CN111401186A - Vehicle line pressing detection system and method - Google Patents
Vehicle line pressing detection system and method Download PDFInfo
- Publication number
- CN111401186A CN111401186A CN202010161355.4A CN202010161355A CN111401186A CN 111401186 A CN111401186 A CN 111401186A CN 202010161355 A CN202010161355 A CN 202010161355A CN 111401186 A CN111401186 A CN 111401186A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- lane line
- image
- camera
- contour
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000001514 detection method Methods 0.000 title claims abstract description 33
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000013135 deep learning Methods 0.000 claims description 6
- 238000013519 translation Methods 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000012360 testing method Methods 0.000 abstract description 8
- 230000003137 locomotive effect Effects 0.000 description 11
- 230000007547 defect Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 2
- 239000012535 impurity Substances 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
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
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
- G06F18/24137—Distances to cluster centroïds
- G06F18/2414—Smoothing the distance, e.g. radial basis function networks [RBFN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/08—Detecting or categorising vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a vehicle line-pressing detection system and a method, wherein the detection system comprises: the first camera collects a first image comprising a vehicle head edge and a lane line located at the vehicle head position, the second camera collects a second image comprising a vehicle tail edge and the lane line located at the vehicle tail position, the third camera collects a third image comprising a vehicle front wheel, a vehicle rear wheel, a vehicle side body outline and the lane line located on the side face of the vehicle body, the control unit respectively identifies the lane line and the vehicle outline from the first image, the second image and the third image, and whether the vehicle is pressed or not is determined by judging whether a superposition area exists between a region where the lane line is located and a region where the vehicle outline is located or not. The invention is based on the camera installed on the vehicle, and detects whether the vehicle presses the line by acquiring the vehicle edge data and the lane line data in the running process of the vehicle in real time, thereby improving the accuracy of judging whether the vehicle is pressed in the test.
Description
Technical Field
The invention relates to the technical field of driving tests, in particular to a vehicle line pressing detection system and method.
Background
With the development of driving examination system equipment, compared with a traditional driving examination mode of accompanying driving supervision examination, the intelligent driving examination method has the advantages that the judgment of human factors is reduced to the maximum extent, the examination fairness, the justice and the publicity are realized, the examination scores of examinees are remotely supervised and intelligently judged, and the examination efficiency is improved.
The existing driver test is based on an RTK (Real-time kinematic) reference station and is realized by using a satellite positioning mode. Specifically, the satellite positioning method is as follows: the vehicle model is used for simulating an actual examination vehicle, the running track of the actual examination vehicle on an actual examination field is simulated by utilizing the running track of the vehicle model on an examination field map through positioning the actual examination vehicle, and whether the actual examination vehicle is qualified or not is determined by judging whether the vehicle model presses a line on the examination field map or not.
However, there are many disadvantages to the satellite positioning approach, including: 1) the vehicle model is easy to cheat, for example, the parameters of the site map are modified to enlarge the site map, or the size of the vehicle model is reduced, so that the vehicle model is not easy to press lines, and the difficulty of the examination is reduced. In addition, when the vehicle model is used for simulating the driving track of the vehicle in the actual test, the driving track simulated in the test record can also be used as the driving track which is already tested before, or other false tracks, so false test behaviors are easy to occur. 2) The satellite signal is easily influenced by the environment and the weather, and when trees or buildings on an examination site are high and/or the weather is severe, the satellite signal is easily shielded, so that the satellite signal is weakened, and the satellite positioning is delayed or inaccurate.
Disclosure of Invention
In view of the above, the invention discloses a vehicle line pressing detection system and method, which are used for detecting whether a vehicle presses a line or not by acquiring vehicle edge data and lane line data in a vehicle running process in real time based on a camera installed on the vehicle, so that the accuracy of judging whether the vehicle is pressed the line or not in an examination is improved, and various defects existing in a satellite positioning mode are effectively avoided.
A vehicle wire press detection system comprising: the system comprises a first camera, a second camera, a third camera and a control unit;
the first camera is arranged at the head position and used for collecting a first image comprising a head edge and a lane line positioned at the head position;
the second camera is arranged at the tail position of the vehicle and used for acquiring a second image comprising the edge of the tail of the vehicle and a lane line positioned at the tail position of the vehicle;
the third camera is used for the roof of setting, gathers and includes: a third image of the front wheels, the rear wheels, the profile of the side body of the vehicle, and the lane lines on the side of the vehicle body;
the control unit is respectively connected with the first camera, the second camera and the third camera, and is used for acquiring the first image, the second image and the third image, respectively identifying a lane line and a vehicle contour from the first image, the second image and the third image, judging whether a superposition area exists between an area where the lane line is located and an area where the vehicle contour is located, if so, judging that the vehicle is pressed, and if not, judging that the vehicle is not pressed.
Optionally, the first camera includes: a first left front camera and a first right front camera;
the first front left camera is arranged at the front left position of the vehicle head and used for collecting images including a front left vehicle head edge and a lane line positioned at the front left position of the vehicle head;
the first front right camera is used for being arranged at the front right position of the vehicle head and collecting images including the front right vehicle head edge and a lane line located at the front right position of the vehicle head.
Optionally, the second camera includes: a second left rear camera and a second right rear camera;
the second left rear camera is used for being arranged at the left rear position of the tail of the vehicle and acquiring an image comprising the edge of the left rear tail of the vehicle and a lane line positioned at the left rear position of the tail of the vehicle;
the second right rear camera is used for being arranged at the rear right rear position of the vehicle tail and collecting images including the rear edge of the right rear vehicle tail and a lane line positioned at the rear right rear position of the vehicle tail.
Optionally, the third camera includes: a third left front camera, a third right front camera, a third left rear camera, and a third right rear camera;
the third left front camera is arranged at the left front position of the roof and used for collecting images including left front wheels, a left side body outline of the vehicle and lane lines at the positions of the left front wheels and the left side body outline of the vehicle;
the third front right camera is used for being arranged at the front right position of the roof and collecting images including a front right wheel, a vehicle right side body outline and a lane line positioned at the positions of the front right wheel and the vehicle right side body outline;
the third left rear camera is arranged at the left rear position of the roof and used for collecting images including a left rear wheel, a vehicle left side body outline and a lane line at the positions of the left rear wheel and the vehicle left side body outline;
the third right rear camera is used for being arranged at the right rear position of the roof and collecting images including a right rear wheel, a vehicle right side body outline and lane lines located at the positions of the right rear wheel and the vehicle right side body outline.
A vehicle line pressing detection method is applied to a control unit in the vehicle line pressing detection system, and comprises the following steps:
the method comprises the steps of obtaining a first image collected by a first camera, a second image collected by a second camera and a third image collected by a third camera, wherein the first image comprises a vehicle head edge and a lane line located at the vehicle head position, and the second image comprises: a vehicle rear edge and a lane line at the vehicle rear position, the third image comprising: the vehicle comprises a front vehicle wheel, a rear vehicle wheel, a side body outline of the vehicle and a lane line positioned on the side surface of the vehicle body;
identifying lane lines and vehicle contours from the first image, the second image and the third image, respectively;
judging whether a superposition area exists in the area where the lane line is located and the area where the vehicle outline is located;
if yes, judging that the vehicle is pressed;
if not, the vehicle is judged not to be pressed.
Optionally, the identifying the lane line and the vehicle contour from the first image, the second image, and the third image respectively specifically includes:
correcting the first image, the second image and the third image respectively to obtain a first target image, a second target image and a third target image;
separating a lane line region and a vehicle contour region from the first target image, the second target image and the third target image respectively by adopting a deep learning semantic segmentation method;
and respectively carrying out edge calculation on the lane line area and the vehicle contour area, and identifying the lane line and the vehicle contour from a pixel level.
Optionally, after determining that the vehicle is not pressed, the method may further include:
translating the area of the vehicle contour to the outer side of the vehicle contour until the area of the vehicle contour intersects with a lane line;
recording the number of pixel points of the vehicle contour translation;
and based on the number of the pixel points, obtaining the distance between the vehicle outline and the lane line according to the corresponding relation between the known lane line width and the number of the pixel points corresponding to the lane line width.
Optionally, after the performing edge calculation on the lane line region and the vehicle contour region respectively and identifying the lane line and the vehicle contour from a pixel level, the method further includes:
judging whether the identified lane line is complete or not;
if the lane line is incomplete, distinguishing the lane line and a lane line shelter from the pixel level;
judging whether the lane line shelter shields one side of the lane line closest to the vehicle outline or not;
if the lane line shelter blocks one side of the lane line closest to the vehicle outline, distinguishing the type of the lane line by adopting a deep learning classification method;
and removing the lane line shielding object from the lane line image, and drawing the missing part of the lane line based on the type of the lane line to obtain the complete lane line.
From the above technical solution, the present invention discloses a vehicle line pressing detection system and method, wherein the detection system comprises: first camera, second camera, third camera and the control unit, first camera setting is in the locomotive position, gathers including the locomotive edge with be located the first image of the lane line in locomotive position, and the second camera is used for setting up in rear of a vehicle position, gathers including the rear of a vehicle edge with be located the second image of the lane line in rear of a vehicle position, and the third camera is used for the roof of setting, gathers and includes: the control unit respectively identifies the lane line and the vehicle contour from the first image, the second image and the third image, and determines whether the vehicle is pressed by judging whether a superposition area exists between an area where the lane line is located and an area where the vehicle contour is located. Compared with the traditional scheme that satellite positioning is adopted to determine whether the vehicle presses the line, the method greatly improves the accuracy of judging whether the vehicle is pressed in an examination, and therefore various defects existing in a satellite positioning mode are effectively avoided.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the disclosed drawings without creative efforts.
FIG. 1 is a schematic structural diagram of a vehicle wire pressing detection system according to an embodiment of the present invention;
FIG. 2 is a test chart for vehicle camera installation disclosed in the embodiment of the present invention;
fig. 3 is a flowchart of a vehicle wire pressing detection method according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. 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 embodiment of the invention discloses a vehicle line pressing detection system and a method, wherein the detection system comprises: first camera, second camera, third camera and the control unit, first camera setting is in the locomotive position, gathers including the locomotive edge with be located the first image of the lane line in locomotive position, and the second camera is used for setting up in rear of a vehicle position, gathers including the rear of a vehicle edge with be located the second image of the lane line in rear of a vehicle position, and the third camera is used for the roof of setting, gathers and includes: the control unit respectively identifies the lane line and the vehicle contour from the first image, the second image and the third image, and determines whether the vehicle is pressed by judging whether a superposition area exists between an area where the lane line is located and an area where the vehicle contour is located. Compared with the traditional scheme that satellite positioning is adopted to determine whether the vehicle presses the line, the method greatly improves the accuracy of judging whether the vehicle is pressed in an examination, and therefore various defects existing in a satellite positioning mode are effectively avoided.
Referring to fig. 1, a schematic structural diagram of a vehicle wire pressing detection system according to an embodiment of the present invention includes: a first camera 11, a second camera 12, a third camera 13 and a control unit 14.
Wherein:
the first camera 11 is used for being arranged at the head position and collecting a first image which comprises a head edge and a lane line located at the head position.
Optionally, the first camera 11 is a wide-angle camera.
In order to ensure that the first image acquired by the first camera 11 can contain more vehicle head edges and lane lines located at the vehicle head position, in practical application, the first camera 11 irradiates downwards, preferably vertically downwards.
Specifically, referring to fig. 2, in an embodiment of the vehicle camera installation test chart disclosed in the present invention, in practical applications, the first camera 11 may include but is not limited to: a first left front camera 111 and a first right front camera 112;
the first front left camera 111 is arranged at a front left position of the vehicle head and used for collecting images including a front left vehicle head edge and a lane line positioned at the front left position of the vehicle head;
the first front right camera 112 is used for being arranged at a front right position of a vehicle head and collecting images including a front right vehicle head edge and a lane line located at the front right position of the vehicle head.
Optionally, the first left front camera 111 may be disposed at a position 20cm to the left of the head emblem. The first right front camera 112 may be disposed 20cm to the right of the head emblem.
The second camera 12 is configured to be disposed at a rear end of the vehicle, and collects a second image including a rear end edge and a lane line located at the rear end of the vehicle.
Optionally, the second camera 12 is a wide-angle camera.
In order to ensure that there are more car tail edges and lane lines located at the car tail positions in the second image acquired by the second camera 12, in practical applications, the second camera 12 irradiates downwards, preferably vertically downwards.
Specifically, referring to fig. 2, in practical applications, the second camera 12 may include, but is not limited to: a second left rear camera and a second right rear camera (not shown in fig. 2);
the second left rear camera is used for being arranged at the left rear position of the tail of the vehicle and acquiring an image comprising the edge of the left rear tail of the vehicle and a lane line positioned at the left rear position of the tail of the vehicle;
the second right rear camera is used for being arranged at the rear right rear position of the vehicle tail and collecting images including the rear edge of the right rear vehicle tail and a lane line positioned at the rear right rear position of the vehicle tail.
Optionally, the second left rear camera may be positioned 20cm above the left side of the rear license plate. The second rear right camera may be positioned 20cm above the right side of the license plate.
The third camera 13 is used for a set roof, and the acquisition includes: a third image of the front wheels, the rear wheels, the side body outline of the vehicle, and the lane lines on the side of the vehicle body.
The third camera 13 may be a wide-angle camera or a general camera, which is determined according to actual needs, and the present invention is not limited herein.
In practice, a rail may be provided on the roof, see fig. 2, and the third camera 13 is provided on the rail 10 on the roof.
Specifically, referring to fig. 2, the third camera 13 may include: a third left front camera 131, a third right front camera 132, a third left rear camera 133, and a third right rear camera 134;
the third left front camera 131 is configured to be arranged at a left front position of a roof of the vehicle, and collects an image including a left front wheel, a left side body contour of the vehicle, and a lane line at a position where the left front wheel and the left side body contour of the vehicle are located;
the third right front camera 132 is configured to be disposed at a front right position of a roof, and collect an image including a front right wheel, a vehicle right side body contour, and a lane line at a position where the front right wheel and the vehicle right side body contour are located;
the third left rear camera 133 is configured to be disposed at a left rear position of the roof of the vehicle, and collects an image including a left rear wheel, a left body contour of the vehicle, and a lane line at a position where the left rear wheel and the left body contour of the vehicle are located;
the third right rear camera 134 is configured to be disposed at a right rear position of the roof, and collects images including a right rear wheel, a right side body contour of the vehicle, and lane lines located at positions where the right rear wheel and the right side body contour of the vehicle are located.
In practical applications, the third left front camera 131, the third right front camera 132, the third left rear camera 133 and the third right rear camera 134 are all mounted on the cross bar 10 of the roof, as shown in detail in fig. 2.
The control unit 14 is connected to the first camera 11, the second camera 12, and the third camera 13, and is configured to obtain the first image, the second image, and the third image, identify a lane line and a vehicle contour from the first image, the second image, and the third image, determine whether an overlapping area exists between an area where the lane line is located and an area where the vehicle contour is located, determine that a vehicle is pressed if the overlapping area exists, and determine that the vehicle is not pressed if the overlapping area does not exist.
In summary, the present invention discloses a vehicle wire pressing detection system, which comprises: first camera 11, second camera 12, third camera 13 and the control unit 14, first camera 11 sets up in the locomotive position, gathers including the locomotive edge and is located the first image of the lane line in locomotive position, and second camera 12 is used for setting up in rear of a vehicle position, gathers including the rear of a vehicle edge and is located the second image of the lane line in rear of a vehicle position, and third camera 13 is used for the roof that sets up, gathers and includes: the control unit 14 respectively identifies the lane line and the vehicle contour from the first image, the second image and the third image, and determines whether the vehicle is pressed by judging whether a superposition area exists between an area where the lane line is located and an area where the vehicle contour is located. Compared with the traditional scheme that satellite positioning is adopted to determine whether the vehicle presses the line, the method greatly improves the accuracy of judging whether the vehicle is pressed in an examination, and therefore various defects existing in a satellite positioning mode are effectively avoided.
Corresponding to the system embodiment, the invention also discloses a vehicle line pressing detection method.
Referring to fig. 3, a flowchart of a vehicle line pressing detection method disclosed in an embodiment of the present invention is applied to the control unit in the embodiment shown in fig. 1, and the detection method includes the steps of:
s101, acquiring a first image acquired by a first camera, a second image acquired by a second camera and a third image acquired by a third camera;
wherein, the first image includes locomotive edge and is located the lane line of locomotive position, the second image includes: a vehicle rear edge and a lane line at the vehicle rear position, the third image comprising: the front wheels, the rear wheels, the profile of the vehicle side body and the lane lines on the side of the vehicle body.
Step S102, respectively identifying a lane line and a vehicle outline from the first image, the second image and the third image;
it can be understood that when the first camera is a wide-angle camera, the first image collected by the first camera is a distorted image, and before the first image is subjected to lane line and vehicle contour recognition, the first image needs to be corrected, and the distorted image is corrected to a normal image, and the specific correction process can refer to the existing mature scheme and is not repeated here.
Also, the second image and the third image may be distorted images, and therefore, the second image and the third image need to be corrected separately before lane line and vehicle contour recognition is performed from the second image and the third image.
Therefore, step S102 may specifically include:
correcting the first image, the second image and the third image respectively to obtain a first target image, a second target image and a third target image;
separating a lane line region and a vehicle contour region from the first target image, the second target image and the third target image respectively by adopting a deep learning semantic segmentation method;
and respectively carrying out edge calculation on the lane line area and the vehicle contour area, and identifying the lane line and the vehicle contour from a pixel level.
The edge calculation process can refer to the existing mature scheme, and is not described herein again.
The invention further accurately determines the pixels of the lane line and the vehicle contour by performing edge calculation on the lane line area and the vehicle contour area.
In this embodiment, it is necessary to identify the lane line and the vehicle contour of the vehicle head position from the first image, identify the lane line and the vehicle contour of the vehicle position from the second image, and identify the lane line and the vehicle contour of the vehicle side from the third image.
Step S103, judging whether an overlapping area exists between the area where the lane line is located and the area where the vehicle outline is located, if so, executing step S104, and if not, executing step S105;
it can be understood that when the area where the lane line is located and the area where the vehicle contour is located have an overlapping area, the vehicle is pressed, whereas when the area where the lane line is located and the area where the vehicle contour is located do not have an overlapping area, the vehicle is not pressed.
S104, judging vehicle line pressing;
and step S105, judging that the vehicle is not pressed.
In summary, the present invention discloses a vehicle line pressing detection method, wherein a control unit obtains a first image collected by a first camera and including a vehicle head edge and a lane line located at the vehicle head position, a second image collected by a second camera and including a vehicle tail edge and a lane line located at the vehicle tail position, and a third camera comprises: the control unit respectively identifies the lane line and the vehicle contour from the first image, the second image and the third image, and determines whether the vehicle is pressed by judging whether a superposition area exists between an area where the lane line is located and an area where the vehicle contour is located. Compared with the traditional scheme that satellite positioning is adopted to determine whether the vehicle presses the line, the method greatly improves the accuracy of judging whether the vehicle is pressed in an examination, and therefore various defects existing in a satellite positioning mode are effectively avoided.
In the above embodiment, after determining that the vehicle is not pressed, the present invention may further determine the distance between the vehicle contour and the lane line.
Therefore, to further optimize the above embodiment, after step S105, the method may further include:
1) translating the area of the vehicle contour to the outer side of the vehicle contour until the area of the vehicle contour intersects with a lane line;
it is understood that the area where the vehicle contour is located is translated to the outside of the vehicle contour until intersecting the lane line, and the translation distance of the vehicle contour is the distance between the vehicle contour and the lane line.
It should be noted that the vehicle contour and the corresponding lane line described in this embodiment refer to a vehicle contour and a lane line in the same image, for example, when determining the distance between the vehicle contour and the lane line in the first image, the vehicle contour in the first image may be translated to the lane line in the first image; when determining the distance between the vehicle contour in the second image and the lane line, the vehicle contour in the second image may be translated toward the lane line in the second image; when determining the distance between the vehicle contour in the third image and the lane line, the vehicle contour in the third image may be translated toward the lane line in the third image.
2) Recording the number of pixel points of the vehicle contour translation;
3) and based on the number of the pixel points, obtaining the distance between the vehicle outline and the corresponding lane line according to the corresponding relation between the known lane line width and the number of the pixel points corresponding to the lane line width.
In this embodiment, the lane line width is a known quantity, and correspondingly, the number of the pixels corresponding to the lane line width is also a known quantity, so that the straight line length, that is, the distance between the vehicle contour and the corresponding lane line, can be obtained based on the recorded number of the pixels of the straight line according to the correspondence between the lane line width and the number of the pixels corresponding to the lane line width.
It can be understood that in practical application, sundries such as fallen leaves cannot be avoided in the ground environment of vehicle operation, so that the lane line of the acquired image is incomplete.
To solve this problem, the steps in the above embodiment are specifically: after the edge calculation is performed on the lane line region and the vehicle contour region, respectively, and the lane line and the vehicle contour are identified from a pixel level, the detection method further includes:
1) judging whether the identified lane line is complete or not;
when the identified lane line is blocked by debris such as fallen leaves, the outline of the identified lane line may be unclear.
2) If the lane line is incomplete, distinguishing the lane line and a lane line shelter from the pixel level;
3) judging whether the lane line shelter shields one side of the lane line closest to the vehicle outline or not;
whether the vehicle is pressed is determined and the distance between the vehicle outline and the lane line is calculated mainly based on the side, closest to the vehicle outline, of the lane line, so that when the side, closest to the vehicle outline, of the lane line is shielded by a lane line shielding object, the lane line needs to be corrected to be complete; on the contrary, when the side of the lane line, which is shielded by the lane line shielding object and is not the side of the lane line closest to the vehicle contour, whether the lane line is complete or not has no influence on determining whether the vehicle is pressed or not and calculating the distance between the vehicle contour and the lane line, and under the condition, the lane line is not required to be corrected.
4) If the lane line shelter blocks one side of the lane line closest to the vehicle outline, distinguishing the type of the lane line by adopting a deep learning classification method;
among them, the type of lane line, for example, a straight line type lane line, a curved line type lane line, a 90-degree right-angle lane line, and the like.
5) And removing the lane line shielding object from the lane line image, and drawing the missing part of the lane line based on the type of the lane line to obtain the complete lane line.
For example, when the lane line is a straight line type lane line, fallen leaves or impurities are removed from the image, and the straight line connects the disconnected part of the lane line.
And when the lane line is a curve type lane line, deducting fallen leaves or impurities from the image, and connecting the disconnected part of the lane line according to the curvature of the curve.
When the lane line is a right-angle lane line with 90 degrees, deducting fallen leaves or sundries from the image, determining whether the right angle of the inner side of the lane line is complete, and if the right angle of the inner side is complete, connecting the disconnected part of the lane line in a straight line; if the inner right angle is incomplete, it is necessary to extend the inner straight line and make the two straight lines intersect to correct the lane line, wherein the inner side of the lane line means: the side that may coincide with the contour of the vehicle.
In summary, the present invention discloses a vehicle line pressing detection method, wherein a control unit obtains a first image collected by a first camera and including a vehicle head edge and a lane line located at the vehicle head position, a second image collected by a second camera and including a vehicle tail edge and a lane line located at the vehicle tail position, and a third camera comprises: the vehicle body side outline comprises a front wheel, a rear wheel, a vehicle side outline and a third image of a lane line positioned on the side face of the vehicle body, the control unit respectively identifies the lane line and the vehicle outline from the first image, the second image and the third image, when the lane line is determined to be incomplete, the missing part of the lane line is drawn according to the type of the lane line to obtain the complete lane line, whether the vehicle is pressed or not is determined by judging whether a superposition area exists between the area where the lane line is located and the area where the vehicle outline is located, and when the vehicle is determined to be pressed, the vehicle outline can be translated to the lane line to determine the vehicle outline and the distance between the vehicle outline and the corresponding lane line. Compared with the traditional scheme that satellite positioning is adopted to determine whether the vehicle presses the line, the method greatly improves the accuracy of judging whether the vehicle is pressed in an examination, and therefore various defects existing in a satellite positioning mode are effectively avoided.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (8)
1. A vehicle wire pressing detection system, comprising: the system comprises a first camera, a second camera, a third camera and a control unit;
the first camera is arranged at the head position and used for collecting a first image comprising a head edge and a lane line positioned at the head position;
the second camera is arranged at the tail position of the vehicle and used for acquiring a second image comprising the edge of the tail of the vehicle and a lane line positioned at the tail position of the vehicle;
the third camera is used for the roof of setting, gathers and includes: a third image of the front wheels, the rear wheels, the profile of the side body of the vehicle, and the lane lines on the side of the vehicle body;
the control unit is respectively connected with the first camera, the second camera and the third camera, and is used for acquiring the first image, the second image and the third image, respectively identifying a lane line and a vehicle contour from the first image, the second image and the third image, judging whether a superposition area exists between an area where the lane line is located and an area where the vehicle contour is located, if so, judging that the vehicle is pressed, and if not, judging that the vehicle is not pressed.
2. The vehicle wire hold detection system according to claim 1, characterized in that the first camera includes: a first left front camera and a first right front camera;
the first front left camera is arranged at the front left position of the vehicle head and used for collecting images including a front left vehicle head edge and a lane line positioned at the front left position of the vehicle head;
the first front right camera is used for being arranged at the front right position of the vehicle head and collecting images including the front right vehicle head edge and a lane line located at the front right position of the vehicle head.
3. The vehicle wire hold detection system according to claim 1, characterized in that the second camera includes: a second left rear camera and a second right rear camera;
the second left rear camera is used for being arranged at the left rear position of the tail of the vehicle and acquiring an image comprising the edge of the left rear tail of the vehicle and a lane line positioned at the left rear position of the tail of the vehicle;
the second right rear camera is used for being arranged at the rear right rear position of the vehicle tail and collecting images including the rear edge of the right rear vehicle tail and a lane line positioned at the rear right rear position of the vehicle tail.
4. The vehicle wire hold detection system according to claim 1, characterized in that the third camera includes: a third left front camera, a third right front camera, a third left rear camera, and a third right rear camera;
the third left front camera is arranged at the left front position of the roof and used for collecting images including left front wheels, a left side body outline of the vehicle and lane lines at the positions of the left front wheels and the left side body outline of the vehicle;
the third front right camera is used for being arranged at the front right position of the roof and collecting images including a front right wheel, a vehicle right side body outline and a lane line positioned at the positions of the front right wheel and the vehicle right side body outline;
the third left rear camera is arranged at the left rear position of the roof and used for collecting images including a left rear wheel, a vehicle left side body outline and a lane line at the positions of the left rear wheel and the vehicle left side body outline;
the third right rear camera is used for being arranged at the right rear position of the roof and collecting images including a right rear wheel, a vehicle right side body outline and lane lines located at the positions of the right rear wheel and the vehicle right side body outline.
5. A vehicle line pressing detection method applied to a control unit in the vehicle line pressing detection system according to any one of claims 1 to 4, the detection method comprising:
the method comprises the steps of obtaining a first image collected by a first camera, a second image collected by a second camera and a third image collected by a third camera, wherein the first image comprises a vehicle head edge and a lane line located at the vehicle head position, and the second image comprises: a vehicle rear edge and a lane line at the vehicle rear position, the third image comprising: the vehicle comprises a front vehicle wheel, a rear vehicle wheel, a side body outline of the vehicle and a lane line positioned on the side surface of the vehicle body;
identifying lane lines and vehicle contours from the first image, the second image and the third image, respectively;
judging whether a superposition area exists in the area where the lane line is located and the area where the vehicle outline is located;
if yes, judging that the vehicle is pressed;
if not, the vehicle is judged not to be pressed.
6. The vehicle line pressing detection method according to claim 5, wherein the identifying a lane line and a vehicle contour from the first image, the second image, and the third image, respectively, specifically comprises:
correcting the first image, the second image and the third image respectively to obtain a first target image, a second target image and a third target image;
separating a lane line region and a vehicle contour region from the first target image, the second target image and the third target image respectively by adopting a deep learning semantic segmentation method;
and respectively carrying out edge calculation on the lane line area and the vehicle contour area, and identifying the lane line and the vehicle contour from a pixel level.
7. The vehicle line pressing detection method according to claim 5, further comprising, after the determination that the vehicle is not pressed:
translating the area of the vehicle contour to the outer side of the vehicle contour until the area of the vehicle contour intersects with a lane line;
recording the number of pixel points of the vehicle contour translation;
and based on the number of the pixel points, obtaining the distance between the vehicle outline and the lane line according to the corresponding relation between the known lane line width and the number of the pixel points corresponding to the lane line width.
8. The vehicle line pressing detection method according to claim 6, further comprising, after the edge calculation is performed on the lane line region and the vehicle contour region, respectively, and the lane line and the vehicle contour are identified from a pixel level:
judging whether the identified lane line is complete or not;
if the lane line is incomplete, distinguishing the lane line and a lane line shelter from the pixel level;
judging whether the lane line shelter shields one side of the lane line closest to the vehicle outline or not;
if the lane line shelter blocks one side of the lane line closest to the vehicle outline, distinguishing the type of the lane line by adopting a deep learning classification method;
and removing the lane line shielding object from the lane line image, and drawing the missing part of the lane line based on the type of the lane line to obtain the complete lane line.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010161355.4A CN111401186B (en) | 2020-03-10 | 2020-03-10 | Vehicle line pressing detection system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010161355.4A CN111401186B (en) | 2020-03-10 | 2020-03-10 | Vehicle line pressing detection system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111401186A true CN111401186A (en) | 2020-07-10 |
| CN111401186B CN111401186B (en) | 2024-05-28 |
Family
ID=71430792
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010161355.4A Active CN111401186B (en) | 2020-03-10 | 2020-03-10 | Vehicle line pressing detection system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111401186B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113538377A (en) * | 2021-07-15 | 2021-10-22 | 河北三国新能源科技有限公司 | Driving test vehicle quarter turn line pressing detection method and system based on panoramic looking-around |
| WO2024017003A1 (en) * | 2022-07-21 | 2024-01-25 | 天津所托瑞安汽车科技有限公司 | Vehicle merging detection method and apparatus based on combined algorithms, and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109360471A (en) * | 2018-11-07 | 2019-02-19 | 无锡合壮智慧交通有限公司 | A kind of driver's real vehicle examination deduction of points evidence collecting method and device |
| CN109657632A (en) * | 2018-12-25 | 2019-04-19 | 重庆邮电大学 | A kind of lane detection recognition methods |
| CN109949578A (en) * | 2018-12-31 | 2019-06-28 | 上海眼控科技股份有限公司 | A kind of illegal automatic auditing method of vehicle crimping based on deep learning |
| CN110210363A (en) * | 2019-05-27 | 2019-09-06 | 中国科学技术大学 | A kind of target vehicle crimping detection method based on vehicle-mounted image |
| WO2019200938A1 (en) * | 2018-04-18 | 2019-10-24 | 福州大学 | Early warning system for vehicles rolling on line |
-
2020
- 2020-03-10 CN CN202010161355.4A patent/CN111401186B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019200938A1 (en) * | 2018-04-18 | 2019-10-24 | 福州大学 | Early warning system for vehicles rolling on line |
| CN109360471A (en) * | 2018-11-07 | 2019-02-19 | 无锡合壮智慧交通有限公司 | A kind of driver's real vehicle examination deduction of points evidence collecting method and device |
| CN109657632A (en) * | 2018-12-25 | 2019-04-19 | 重庆邮电大学 | A kind of lane detection recognition methods |
| CN109949578A (en) * | 2018-12-31 | 2019-06-28 | 上海眼控科技股份有限公司 | A kind of illegal automatic auditing method of vehicle crimping based on deep learning |
| CN110210363A (en) * | 2019-05-27 | 2019-09-06 | 中国科学技术大学 | A kind of target vehicle crimping detection method based on vehicle-mounted image |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113538377A (en) * | 2021-07-15 | 2021-10-22 | 河北三国新能源科技有限公司 | Driving test vehicle quarter turn line pressing detection method and system based on panoramic looking-around |
| CN113538377B (en) * | 2021-07-15 | 2022-08-12 | 河北三国新能源科技有限公司 | Driving test vehicle quarter turn line pressing detection method and system based on panoramic looking-around |
| WO2024017003A1 (en) * | 2022-07-21 | 2024-01-25 | 天津所托瑞安汽车科技有限公司 | Vehicle merging detection method and apparatus based on combined algorithms, and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111401186B (en) | 2024-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108537197B (en) | Lane line detection early warning device and method based on deep learning | |
| CN108446678B (en) | Dangerous driving behavior identification method based on skeletal features | |
| DE102013008451B4 (en) | Vehicle parking control system and vehicle parking control method using the same | |
| CN110298300B (en) | Method for detecting vehicle illegal line pressing | |
| CN102663352B (en) | Track identification method | |
| CN111401186B (en) | Vehicle line pressing detection system and method | |
| CN107478590A (en) | A kind of method of combination motor vehicle intelligent vision identification and remote exhaust emission detection | |
| CN112487908B (en) | Front vehicle line pressing behavior detection and dynamic tracking method based on vehicle-mounted video | |
| EP1192597A1 (en) | Method of detecting objects within a wide range of a road vehicle | |
| DE102006012914A1 (en) | System and method for determining the distance to a preceding vehicle | |
| CN104700072A (en) | Lane line historical frame recognition method | |
| CN109035868B (en) | Method for lane division by automatically driving vehicle under condition of no lane line | |
| CN106485694A (en) | A kind of high ferro contact net double-jacket tube connector six-sided nut based on cascade classifier comes off defective mode detection method | |
| DE102014221803A1 (en) | Method and device for determining a current driving situation | |
| CN105787438A (en) | Video-based locomotive driver driving state detection method and system | |
| CN110376211B (en) | Wet-process-gummed synthetic leather hemming on-line detection device and method | |
| CN108830822B (en) | Pantograph carbon skateboard abrasion identification method based on improved Canny operator | |
| CN116631187B (en) | Intelligent acquisition and analysis system for case on-site investigation information | |
| CN112285111A (en) | Pantograph front carbon sliding plate defect detection method, device, system and medium | |
| CN115131584A (en) | Visual recognition system for vehicle cleaning robot | |
| CN114266770B (en) | Method for detecting hanger defect of high-speed rail contact net through neural network learning method | |
| DE102005044981A1 (en) | Lane markings identifying method for moving motor vehicle, involves combining coherent images having similar brightness to cluster for determining form of objects, and identifying lane markings based on form and periodic arrangement in lane | |
| EP4222039A1 (en) | Optical railway detection | |
| CN114724091A (en) | Method and device for identifying foreign matters on transmission line wire | |
| CN203305895U (en) | System capable of giving alarm when vehicle deviates from lane based on image recognition |
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 |