CN115713740A - Lane line detection method and system based on key points, electronic equipment and vehicle - Google Patents
Lane line detection method and system based on key points, electronic equipment and vehicle Download PDFInfo
- Publication number
- CN115713740A CN115713740A CN202211379395.1A CN202211379395A CN115713740A CN 115713740 A CN115713740 A CN 115713740A CN 202211379395 A CN202211379395 A CN 202211379395A CN 115713740 A CN115713740 A CN 115713740A
- Authority
- CN
- China
- Prior art keywords
- lane line
- offset information
- points
- line detection
- key
- 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.)
- Pending
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a method, a system, electronic equipment and a vehicle for detecting lane lines based on key points, wherein the method comprises the following steps: acquiring image data about a lane line; acquiring a characteristic diagram through a backbone network of a lane line detection model and a decoder based on the image data; based on the characteristic diagram, respectively predicting the position information of a key point, first offset information of the key point due to down sampling and second offset information of the key point from the starting point of the lane line by using a task head of a lane line detection model; forming a plurality of groups of determination points based on the position information, the first offset information and the second offset information; and forming a plurality of corresponding lane lines based on the plurality of groups of determined points. The detection of the lane lines is not required to be based on segmentation and is only based on key point detection, the post-processing is faster, and the model reasoning speed is improved.
Description
Technical Field
The invention relates to the technical field of target detection, in particular to a method and a system for detecting lane lines based on key points, electronic equipment and a vehicle.
Background
Lane line detection is an environmental awareness application that aims to detect lane lines through a vehicle-mounted camera or lidar. In recent years, with the development and landing of computer vision applications, lane line detection tasks have attracted much attention, and a series of lane line detection methods have appeared. In the existing mode, the method based on segmentation is long in time consumption, time consumption is more due to the fact that the method based on key points is often required to be combined with the method based on segmentation, and the method based on polynomial regression is sensitive to the shape of a lane line.
The information disclosed in this background section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art that is already known to a person skilled in the art.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a method and a system for detecting a lane line based on key points, electronic equipment and a vehicle.
The invention provides a method for detecting a lane line based on key points, which comprises the following steps:
acquiring image data about a lane line;
acquiring a characteristic diagram through a backbone network of a lane line detection model and a decoder based on the image data;
based on the characteristic diagram, respectively predicting the position information of a key point, first offset information of the key point due to down sampling and second offset information of the key point from the starting point of the lane line by using a task head of a lane line detection model;
forming a plurality of groups of determination points based on the position information, the first offset information and the second offset information;
and forming a plurality of corresponding lane lines based on the plurality of groups of determined points.
According to the method for detecting the lane line based on the key points, provided by the invention, when the lane line detection model is trained,
based on the feature map, respectively predicting the position information of the key point, the first offset information of the key point due to down-sampling, and the second offset information of the key point from the starting point of the lane line by using the task head of the lane line detection model, and further comprising:
predicting third offset information from the key point to the image central point through a task head of the lane line detection model based on the feature map;
adding an average absolute error based on the third offset information to a loss function of the lane line detection model with respect to a task head.
According to the method for detecting the lane line based on the key points, provided by the invention, when the lane line detection model is trained,
based on the feature map, respectively predicting the position information of the key point, the first offset information of the key point due to down-sampling, and the second offset information of the key point from the starting point of the lane line by using the task head of the lane line detection model, and further comprising:
predicting direction vectors from the key points to adjacent key points of the key points through a task head of a lane line detection model based on the feature map;
adding a mean square error based on the direction vector to a loss function of the lane line detection model with respect to a task head.
According to the method for detecting the lane line based on the key point, provided by the invention, a plurality of groups of determined points are formed based on the position information, the first offset information and the second offset information, and the method comprises the following steps:
transforming all the key points into a preliminary starting point based on the position information and the second offset information;
clustering the prepared starting points obtained by transformation to form a plurality of starting point groups;
and carrying out position averaging on the prepared starting points in each starting point group to serve as the final starting point of the corresponding lane line.
According to the method for detecting the lane line based on the key points, the position averaging is carried out on the prepared starting points in each group of the starting point groups to serve as the final starting points of the corresponding lane lines, and the method comprises the following steps:
correcting the preparation starting points in each starting point group based on the first offset information;
and carrying out position averaging on the prepared starting points in each corrected starting point group to obtain the final starting point.
According to the method for detecting the lane line based on the key point, provided by the invention, a plurality of groups of determined points are formed based on the position information, the first offset information and the second offset information, and the method further comprises the following steps:
aiming at each key point, acquiring the final starting point which is closest to the preparation starting point corresponding to the key point;
grouping the key points into a group with the final starting point closest to the distance;
and forming a plurality of groups of determined points after all the key points are grouped.
The invention also provides a system for detecting the lane line based on the key points, which comprises:
an acquisition module for acquiring image data about a lane line;
the preprocessing module is used for acquiring a characteristic diagram through a backbone network of a lane line detection model and a decoder based on the image data;
the prediction module is used for respectively predicting the position information of the key points, the first offset information of the key points caused by downsampling and the second offset information of the key points from the starting point of the lane line on which the key points are located through the task head of the lane line detection model based on the feature map;
a post-processing module for forming a plurality of groups of determination points based on the position information, the first offset information and the second offset information;
and the lane line forming module is used for forming a plurality of corresponding lane lines based on the plurality of groups of determined points.
The invention also provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the program to realize the steps of the method for detecting the lane line based on the key point.
The invention also provides a vehicle which can realize auxiliary driving and/or automatic driving, and the vehicle comprises the steps of the key point-based lane line detection method.
According to the method, the system, the electronic equipment and the vehicle for detecting the lane line based on the key points, provided by the invention, the detection of the lane line is not required to be based on the segmentation and only based on the key point detection, the post-processing is faster, and the model reasoning speed is improved.
Drawings
In order to more clearly illustrate the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a schematic flow chart of a method for detecting a lane line based on key points according to the present invention;
fig. 2 is a schematic structural diagram of a lane line detection system based on key points according to the present invention;
fig. 3 is a schematic physical structure diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The following describes in detail a method for detecting a lane line based on a key point according to an embodiment of the present application with reference to the accompanying drawings.
Fig. 1 is a schematic flow chart of a method for detecting a lane line based on key points according to the present invention, and as shown in fig. 1, the method for detecting a lane line based on key points according to the present invention may include the following steps.
S100, acquiring image data about the lane line.
Preferably, the image data is acquired by an image sensor mounted on the vehicle.
Further, the image sensor includes a camera, an infrared sensor, a millimeter wave radar, and the like.
S200, acquiring a characteristic diagram through a backbone network of the lane line detection model and a decoder based on the image data.
Preferably, the backbone network (backbone) comprises a RegNet network.
Preferably, the decoder comprises a Feature Pyramid Network (FPN).
And S300, respectively predicting the position information of the key point, the first offset information of the key point caused by down-sampling and the second offset information of the key point from the starting point of the lane line on which the key point is positioned by the task head of the lane line detection model based on the feature map.
Preferably, the position information of the key point, the first offset information of the key point due to downsampling, and the second offset information of the key point from the start point of the lane line where the key point is located are predicted by the three corresponding task head branches respectively.
Preferably, when the task head predicts the position information of the key points, the embodiment of the invention determines all the key points in the image based on one heat map.
Further, for the branch used to predict the location information of the key points, the lane lines are first equally divided by y-coordinate (y-axis of the vehicle coordinate system) as the ordinate of the plurality of key points, then the heat maps of the key points are predicted, and finally a maximum pooling of 1 × 3 is employed to obtain the highest scoring pixel location in the heat maps in the neighborhood, i.e., the key point.
Preferably, the branch used to predict the location information of the key point during training adds a focal loss (focal loss) based on the location information to a loss function of the lane line detection model with respect to the task head.
Preferably, the branch used for predicting the first offset information of the key point due to the down-sampling during the training adds the average absolute error (L1 loss) based on the first offset information to the loss function of the lane line detection model with respect to the task head.
Preferably, the branch of the second offset information used for predicting the distance between the key point and the start point of the lane line is trained, and the average absolute error (L1 loss) based on the second offset information is added to the loss function of the lane line detection model about the task head.
Preferably, the first offset information caused by the down-sampling refers to an error caused by the loss of the fine position information of the key point due to rounding when the feature map is subjected to the down-sampling, and the error is amplified to form an offset after the up-sampling.
It should be noted that S300 is a step that is implemented during both training and applying inference. Further, the embodiment of the invention also discloses the following auxiliary branch special for training.
Alternatively, the lane marking detection model may be trained,
based on the characteristic diagram, respectively predicting the position information of the key points, the first offset information of the key points caused by downsampling and the second offset information of the key points from the starting point of the lane line by the task head of the lane line detection model, and further comprising the following steps of:
predicting third offset information from the key point to the image central point through a task head of the lane line detection model based on the feature map;
the average absolute error (L1 loss) based on the third offset information is added to a loss function of the lane line detection model with respect to the task head.
Alternatively, the lane line detection model may be trained,
based on the feature map, respectively predicting the position information of the key point, the first offset information of the key point caused by down-sampling and the second offset information of the key point from the starting point of the lane line by using the task head of the lane line detection model, and further comprising:
predicting direction vectors from the key points to adjacent key points through a task head of the lane line detection model based on the characteristic diagram;
the mean square error (MSE loss) based on the direction vector is added to the loss function of the lane line detection model with respect to the task head.
Preferably, predicting direction vectors of the key points to their neighboring key points by a task head of the lane line detection model includes:
and predicting direction vectors from the key points to 2 adjacent key points through a task head of the lane line detection model.
Preferably, the samples used in the training include the key point labels on the lane line and the starting point labels of the lane line.
Preferably, the annotation starting point is realized by acquiring an intersection point of the lane line and the lower boundary of the image. In contrast, when applying inference to the lane marking detection model, the starting point of the lane marking is additionally predicted.
It should be noted that, a special training auxiliary branch is designed during training, and information obtained by model training is increased through third offset information from the key point to the image center point and direction vectors from the key point to the adjacent key points, so that the robustness of the model is better, and the training effect is improved.
S400, forming a plurality of groups of determined points based on the position information, the first offset information and the second offset information.
Optionally, forming several groups of determined points based on the position information, the first offset information and the second offset information, including:
transforming all the key points into a preliminary starting point based on the position information and the second offset information;
clustering the prepared starting points obtained by transformation to form a plurality of starting point groups;
and carrying out position averaging on the prepared starting points in each group of starting point groups to serve as the final starting points of the corresponding lane lines.
Optionally, performing position averaging on the prepared starting points in each group of starting points to serve as a final starting point of the corresponding lane line, includes:
correcting the preparation starting points in each group of the starting point groups based on the first offset information;
and carrying out position averaging on the prepared starting points in each group of corrected starting point groups to obtain a final starting point.
Optionally, forming several groups of determination points based on the position information, the first offset information and the second offset information, further comprising:
aiming at each key point, acquiring a final starting point which is closest to the preparation starting point corresponding to the key point;
dividing the key points and the final starting points with the nearest distance into a group;
after all the key points are grouped, a plurality of groups of determined points are formed.
S500, determining points based on the groups to form a plurality of corresponding lane lines.
Preferably, each group of the determined points is connected from small to large according to the y coordinate, and the shape of the lane line corresponding to the group of the determined points can be obtained.
The embodiment detects the lane lines without being based on segmentation and only based on key point detection, so that the post-processing is faster, and the model reasoning speed is improved.
The following describes the key point-based lane line detection system provided by the present invention, and the key point-based lane line detection system described below and the key point-based lane line detection method described above may be referred to in correspondence with each other.
Fig. 2 is a schematic structural view of a lane line detection system based on a key point, as shown in fig. 2, the lane line detection system based on a key point further provided by the present invention includes:
an acquisition module for acquiring image data about a lane line;
the preprocessing module is used for acquiring a characteristic diagram through a backbone network of the lane line detection model and a decoder based on the image data;
the prediction module is used for respectively predicting the position information of the key point, first offset information of the key point caused by down sampling and second offset information of the key point from the starting point of the lane line on which the key point is positioned on the basis of the characteristic diagram by a task head of the lane line detection model;
a post-processing module for forming a plurality of groups of determination points based on the position information, the first offset information and the second offset information;
and the lane line forming module is used for forming a plurality of corresponding lane lines based on the plurality of groups of determined points.
The embodiment detects the lane lines without being based on segmentation and only based on key point detection, so that the post-processing is faster, and the model reasoning speed is improved.
Fig. 3 is a schematic physical structure diagram of an electronic device provided in the present invention, and as shown in fig. 3, the electronic device may include: a processor (processor) 310, a communication Interface (communication Interface) 320, a memory (memory) 330 and a communication bus 340, wherein the processor 310, the communication Interface 320 and the memory 330 communicate with each other via the communication bus 340. The processor 310 may invoke logic instructions in the memory 330 to perform a keypoint-based lane line detection method, the method comprising:
acquiring image data about a lane line;
acquiring a characteristic diagram through a backbone network of a lane line detection model and a decoder based on the image data;
based on the characteristic diagram, respectively predicting the position information of a key point, first offset information of the key point due to down sampling and second offset information of the key point from the starting point of the lane line by using a task head of a lane line detection model;
forming a plurality of groups of determination points based on the position information, the first offset information, and the second offset information;
and forming a plurality of corresponding lane lines based on the plurality of groups of determined points.
In addition, the logic instructions in the memory 330 may be implemented in the form of software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk, and various media capable of storing program codes.
The invention also provides a vehicle which can realize auxiliary driving and/or automatic driving, and the vehicle comprises the steps of any one of the key point-based lane line detection methods.
In another aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the keypoint-based lane line detection method provided by the above methods, the method comprising:
acquiring image data about a lane line;
acquiring a characteristic diagram through a backbone network of a lane line detection model and a decoder based on the image data;
based on the characteristic diagram, respectively predicting the position information of a key point, first offset information of the key point due to down sampling and second offset information of the key point from the starting point of the lane line by using a task head of a lane line detection model;
forming a plurality of groups of determination points based on the position information, the first offset information and the second offset information;
and forming a plurality of corresponding lane lines based on the plurality of groups of determined points.
In yet another aspect, the present invention also provides a non-transitory computer-readable storage medium having stored thereon a computer program, which when executed by a processor is implemented to perform the above-provided keypoint-based lane line detection methods, the method comprising:
acquiring image data about a lane line;
acquiring a feature map through a backbone network of a lane line detection model and a decoder based on the image data;
based on the characteristic diagram, respectively predicting the position information of a key point, first offset information of the key point due to down sampling and second offset information of the key point from the starting point of the lane line by using a task head of a lane line detection model;
forming a plurality of groups of determination points based on the position information, the first offset information and the second offset information;
and forming a plurality of corresponding lane lines based on the plurality of groups of determined points.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment may be implemented by software plus a necessary general hardware platform, and may also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for detecting lane lines based on key points is characterized by comprising the following steps:
acquiring image data about a lane line;
acquiring a feature map through a backbone network of a lane line detection model and a decoder based on the image data;
based on the characteristic diagram, respectively predicting the position information of a key point, first offset information of the key point due to down sampling and second offset information of the key point from the starting point of the lane line by using a task head of a lane line detection model;
forming a plurality of groups of determination points based on the position information, the first offset information, and the second offset information;
and forming a plurality of corresponding lane lines based on the plurality of groups of determined points.
2. The keypoint-based lane line detection method of claim 1, wherein said lane line detection model, when trained,
based on the feature map, respectively predicting the position information of the key point, the first offset information of the key point due to down-sampling, and the second offset information of the key point from the starting point of the lane line by using the task head of the lane line detection model, and further comprising:
predicting third offset information from the key point to the image central point through a task head of the lane line detection model based on the feature map;
adding an average absolute error based on the third offset information to a loss function of the lane line detection model with respect to a task head.
3. The keypoint-based lane line detection method of claim 1, wherein said lane line detection model, when trained,
based on the feature map, respectively predicting the position information of the key points, the first offset information of the key points caused by downsampling and the second offset information of the key points from the starting point of the lane line by the task head of the lane line detection model, and further comprising:
predicting direction vectors from the key points to adjacent key points of the key points through a task head of the lane line detection model based on the feature map;
adding a mean square error based on the direction vector to a loss function of the lane marking detection model with respect to a task head.
4. The keypoint-based lane line detection method of claim 1, wherein forming several groups of determined points based on said location information, said first offset information, and said second offset information comprises:
transforming all the key points into a preliminary starting point based on the position information and the second offset information;
clustering the prepared starting points obtained by transformation to form a plurality of starting point groups;
and carrying out position averaging on the prepared starting points in each starting point group to serve as the final starting point of the corresponding lane line.
5. The method of claim 4, wherein the performing a position average of the preliminary starting points in each set of the starting point groups as a final starting point of the corresponding lane line comprises:
correcting the preparation starting points in each starting point group based on the first offset information;
and carrying out position averaging on the prepared starting points in each corrected starting point group to obtain the final starting point.
6. The keypoint-based lane line detection method of claim 4, wherein forming several groups of determined points based on said position information, said first offset information, and said second offset information, further comprises:
aiming at each key point, acquiring the final starting point which is closest to the preparation starting point corresponding to the key point;
grouping the key points into a group with the final starting point closest to the distance;
and forming a plurality of groups of determined points after all the key points are grouped.
7. A keypoint-based lane line detection system, the system comprising:
an acquisition module for acquiring image data about a lane line;
the preprocessing module is used for acquiring a characteristic diagram through a backbone network of a lane line detection model and a decoder based on the image data;
the prediction module is used for respectively predicting the position information of the key points, the first offset information of the key points caused by downsampling and the second offset information of the key points from the starting point of the lane line on which the key points are located through the task head of the lane line detection model based on the feature map;
a post-processing module for forming a plurality of groups of determination points based on the position information, the first offset information and the second offset information;
and the lane line forming module is used for forming a plurality of corresponding lane lines based on the plurality of groups of determined points.
8. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program performs the steps of the keypoint-based lane line detection method of any of claims 1 to 6.
9. A vehicle capable of assisted driving and/or autonomous driving, characterized in that it comprises an electronic device as claimed in claim 8.
10. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor, performs the steps of the keypoint-based lane line detection method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211379395.1A CN115713740A (en) | 2022-11-04 | 2022-11-04 | Lane line detection method and system based on key points, electronic equipment and vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211379395.1A CN115713740A (en) | 2022-11-04 | 2022-11-04 | Lane line detection method and system based on key points, electronic equipment and vehicle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115713740A true CN115713740A (en) | 2023-02-24 |
Family
ID=85232374
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211379395.1A Pending CN115713740A (en) | 2022-11-04 | 2022-11-04 | Lane line detection method and system based on key points, electronic equipment and vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115713740A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116229406A (en) * | 2023-05-09 | 2023-06-06 | 华东交通大学 | Lane line detection method, system, electronic equipment and storage medium |
-
2022
- 2022-11-04 CN CN202211379395.1A patent/CN115713740A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116229406A (en) * | 2023-05-09 | 2023-06-06 | 华东交通大学 | Lane line detection method, system, electronic equipment and storage medium |
| CN116229406B (en) * | 2023-05-09 | 2023-08-25 | 华东交通大学 | Lane line detection method, system, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107972662B (en) | Vehicle forward collision early warning method based on deep learning | |
| CN111666921B (en) | Vehicle control method, apparatus, computer device, and computer-readable storage medium | |
| CN108345875B (en) | Driving region detection model training method, detection method and device | |
| US9811742B2 (en) | Vehicle-surroundings recognition device | |
| US10853964B2 (en) | Image recognition system | |
| CN112441022A (en) | Lane center line determining method and device | |
| US20200394838A1 (en) | Generating Map Features Based on Aerial Data and Telemetry Data | |
| CN112528807B (en) | Method and device for predicting running track, electronic equipment and storage medium | |
| CN111583716B (en) | Vehicle obstacle avoidance method and device, electronic equipment and storage medium | |
| CN116783620A (en) | Efficient three-dimensional object detection from point clouds | |
| CN112721950A (en) | Autonomous vehicle control method and system based on deep learning | |
| CN115713740A (en) | Lane line detection method and system based on key points, electronic equipment and vehicle | |
| CN113255445A (en) | Multitask model training and image processing method, device, equipment and storage medium | |
| CN114841910A (en) | Vehicle-mounted lens shielding identification method and device | |
| CN114155491B (en) | Ship behavior recognition method and system based on AIS data | |
| CN114495050A (en) | Multitask integrated detection method for automatic driving forward vision detection | |
| CN112597996B (en) | Method for detecting traffic sign significance in natural scene based on task driving | |
| CN112654998B (en) | Lane line detection method and device | |
| CN115083199A (en) | Parking space information determination method and related equipment thereof | |
| CN112733678A (en) | Ranging method, ranging device, computer equipment and storage medium | |
| CN111144361A (en) | Road lane detection method based on binaryzation CGAN network | |
| CN111079634A (en) | Method, device and system for detecting obstacle in vehicle running and vehicle | |
| Ben Romdhane et al. | A lane detection and tracking method for driver assistance system | |
| CN114170275A (en) | Lane line processing method and system based on Kalman filtering | |
| CN114550139A (en) | Lane line detection method and device |
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 |