherein the lane vanishing point set comprises the lane vanishing point and the at least one new lane vanishing point;
step S23, respectively calculating a lane line detection value corresponding to each lane vanishing point in the lane vanishing point set;
step S24, determining a target lane vanishing point according to a lane detection value corresponding to each lane vanishing point in the lane vanishing point set;
and S25, determining a lane line detection result according to the target lane vanishing point.
In practical application, factors such as vehicle body shake, other line segment interference in a road image and the like may cause inaccuracy of the obtained lane line vanishing point, and therefore, in the embodiment provided by the disclosure, in order to reduce the influence of the inaccuracy of the lane line vanishing point on the lane line detection result, after the lane line vanishing point is obtained, random disturbance is performed on the lane line vanishing point within a preset range, that is, the coordinate value of the lane line vanishing point is floated up and down within the preset range to obtain at least one new lane line vanishing point, and then a lane line vanishing point set is generated based on the initially determined lane line vanishing point and the new lane line vanishing point obtained after the random disturbance.
And performing lane line detection based on each lane line vanishing point in the set of lane line vanishing points, namely performing the content of the step 104 according to each lane line vanishing point, calculating a lane line detection value corresponding to each lane line vanishing point, then determining a target lane line vanishing point according to each lane line detection value, and taking the lane line detected based on the target lane line vanishing point as a final lane line detection result.
In an optional embodiment of the present disclosure, the step S23 of respectively calculating a lane line detection value corresponding to each lane vanishing point in the set of lane vanishing points includes:
substep S231, performing curve fitting on each lane line in the road image according to any lane line vanishing point in the lane vanishing point set and the lane line probability map to obtain a score value of each lane line corresponding to the lane vanishing point;
and a substep S232, summing the score values of each lane line to obtain a lane line detection value of the lane vanishing point.
Assuming that the set of lane line vanishing points includes the lane line vanishing point _ vp _ i, the process of calculating the lane line detection value of the lane line vanishing point _ vp _ i is as follows:
firstly, respectively determining a first sampling point set corresponding to each lane line in a road image according to a lane line probability graph, and then performing curve fitting on each lane line according to a lane line vanishing point _ vp _ i and at least one sampling point in the first sampling point set to obtain at least one fitting curve corresponding to each lane line. Determining the score value of each fitting curve, determining the score value of the highest score in the score values of each fitting curve as the score value of the lane line, and summing the score values of each lane line to obtain the lane line detection value of the vanishing point _ vp _ i of the lane line. Other values, such as the average score of the lane line, may also be used as the lane line detection value of the lane line vanishing point, and this disclosure is not particularly limited as long as the lane line detection results corresponding to each lane line vanishing point can be compared.
The process of calculating the lane line detection value of each lane line vanishing point is essentially to execute the steps S11 to S14 based on each lane line vanishing point, and then determine the lane line detection value of the lane line vanishing point according to the score value of each lane line.
In an optional embodiment of the present disclosure, the acquiring a lane line vanishing point in the road image in step 103 includes:
step S31, acquiring a high-precision map corresponding to the current pose of the vehicle, wherein the high-precision map comprises at least two lane lines corresponding to the road image;
step S32, projecting the lane lines in the high-precision map to the road image according to the current pose of the vehicle to obtain the projection of at least two lane lines;
and S33, determining a lane line vanishing point according to the intersection point of the projections of the at least two lane lines.
In the embodiment provided by the present disclosure, the lane line vanishing point in the road image may also be obtained based on a high-precision map. Specifically, a corresponding high-precision map is obtained according to the current pose of the vehicle, namely the current position and posture of the vehicle. The high-precision map includes information such as an accurate road shape, a gradient, a curvature, and a tendency of each lane. And the high-precision map comprises at least two lane lines of the road where the vehicle is located currently. Because the pose of the vehicle when acquiring the road image is different, the directions of the lane lines on the high-precision map and the lane lines in the road image acquired by the vehicle are possibly different, and therefore, the lane lines on the high-precision map are projected into the road image according to the current pose of the vehicle, and the projection of at least two lane lines can be obtained. And determining the intersection point of the projections of at least two lane lines as a lane line vanishing point. Specifically, a curve equation or a line equation corresponding to each projection may be determined according to the projections of at least two lane lines, and then the intersection coordinates of each curve equation or line equation obtained by calculation are the coordinates of the vanishing point of the lane lines. According to the projection of the lane line in the high-precision map in the road image, a more reliable lane line vanishing point can be obtained, and therefore the accuracy of the lane line detection result determined based on the lane line vanishing point is improved.
In summary, embodiments of the present disclosure provide a lane line detection method, in which a lane line vanishing point and a lane line probability map are used to perform curve fitting on a lane line together, and compared with the prior art in which a lane line curve fitting is performed only using a sampling point obtained by the lane line probability map, the method and the system disclosed herein use the property that the lane line inevitably passes through the lane line vanishing point, and perform geometric constraint on a fitting result through the lane line vanishing point, so as to overcome the problem that deviation occurs in curve fitting due to deviation of the sampling point obtained by the lane line probability map, and improve accuracy of lane line detection.
Example two
Referring to fig. 4, a structural diagram of a lane line detection apparatus in an embodiment of the present disclosure is shown, specifically as follows:
the system comprises an image acquisition module 201, a road image acquisition module, a traffic information acquisition module and a traffic information acquisition module, wherein the image acquisition module is used for acquiring a road image which comprises at least one lane line;
the image preprocessing module 202 is configured to preprocess the road image to obtain a lane line probability map corresponding to the road image;
a lane vanishing point obtaining module 203, configured to obtain a lane line vanishing point in the road image;
and the curve fitting module 204 is configured to perform curve fitting on each lane line in the road image according to the lane line vanishing point and the lane line probability map to obtain a lane line detection result.
In an optional embodiment of the present disclosure, the curve fitting module 204 includes:
the first sampling point set determining submodule is used for respectively determining a first sampling point set corresponding to each lane line in the road image according to the lane line probability map;
the curve fitting submodule is used for performing curve fitting on each lane line according to the lane line vanishing point and at least one sampling point in the first sampling point set to obtain at least one fitting curve corresponding to each lane line;
the fitting curve scoring submodule is used for determining the scoring value of each fitting curve;
and the first detection result determining submodule is used for determining the detection result of each lane line according to the score value of each fitting curve corresponding to each lane line.
In an optional embodiment of the present disclosure, the fitting curve scoring submodule includes:
the resampling unit is used for resampling each fitted curve corresponding to each lane line to obtain a second sampling point set of each fitted curve corresponding to each lane line;
a probability value determining unit, configured to determine, according to the lane line probability map, a probability value corresponding to each sampling point in the second sampling point set;
and the fitting curve scoring unit is used for calculating the sum of probability values corresponding to all sampling points in the second sampling point set of each fitting curve corresponding to each lane line to obtain the score value of each fitting curve corresponding to each lane line.
In an optional embodiment of the present disclosure, the probability value determining unit includes:
the coordinate value determining subunit is used for determining the coordinate value of each sampling point in the second sampling point set;
a coordinate point matching subunit, configured to search, in the lane line probability map, a coordinate point that is the same as the coordinate value of each sampling point, and obtain a probability value corresponding to the coordinate point, where the sampling points in the second sampling point set correspond to the coordinate points in the lane line probability map in a one-to-one manner;
and the probability value determining subunit is used for determining the probability value corresponding to the coordinate point corresponding to the sampling point as the probability value of the sampling point.
In an optional embodiment of the disclosure, the apparatus further comprises:
the random disturbance module is used for carrying out random disturbance on the lane vanishing point in a preset range to obtain at least one new lane vanishing point;
a lane vanishing point set generating module, configured to generate a lane vanishing point set according to the lane vanishing point and the at least one new lane vanishing point, where the lane vanishing point set includes the lane vanishing point and the at least one new lane vanishing point;
the curve fitting module comprises:
the lane line detection value calculation submodule is used for calculating the lane line detection value corresponding to each lane vanishing point in the lane vanishing point set respectively;
the target vanishing point determining submodule is used for determining a target lane vanishing point according to the lane detection value corresponding to each lane vanishing point in the lane vanishing point set;
and the second detection result determining submodule is used for determining a lane line detection result according to the target lane vanishing point.
In an optional embodiment of the present disclosure, the lane line detection value calculation sub-module includes:
a curve fitting unit, configured to perform curve fitting on each lane line in the road image according to any lane line vanishing point in the lane vanishing point set and the lane line probability map, to obtain a score value of each lane line corresponding to the lane vanishing point;
and the lane line detection value determining unit is used for summing the score values of each lane line to obtain the lane line detection value of the lane vanishing point.
In an optional embodiment of the present disclosure, the lane vanishing point obtaining module 203 includes:
the high-precision map acquisition sub-module is used for acquiring a high-precision map corresponding to the current pose of the vehicle, and the high-precision map comprises at least two lane lines corresponding to the road image;
the lane line projection sub-module is used for projecting lane lines in the high-precision map into the road image according to the current pose of the vehicle to obtain the projection of at least two lane lines;
and the lane line vanishing point determining submodule is used for determining the lane line vanishing point according to the intersection point of the projections of the at least two lane lines.
In summary, the embodiments of the present disclosure provide a lane line detection apparatus, which performs curve fitting on a lane line by using a lane line vanishing point and a lane line probability map, and compared with performing curve fitting on a lane line by using a sampling point obtained by using a lane line probability map in the prior art, the apparatus and the method disclosed herein use the property that a lane line inevitably passes through a lane line vanishing point, perform geometric constraint on a fitting result by using a lane line vanishing point, can overcome the problem that deviation occurs in curve fitting due to deviation of a sampling point obtained by a lane line probability map, and improve accuracy of lane line detection.
The second embodiment is a corresponding device embodiment to the first embodiment, and the detailed description may refer to the first embodiment, which is not repeated herein.
Embodiments of the present disclosure also provide an autonomous vehicle, referring to fig. 5, including: a processor 301, a memory 302 and a computer program 3021 stored on and executable on the memory 302, the processor 301 implementing the lane line detection method of the foregoing embodiment when executing the program.
Embodiments of the present disclosure also provide a readable storage medium having instructions that, when executed by a processor of an autonomous vehicle, enable the autonomous vehicle to perform the lane line detection method of the foregoing embodiments.
For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
The algorithms and displays presented herein are not inherently related to any particular computer, virtual machine, or other apparatus. Various general purpose systems may also be used with the teachings herein. The required structure for constructing such a system will be apparent from the description above. In addition, embodiments of the present disclosure are not directed to any particular programming language. It is appreciated that a variety of programming languages may be used to implement the teachings of the embodiments of the present disclosure as described herein, and any descriptions of specific languages are provided above to disclose the best modes of the embodiments of the present disclosure.
In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the disclosure may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the disclosure, various features of the embodiments of the disclosure are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be interpreted as reflecting an intention that: that is, claimed embodiments of the disclosure require more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of an embodiment of this disclosure.
Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
The various component embodiments of the disclosure may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that a microprocessor or Digital Signal Processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components in a document processing apparatus according to embodiments of the present disclosure. Embodiments of the present disclosure may also be implemented as an apparatus or device program for performing a portion or all of the methods described herein. Such programs implementing embodiments of the present disclosure may be stored on a computer readable medium or may be in the form of one or more signals. Such a signal may be downloaded from an internet website or provided on a carrier signal or in any other form.
It should be noted that the above-mentioned embodiments illustrate rather than limit embodiments of the disclosure, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. Embodiments of the disclosure may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The above description is only for the purpose of illustrating the preferred embodiments of the present disclosure and is not to be construed as limiting the embodiments of the present disclosure, and any modifications, equivalents, improvements and the like that are made within the spirit and principle of the embodiments of the present disclosure are intended to be included within the scope of the embodiments of the present disclosure.
The above description is only a specific implementation of the embodiments of the present disclosure, but the scope of the embodiments of the present disclosure is not limited thereto, and any person skilled in the art can easily conceive of changes or substitutions within the technical scope of the embodiments of the present disclosure, and all the changes or substitutions should be covered by the scope of the embodiments of the present disclosure. Therefore, the protection scope of the embodiments of the present disclosure shall be subject to the protection scope of the claims.
Claims (14)
1. A lane line detection method, comprising:
acquiring a road image, wherein the road image comprises at least two lane lines;
preprocessing the road image to obtain a lane line probability map corresponding to the road image;
acquiring a lane line vanishing point in the road image;
performing curve fitting on each lane line in the road image according to the lane line vanishing point and the lane line probability graph to obtain a lane line detection result;
after the obtaining of the lane line vanishing point in the road image, the method further includes:
randomly disturbing the lane line vanishing point within a preset range to obtain at least one new lane line vanishing point;
generating a lane line vanishing point set according to the lane line vanishing point and the at least one new lane line vanishing point, wherein the lane line vanishing point set comprises the lane line vanishing point and the at least one new lane line vanishing point;
and performing curve fitting on each lane line in the road image according to the lane line vanishing point and the lane line probability graph to obtain a lane line detection result, wherein the curve fitting comprises the following steps:
respectively calculating a lane line detection value corresponding to each lane line vanishing point in the set of lane line vanishing points;
determining a target lane line vanishing point according to a lane line detection value corresponding to each lane line vanishing point in the set of lane line vanishing points;
and determining a lane line detection result according to the target lane line vanishing point.
2. The method of claim 1, and performing curve fitting on each lane line in the road image according to the lane line vanishing point and the lane line probability graph to obtain a lane line detection result, wherein the curve fitting comprises the following steps:
respectively determining a first sampling point set corresponding to each lane line in the road image according to the lane line probability map;
performing curve fitting on each lane line according to the lane line vanishing point and at least one sampling point in the first sampling point set to obtain at least one fitting curve corresponding to each lane line;
determining the score value of each fitted curve;
and determining the detection result of each lane line according to the score value of each fitting curve corresponding to each lane line.
3. The method of claim 2, wherein determining a score value for each fitted curve comprises:
resampling each fitted curve corresponding to each lane line to obtain a second sampling point set of each fitted curve corresponding to each lane line;
determining a probability value corresponding to each sampling point in the second sampling point set according to the lane line probability map;
and calculating the sum of probability values corresponding to all sampling points in the second sampling point set of each fitting curve corresponding to each lane line to obtain the score value of each fitting curve corresponding to each lane line.
4. The method of claim 3, wherein determining a probability value corresponding to each sample point in the second set of sample points from the lane line probability map comprises:
determining a coordinate value of each sampling point in the second set of sampling points;
searching a coordinate point which is the same as the coordinate value of each sampling point in the lane line probability map, and acquiring a probability value corresponding to the coordinate point, wherein the sampling points in the second sampling point set correspond to the coordinate points in the lane line probability map in a one-to-one manner;
and determining the probability value corresponding to the coordinate point corresponding to the sampling point as the probability value of the sampling point.
5. The method according to claim 1, wherein the separately calculating the lane line detection value corresponding to each lane line vanishing point in the set of lane line vanishing points comprises:
performing curve fitting on each lane line in the road image according to any lane line vanishing point in the set of lane line vanishing points and the lane line probability map to obtain the score value of each lane line corresponding to the lane line vanishing point;
and summing the scoring values of each lane line to obtain the lane line detection value of the lane line vanishing point.
6. The method of claim 1, wherein the obtaining of the lane line vanishing point in the road image comprises:
acquiring a high-precision map corresponding to the current pose of the vehicle, wherein the high-precision map comprises at least two lane lines corresponding to the road image;
projecting the lane lines in the high-precision map into the road image according to the current pose of the vehicle to obtain the projection of the at least two lane lines;
and determining a lane line vanishing point according to the intersection point of the projections of the at least two lane lines.
7. A lane line detection apparatus, characterized in that the apparatus comprises:
the system comprises an image acquisition module, a data acquisition module and a data processing module, wherein the image acquisition module is used for acquiring a road image, and the road image comprises at least two lane lines;
the image preprocessing module is used for preprocessing the road image to obtain a lane line probability map corresponding to the road image;
the lane line vanishing point acquiring module is used for acquiring lane line vanishing points in the road image;
the curve fitting module is used for performing curve fitting on each lane line in the road image according to the lane line vanishing point and the lane line probability graph to obtain a lane line detection result;
the random disturbance module is used for carrying out random disturbance on the lane line vanishing point in a preset range to obtain at least one new lane line vanishing point;
a lane line vanishing point set generating module, configured to generate a lane line vanishing point set according to the lane line vanishing point and the at least one new lane line vanishing point, where the lane line vanishing point set includes the lane line vanishing point and the at least one new lane line vanishing point;
the curve fitting module comprises:
the lane line detection value calculation submodule is used for calculating the lane line detection value corresponding to each lane line vanishing point in the lane line vanishing point set respectively;
the target vanishing point determining submodule is used for determining a target lane line vanishing point according to a lane detection value corresponding to each lane line vanishing point in the lane line vanishing point set;
and the second detection result determining submodule is used for determining a lane line detection result according to the target lane line vanishing point.
8. The apparatus of claim 7, wherein the curve fitting module comprises:
the first sampling point set determining submodule is used for respectively determining a first sampling point set corresponding to each lane line in the road image according to the lane line probability map;
the curve fitting submodule is used for performing curve fitting on each lane line according to the lane line vanishing point and at least one sampling point in the first sampling point set to obtain at least one fitting curve corresponding to each lane line;
the fitting curve scoring submodule is used for determining the scoring value of each fitting curve;
and the first detection result determining submodule is used for determining the detection result of each lane line according to the score value of each fitting curve corresponding to each lane line.
9. The apparatus of claim 8, wherein the fitted curve scoring sub-module comprises:
the resampling unit is used for resampling each fitted curve corresponding to each lane line to obtain a second sampling point set of each fitted curve corresponding to each lane line;
a probability value determining unit, configured to determine, according to the lane line probability map, a probability value corresponding to each sampling point in the second sampling point set;
and the fitting curve scoring unit is used for calculating the sum of probability values corresponding to all sampling points in the second sampling point set of each fitting curve corresponding to each lane line to obtain the score value of each fitting curve corresponding to each lane line.
10. The apparatus of claim 9, wherein the probability value determining unit comprises:
the coordinate value determining subunit is used for determining the coordinate value of each sampling point in the second sampling point set;
a coordinate point matching subunit, configured to search a coordinate point in the lane line probability map, where the coordinate value of each sampling point is the same as the coordinate value of each sampling point, and obtain a probability value corresponding to the coordinate point, where the sampling points in the second sampling point set correspond to the coordinate points in the lane line probability map in a one-to-one manner;
and the probability value determining subunit is used for determining the probability value corresponding to the coordinate point corresponding to the sampling point as the probability value of the sampling point.
11. The apparatus of claim 7, wherein the lane line detection value calculation sub-module comprises:
a curve fitting unit, configured to perform curve fitting on each lane line in the road image according to any lane line vanishing point in the set of lane line vanishing points and the lane line probability map, to obtain a score value of each lane line corresponding to the lane line vanishing point;
and the lane line detection value determining unit is used for summing the score values of each lane line to obtain the lane line detection value of the lane line vanishing point.
12. The apparatus of claim 7, wherein the lane line vanishing point obtaining module comprises:
the high-precision map acquisition sub-module is used for acquiring a high-precision map corresponding to the current pose of the vehicle, wherein the high-precision map comprises at least two lane lines corresponding to the road image;
the lane line projection sub-module is used for projecting lane lines in the high-precision map into the road image according to the current pose of the vehicle to obtain the projection of at least two lane lines;
and the lane line vanishing point determining submodule is used for determining the lane line vanishing point according to the intersection point of the projections of the at least two lane lines.
13. An autonomous vehicle, comprising:
a processor, a memory and a computer program stored on the memory and executable on the processor, wherein the processor when executing the computer program implements the lane line detection method according to any of claims 1-6.
14. A readable storage medium, wherein instructions in the storage medium, when executed by a processor of an autonomous vehicle, enable the autonomous vehicle to perform the lane line detection method of any of method claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110229709.9A CN113011285B (en) | 2021-03-02 | 2021-03-02 | Lane line detection method and device, automatic driving vehicle and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110229709.9A CN113011285B (en) | 2021-03-02 | 2021-03-02 | Lane line detection method and device, automatic driving vehicle and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113011285A CN113011285A (en) | 2021-06-22 |
| CN113011285B true CN113011285B (en) | 2023-04-07 |
Family
ID=76402255
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110229709.9A Active CN113011285B (en) | 2021-03-02 | 2021-03-02 | Lane line detection method and device, automatic driving vehicle and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113011285B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11776282B2 (en) * | 2021-03-26 | 2023-10-03 | Here Global B.V. | Method, apparatus, and system for removing outliers from road lane marking data |
| CN115019268A (en) * | 2022-05-30 | 2022-09-06 | 深圳市商汤科技有限公司 | Lane line detection method and device, computer equipment and storage medium |
| CN114973180B (en) * | 2022-07-18 | 2022-11-01 | 福思(杭州)智能科技有限公司 | Lane line tracking method, device, equipment and storage medium |
| CN117719515B (en) * | 2024-01-05 | 2024-06-25 | 深圳技术大学 | Lane departure warning method and system |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109284674B (en) * | 2018-08-09 | 2020-12-08 | 浙江大华技术股份有限公司 | Method and device for determining lane line |
| CN111191487A (en) * | 2018-11-14 | 2020-05-22 | 北京市商汤科技开发有限公司 | Lane line detection and driving control method and device and electronic equipment |
| CN111476062A (en) * | 2019-01-23 | 2020-07-31 | 北京市商汤科技开发有限公司 | Lane line detection method and device, electronic equipment and driving system |
-
2021
- 2021-03-02 CN CN202110229709.9A patent/CN113011285B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113011285A (en) | 2021-06-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113011285B (en) | Lane line detection method and device, automatic driving vehicle and readable storage medium | |
| CN112102369B (en) | Autonomous inspection method, device, equipment and storage medium for water surface floating target | |
| CN101609504B (en) | Method for detecting, distinguishing and locating infrared imagery sea-surface target | |
| JP7138718B2 (en) | Feature detection device, feature detection method, and feature detection program | |
| Hinz | Detection and counting of cars in aerial images | |
| CN110658539B (en) | Vehicle positioning method, device, vehicle and computer readable storage medium | |
| EP2887315B1 (en) | Camera calibration device, method for implementing calibration, program and camera for movable body | |
| CN109657686A (en) | Lane line generation method, device, equipment and storage medium | |
| CN103770704A (en) | System and method for recognizing parking space line markings for vehicle | |
| EP3594902A1 (en) | Method for estimating a relative position of an object in the surroundings of a vehicle and electronic control unit for a vehicle and vehicle | |
| CN117576652B (en) | Road object identification method and device, storage medium and electronic equipment | |
| WO2021017211A1 (en) | Vehicle positioning method and device employing visual sensing, and vehicle-mounted terminal | |
| CN112084835A (en) | Generating map features based on aerial data and telemetry data | |
| CN110751040A (en) | Three-dimensional object detection method and device, electronic equipment and storage medium | |
| CN111256693A (en) | Pose change calculation method and vehicle-mounted terminal | |
| KR102003387B1 (en) | Method for detecting and locating traffic participants using bird's-eye view image, computer-readerble recording medium storing traffic participants detecting and locating program | |
| CN115457130A (en) | Electric vehicle charging port detection and positioning method based on depth key point regression | |
| CN113989359A (en) | Target diameter automatic measurement method and system based on visual information | |
| US20230350418A1 (en) | Position determination by means of neural networks | |
| CN116912688A (en) | Sea surface target detection method and system based on laser radar and camera fusion | |
| CN116343143A (en) | Target detection method, storage medium, road side equipment and automatic driving system | |
| JP7427569B2 (en) | Condition determination device, condition determination system, and condition determination method | |
| JP2005148784A (en) | Road lane detection device | |
| CN109690555A (en) | Face detector based on curvature | |
| CN118587244B (en) | Vehicle track prediction method, device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |