CN109271857A - A kind of puppet lane line elimination method and device - Google Patents

A kind of puppet lane line elimination method and device Download PDF

Info

Publication number
CN109271857A
CN109271857A CN201810914211.4A CN201810914211A CN109271857A CN 109271857 A CN109271857 A CN 109271857A CN 201810914211 A CN201810914211 A CN 201810914211A CN 109271857 A CN109271857 A CN 109271857A
Authority
CN
China
Prior art keywords
lane line
road
pseudo
vehicle
information
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
Application number
CN201810914211.4A
Other languages
Chinese (zh)
Inventor
蒋少峰
冯锴
肖志光
陈昊
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangzhou Xiaopeng Motors Technology Co Ltd
Original Assignee
Guangzhou Xiaopeng Motors Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangzhou Xiaopeng Motors Technology Co Ltd filed Critical Guangzhou Xiaopeng Motors Technology Co Ltd
Priority to CN201810914211.4A priority Critical patent/CN109271857A/en
Publication of CN109271857A publication Critical patent/CN109271857A/en
Priority to PCT/CN2019/093423 priority patent/WO2020029706A1/en
Pending legal-status Critical Current

Links

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
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
    • 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
    • G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Mathematics (AREA)
  • Multimedia (AREA)
  • Operations Research (AREA)
  • Algebra (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Traffic Control Systems (AREA)

Abstract

The present invention relates to Vehicular intelligents to assist driving technology field, disclose a kind of pseudo- lane line elimination method and device, comprising: obtain the road image of vehicle place road, and identify lane line from road image；The obstacle information of the stationary obstruction of road is obtained, and generates the boundary information of road by being fitted to obstacle information；The lane line being located at except the boundary of road in lane line is determined as pseudo- lane line, pseudo- lane line is rejected from lane line.Implement the embodiment of the present invention, lane line can be obtained by the road image of road where vehicle, and the boundary information of road is determined by the obstacle information around road, if barrier similar with lane line color is identified as lane line, by detecting whether lane line is in boundary, the pseudo- lane line being not in boundary can be rejected, ensure that the lane line identified is true lane line, to promote the accuracy of Lane detection.

Description

A kind of puppet lane line elimination method and device
Technical field
The present invention relates to Vehicular intelligents to assist driving technology field, and in particular to a kind of puppet lane line elimination method and dress It sets.
Background technique
It is to realize the driving of Vehicular intelligent auxiliary and unpiloted basis to the perception of ambient enviroment in vehicle travel process, Lane detection technology is to realize the important link of Vehicular intelligent path planning and Decision Control, and realize that lane keeps auxiliary The auxiliary such as (Lane Keeping Assist, LKA) and lane departure warning (Lane Departure Warning, LDW) drive Basis.
Existing Lane detection technology is usually: the present road scene photo of road where obtaining vehicle utilizes figure As identification technology identifies lane line from present road scene photo.However, being found in practice process, due to real road The influence of middle light environment, it may appear that the chaff interferents such as roadside guardrail similar in color in present road scene photo are identified as vehicle It is the case where diatom, higher so as to cause the error rate of Lane detection, the accuracy of Vehicular intelligent Driving Decision-making is influenced, is caused tight The security risk of weight.
Summary of the invention
The embodiment of the present invention discloses a kind of pseudo- lane line elimination method and device, is able to ascend the correct of Lane detection Rate.
First aspect of the embodiment of the present invention discloses a kind of pseudo- lane line elimination method, which comprises
The road image of road where obtaining vehicle, and lane line is identified from the road image；
The obstacle information of the stationary obstruction of the road is obtained, and by being fitted life to the obstacle information At the boundary information of the road, the boundary information includes at least the boundary of the road；
The lane line being located at except the boundary of the road in the lane line is determined as pseudo- lane line, from the lane The pseudo- lane line is rejected in line.
As an alternative embodiment, in first aspect of the embodiment of the present invention, road where the acquisition vehicle Road image, and identify lane line from the road image, comprising:
The road image of road where obtaining the vehicle using the visual sensor being arranged on vehicle；
Semantic feature identification is carried out to the road image, extracts the semantic feature in the road image；
Judge to whether there is and the matched semantic feature of lane line in the semantic feature；
If it does, identifying the lane line in the road image by image recognition technology.
As an alternative embodiment, in first aspect of the embodiment of the present invention, it is described to obtain the quiet of the road The only obstacle information of barrier, and the boundary information of the road is generated by being fitted to the obstacle information, packet It includes:
The obstacle information of stationary obstruction, the barrier letter are obtained using the radar sensor being arranged on the vehicle Breath includes the orientation of estimating traveling curve of the stationary obstruction relative to the vehicle；
By analyzing the orientation in the obstacle information, the obstacle information is divided into left side obstacle information With right side obstacle information；
The left side obstacle information and the right side obstacle information are fitted respectively using least square method, it is raw At and the corresponding road left margin curvilinear equation of the left side obstacle information and corresponding with the right side obstacle information Road right margin curvilinear equation；
It is calculated according to the road left margin curvilinear equation and the road right margin curvilinear equation and generates the road Boundary information.
As an alternative embodiment, described utilize sets on the vehicle in first aspect of the embodiment of the present invention Before the radar sensor set obtains the obstacle information of stationary obstruction, the method also includes:
The running condition information of the vehicle is obtained, the running condition information includes at least the speed of the vehicle, turns The corner and yaw velocity of wheel；
It estimates the driving trace of the vehicle according to the running condition information, and the driving trace is determined as described Vehicle estimates traveling curve.
As an alternative embodiment, in first aspect of the embodiment of the present invention, road where the acquisition vehicle Road image, and from the road image identify lane line after and it is described will in the lane line be located at the road Lane line except the boundary on road is determined as before pseudo- lane line, the method also includes:
The lane line in the road image is fitted, lane line curve corresponding with the lane line is generated Equation, wherein the corresponding lane line curvilinear equation of a lane line；
It is described that the lane line being located at except the boundary of the road in the lane line is determined as pseudo- lane line, from described The pseudo- lane line is rejected in lane line, comprising:
Using the vehicle as foundation, in conjunction with the boundary information of the road and the corresponding lane line of the lane line Curvilinear equation generates comprehensive road information；
From the pseudo- lane line curvilinear equation determined in the comprehensive road information except the boundary in the road, and from The corresponding pseudo- lane line of the pseudo- lane line curvilinear equation is rejected in the lane line.
As an alternative embodiment, in first aspect of the embodiment of the present invention, it is described to believe from the comprehensive road The pseudo- lane line curvilinear equation except the boundary in the road is determined in breath, and the pseudo- vehicle is rejected from the lane line After the corresponding pseudo- lane line of diatom curvilinear equation, the method also includes:
The lane line curvilinear equation in the comprehensive road information in addition to the pseudo- lane line curvilinear equation is true It is set to true lane line curvilinear equation；
The display output being arranged by the vehicle shows the corresponding true lane of the true lane line curvilinear equation Line.
Second aspect of the embodiment of the present invention discloses a kind of pseudo- lane line device for eliminating, comprising:
Recognition unit for the road image of road where obtaining vehicle, and identifies lane line from the road image；
First fitting unit, the obstacle information of the stationary obstruction for obtaining the road, and by the barrier Object information is hindered to be fitted the boundary information for generating the road, the boundary information includes at least the boundary of the road；
Culling unit, for the lane line being located at except the boundary of the road in the lane line to be determined as pseudo- lane Line rejects the pseudo- lane line from the lane line.
As an alternative embodiment, in second aspect of the embodiment of the present invention, the recognition unit includes:
First obtains subelement, the road for road where obtaining the vehicle using the visual sensor being arranged on vehicle Road image；
Subelement is extracted, for carrying out semantic feature identification to the road image, extracts the language in the road image Adopted feature；
Judgment sub-unit whether there is and the matched semantic feature of lane line in the semantic feature for judging；
It identifies subelement, when the result for judging in the judgment sub-unit is is, is identified by image recognition technology Lane line in the road image.
As an alternative embodiment, in second aspect of the embodiment of the present invention, first fitting unit includes:
Second obtains subelement, for obtaining the obstacle of stationary obstruction using the radar sensor being arranged on the vehicle Object information, the obstacle information include the orientation of estimating traveling curve of the stationary obstruction relative to the vehicle；
Subelement is analyzed, for by analyzing the orientation in the obstacle information, the obstacle information to be divided For left side obstacle information and right side obstacle information；
It is fitted subelement, for utilizing least square method respectively to the left side obstacle information and the right side barrier Information is fitted, and is generated road left margin curvilinear equation corresponding with the left side obstacle information and is hindered with the right side Hinder the corresponding road right margin curvilinear equation of object information；
Computation subunit, for being calculated according to the road left margin curvilinear equation and the road right margin curvilinear equation Generate the boundary information of the road.
As an alternative embodiment, in second aspect of the embodiment of the present invention, the puppet lane line device for eliminating Further include:
Acquiring unit, for quiet using the radar sensor acquisition being arranged on the vehicle in the second acquisition subelement Only before the obstacle information of barrier, the running condition information of the vehicle is obtained, the running condition information includes at least The speed of the vehicle, the corner of runner and yaw velocity；
Unit is estimated, for estimating the driving trace of the vehicle according to the running condition information, and by the traveling What track was determined as the vehicle estimates traveling curve, and triggering described second obtains subelement and executes described in the utilization The radar sensor being arranged on vehicle obtains the obstacle information of stationary obstruction.
As an alternative embodiment, in second aspect of the embodiment of the present invention, the puppet lane line device for eliminating Further include:
Second fitting unit, for the road image of road where obtaining vehicle in the recognition unit, and from the road In the image of road identify lane line after, and the culling unit will in the lane line positioned at the road boundary except Lane line be determined as before pseudo- lane line, the lane line in the road image is fitted, generate with the vehicle The corresponding lane line curvilinear equation of diatom, wherein the corresponding lane line curvilinear equation of a lane line；
The culling unit includes:
Subelement is generated, is used for using the vehicle as foundation, in conjunction with the boundary information and the lane line of the road The corresponding lane line curvilinear equation generates comprehensive road information；
Subelement is rejected, for being in the pseudo- lane except the boundary of the road from determination in the comprehensive road information Line curvilinear equation, and the corresponding pseudo- lane line of the pseudo- lane line curvilinear equation is rejected from the lane line.
As an alternative embodiment, in second aspect of the embodiment of the present invention, the puppet lane line device for eliminating Further include:
Determination unit, for determining the side for being in the road from the comprehensive road information in the rejecting subelement Pseudo- lane line curvilinear equation except boundary, and the corresponding pseudo- lane of the pseudo- lane line curvilinear equation is rejected from the lane line After line, the lane line curvilinear equation in the comprehensive road information in addition to the pseudo- lane line curvilinear equation is determined For true lane line curvilinear equation；
Output unit, the display output for being arranged by the vehicle show the true lane line curvilinear equation pair The true lane line answered.
The third aspect of the embodiment of the present invention discloses another pseudo- lane line device for eliminating, comprising:
It is stored with the memory of executable program code；
The processor coupled with the memory；
The processor calls the executable program code stored in the memory, executes any of first aspect A kind of some or all of method step.
Fourth aspect of the embodiment of the present invention discloses a kind of computer readable storage medium, the computer readable storage medium Store program code, wherein said program code includes the part or complete for executing any one method of first aspect The instruction of portion's step.
The 5th aspect of the embodiment of the present invention discloses a kind of computer program product, when the computer program product is calculating When being run on machine, so that the computer executes some or all of any one method of first aspect step.
The aspect of the embodiment of the present invention the 6th disclose a kind of using distribution platform, and the application distribution platform is for publication calculating Machine program product, wherein when the computer program product is run on computers, so that the computer executes first party Some or all of any one method in face step.
Compared with prior art, the embodiment of the present invention has the advantages that
In the embodiment of the present invention, the road image of road where obtaining vehicle, and lane line is identified from road image；It obtains By way of the obstacle information of the stationary obstruction on road, and the boundary information by being fitted generation road to obstacle information； The lane line being located at except the boundary of road in lane line is determined as pseudo- lane line, pseudo- lane line is rejected from lane line.It can See, implement the embodiment of the present invention, lane line can be obtained by the road image of road where vehicle, and by around road Obstacle information determine the boundary information of road, if the barriers such as guardrail similar with lane line color are identified as lane When line, by detecting whether lane line is in boundary, the pseudo- lane line being not in boundary can be rejected, ensure that identification Lane line out is true lane line, to promote the accuracy of Lane detection.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to needed in the embodiment Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for ability For the those of ordinary skill of domain, without creative efforts, it can also be obtained according to these attached drawings other attached Figure.
Fig. 1 is a kind of flow diagram of pseudo- lane line elimination method disclosed by the embodiments of the present invention；
Fig. 2 is the flow diagram of another pseudo- lane line elimination method disclosed by the embodiments of the present invention；
Fig. 3 is the flow diagram of another pseudo- lane line elimination method disclosed by the embodiments of the present invention；
Fig. 4 is a kind of schematic diagram of comprehensive road information disclosed by the embodiments of the present invention；
Fig. 5 is a kind of structural schematic diagram of pseudo- lane line device for eliminating disclosed by the embodiments of the present invention；
Fig. 6 is the structural schematic diagram of another pseudo- lane line device for eliminating disclosed by the embodiments of the present invention；
Fig. 7 is the structural schematic diagram of another pseudo- lane line device for eliminating disclosed by the embodiments of the present invention；
Fig. 8 is the structural schematic diagram of another pseudo- lane line device for eliminating disclosed by the embodiments of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete Site preparation description, it is clear that the described embodiment is only a part of the embodiment of the present invention, instead of all the embodiments.Based on this Embodiment in invention, every other reality obtained by those of ordinary skill in the art without making creative efforts Example is applied, shall fall within the protection scope of the present invention.
It should be noted that term " includes " and " having " and their any changes in the embodiment of the present invention and attached drawing Shape, it is intended that cover and non-exclusive include.Such as contain the process, method of a series of steps or units, system, product or Equipment is not limited to listed step or unit, but optionally further comprising the step of not listing or unit or optional Ground further includes the other step or units intrinsic for these process, methods, product or equipment.
The embodiment of the present invention discloses a kind of pseudo- lane line elimination method and device, is able to ascend the correct of Lane detection Rate.It is described in detail separately below.
Embodiment one
Referring to Fig. 1, Fig. 1 is a kind of flow diagram of pseudo- lane line elimination method disclosed by the embodiments of the present invention.Such as Shown in Fig. 1, which be may comprise steps of:
101, the road image of road where pseudo- lane line device for eliminating obtains vehicle, and lane is identified from road image Line.
In the embodiment of the present invention, pseudo- lane line device for eliminating can be the part dress in the terminal device being arranged on vehicle It sets.Road image can be obtained by camera or visual sensor etc..Road image is the road on vehicle current driving direction Road image can pass through image recognition technology (Image Recognition Technology), image segmentation (Image Segmentation) technology etc. automatically identifies lane line in road image, and to identifying in the embodiment of the present invention The quantity of lane line is without limitation.
102, pseudo- lane line device for eliminating obtains the obstacle information of the stationary obstruction of road, and by believing barrier Breath is fitted the boundary information for generating road, and boundary information includes at least the boundary of road.
In the embodiment of the present invention, obstacle information only needs to obtain the information of stationary obstruction, because in addition to this vehicle Vehicle it is also assumed that be barrier, pedestrian is it is also assumed that be barrier, however the position of the vehicle and pedestrian in moving And it is not fixed, and the position of the vehicle and pedestrian in movement can not indicate the boundary of road, therefore obstacle information only needs Obtain the obstacle information of stationary obstruction.Stationary obstruction can be the barrier positioned at both sides of the road, as roadside guardrail, Greenbelt, direction board, street lamp or building etc., in this regard, the embodiment of the present invention is without limitation.Since vehicle is traveling in road Intermediate, therefore the obstacle information of the stationary obstruction obtained can be distributed across both sides of the road, according to both sides of the road Obstacle information can be fitted generate road boundary information, the boundary information can according to both sides of the road obstacle information Obtained road left margin and road right margin, may be considered road between road left margin and road right margin.
As an alternative embodiment, can also be performed before pseudo- lane line device for eliminating execution step 102 following Step:
Pseudo- lane line device for eliminating obtains the barrier of road；
Pseudo- lane line device for eliminating detects the speed of barrier by velocity radar；
The barrier that speed in barrier is zero is determined as stationary obstruction by pseudo- lane line device for eliminating.
Wherein, implement this embodiment, the vehicle can be moved about to avoid pseudo- lane line device for eliminating other Vehicle and/or pedestrian etc. are identified as barrier, reduce the error that fitting generates the boundary information of road.
103, the lane line being located at except the boundary of road in lane line is determined as pseudo- lane by pseudo- lane line device for eliminating Line rejects pseudo- lane line from lane line.
In the embodiment of the present invention, if identifying that some lane line is located at except the boundary of road, it is believed that the lane Line and there is no on road, it may be possible to barriers such as roadside guardrail, therefore the lane line can be determined as to pseudo- lane line, and The puppet lane line is rejected from lane line.
As an alternative embodiment, pseudo- lane line device for eliminating will be located at except the boundary of road in lane line Lane line is determined as pseudo- lane line, and the mode that pseudo- lane line is rejected from lane line may comprise steps of:
Pseudo- lane line device for eliminating combines the lane line identified to generate lane line set；
Pseudo- lane line device for eliminating successively traverses the lane line in the lane line set, will be located at except the boundary of road Lane line is labeled as pseudo- lane line, and the lane line within the boundary for being located at road is labeled as true lane line；
Pseudo- lane line device for eliminating rejects in lane line set and is marked as the lane line of pseudo- lane line, to make lane line Lane line in set is true lane line.
Wherein, implement this embodiment, can determine that each lane line was traversed, and lane line is carried out true Puppet label so that pseudo- lane line device for eliminating can batch deletion puppet lane line, improve the fortune of pseudo- lane line device for eliminating Scanning frequency degree.
In the method depicted in fig. 1, lane line can be obtained by the road image of road where vehicle, and passed through Obstacle information around road determines the boundary information of road, and the pseudo- lane line being not in road boundary can be picked It removes, ensure that the lane line identified is true lane line, to promote the accuracy of Lane detection.In addition, implementing Method described in Fig. 1, can be to avoid other vehicles and/or pedestrian etc. that pseudo- lane line device for eliminating moves about the vehicle It is identified as barrier, reduces the error that fitting generates the boundary information of road.In addition, implement method described in Fig. 1, it can be with It determines that each lane line was traversed, and true or false mark is carried out to lane line, so that pseudo- lane line device for eliminating can be criticized The deletion puppet lane line of amount, improves the speed of service of pseudo- lane line device for eliminating.
Embodiment two
Referring to Fig. 2, Fig. 2 is the flow diagram of another pseudo- lane line elimination method disclosed by the embodiments of the present invention. As shown in Fig. 2, the puppet lane line elimination method may comprise steps of:
201, the road of road where pseudo- lane line device for eliminating obtains vehicle using the visual sensor being arranged on vehicle Image.
In the embodiment of the present invention, the visual sensor being arranged on vehicle can be monocular vision sensor, be regarded by monocular The road image for feeling that sensor obtains can be identified as 3-D image, to make pseudo- lane line device for eliminating fast and accurately Obtain the information in road image.Visual sensor can be set in the front of vehicle, to obtain the road image of vehicle front； Visual sensor is also possible to rotatable, and rotatable visual sensor can be set in any position of vehicle, pseudo- lane Line device for eliminating can control the shooting direction of the rotatable visual sensor, so that rotatable visual sensor obtains always The road image that type B error lane line device for eliminating needs improves mileage chart to improve the controllability of visual sensor As the diversity of content.
202, pseudo- lane line device for eliminating carries out semantic feature identification to road image, and the semanteme extracted in road image is special Sign.
In the embodiment of the present invention, semantic feature can be the meaning of object present in image, such as the language in road image Adopted feature may include the semantic features such as vehicle, road surface, lane line, tree, house or direction board.Pseudo- lane line device for eliminating can To identify the semantic feature in road image, and identical semantic feature is merged, with simplify judge be in semantic feature It is no to there is the step of matching with lane line semantic feature, it, only need to be to not without judging each semantic feature identified Same semantic feature is judged, the pseudo- lane line device for eliminating speed of service is improved.
203, pseudo- lane line device for eliminating judge to whether there is in semantic feature with the matched semantic feature of lane line, if It is to execute step 204~step 211；If not, terminating this process.
In the embodiment of the present invention, if identifying presence and the matched semantic feature of lane line in road image, it can recognize To need to carry out lane detection to present road, to judge whether lane line present in road is pseudo- lane line, if worked as There is pseudo- lane line in preceding road, it is necessary to pseudo- lane line be rejected, to guarantee the accuracy of Lane detection, further protected Demonstrate,prove the safety driven.
204, pseudo- lane line device for eliminating identifies the lane line in road image by image recognition technology.
As an alternative embodiment, pseudo- lane line device for eliminating is identified in road image by image recognition technology The mode of lane line may comprise steps of:
The Road image segmentation is several subgraphs by image Segmentation Technology by pseudo- lane line device for eliminating；
Pseudo- lane line device for eliminating obtains the image of several subgraphs using deep learning (Deep Learning) algorithm Semantic feature；
Pseudo- lane line device for eliminating determining and matched target image language of lane line semantic feature from image, semantic feature Adopted feature；
Pseudo- lane line device for eliminating is by the corresponding lane line of image recognition technology recognition target image semantic feature, and one Target image semantic feature corresponds to a lane line.
Wherein, implement this embodiment, can use image Segmentation Technology, deep learning algorithm and image recognition skill Art identifies lane line present in road image jointly, so that lane line present in road image can be known as far as possible It does not come out, improves the accuracy rate of Lane detection.
In the embodiment of the present invention, implement above-mentioned step 201~step 204, can confirmation road image in there are vehicles The lane line in identification road image is removed after diatom again, to simplify the lane line for being carried out identification to each road image Operation, improve to road image processing speed.
205, pseudo- lane line device for eliminating obtains the running condition information of vehicle, and running condition information includes at least vehicle The corner and yaw velocity of speed, runner.
In the embodiment of the present invention, the running condition information of vehicle can pass through Inertial Measurement Unit (Inertial Measurement Unit, IMU) it obtains, IMU may include the dress such as velocity sensor, accelerometer, gyroscope and wheel speed meter It sets, for obtaining the information such as the travel speed of vehicle, the acceleration of vehicle, the corner of runner and yaw velocity.
206, information estimates the driving trace of vehicle to pseudo- lane line device for eliminating according to transport condition, and driving trace is true Be set to vehicle estimates traveling curve.
In the embodiment of the present invention, vehicle can be determined as working as vehicle in the tangent line for estimating the current location on traveling curve Preceding driving direction, the corresponding current driving direction of vehicle can estimate traveling curvilinear motion with vehicle.
In the embodiment of the present invention, implement above-mentioned step 205~step 206, current driving status is estimated according to vehicle The driving trace of vehicle out, and then be calculated vehicle estimates traveling curve, so that pseudo- lane line device for eliminating can use This is estimated traveling curve and accurately identifies pseudo- lane line.
In the embodiment of the present invention, step 205~step 206 can be before the arbitrary steps in step 201~step 204 Or execute later, to will not the embodiment of the present invention have an impact.
207, pseudo- lane line device for eliminating obtains the barrier of stationary obstruction using the radar sensor being arranged on vehicle Information, obstacle information include the orientation of estimating traveling curve of the stationary obstruction relative to vehicle.
In the embodiment of the present invention, the radar sensor being arranged on vehicle can be millimetre-wave radar sensor (Millimeter Radar Sensor), and the embodiment of the present invention is for the number for the millimetre-wave radar sensor being arranged on vehicle Amount without limitation, the information of the stationary obstruction around the road of vehicle place can be accurately obtained by radar sensor, such as Linear distance of the available stationary obstruction away from vehicle can also obtain the current driving direction of stationary obstruction Yu the vehicle Angle, which can be calculated by the above-mentioned linear distance got and angle and estimate row relative to vehicle Sail the specific orientation of curve.
Specifically, the millimeter wave that millimetre-wave radar sensor uses may be in 30~300GHz frequency domain, millimeter Wave radar sensor has many advantages, such as that small in size, easy of integration and spatial resolution is high, further, since millimetre-wave radar sensor The ability of penetrating fog, cigarette and dust is strong, therefore millimetre-wave radar sensor also has the characteristics that strong antijamming capability.
Further, millimetre-wave radar sensor can built-in several transmitting antennas and several receiving antennas, milli Metre wave radar sensor can emit radar wave by transmitting antenna, form radar after radar wave is reflected by barrier and return Wave, receiving antenna can receive radar return, wherein the speed of radar wave and radar return is equal to the light velocity；Millimetre-wave radar Sensor can be calculated according to the time difference that the light velocity and transmitting antenna transmitting radar wave and receiving antenna receive radar return Distance to barrier away from vehicle.Millimetre-wave radar sensor can also calculate the side in the current driving direction of barrier and vehicle Parallactic angle can be calculated when two receiving antennas any in millimetre-wave radar sensor receive identical radar return Any two receiving antennas respectively away from the distance of barrier, according between any two receiving antennas geometric distance and The barrier and vehicle can be calculated respectively away from the distance of barrier, using trigonometric function in any two receiving antennas The azimuth in current driving direction, and then can determine that barrier is located at the left side in the current driving direction of vehicle according to azimuth Or right side.In addition, millimetre-wave radar sensor can also judge the motion state of barrier, millimetre-wave radar sensor Radar wave can be emitted by transmitting antenna with fixed frequency, and the frequency for the radar return that receiving antenna receives is carried out Analysis is calculated, movement velocity of the barrier relative to vehicle is obtained, if the driving direction of the direction of motion of barrier and vehicle Opposite and the speed of barrier and the speed of vehicle are identical, it is believed that the barrier remains static, and is stationary obstruction； If the direction of motion of barrier not from the driving direction of vehicle is opposite or the speed of barrier and the speed of vehicle it is different, can It is moving obstacle to think that the barrier is kept in motion.As it can be seen that can accurately really using millimetre-wave radar sensor The orientation for determining stationary obstruction and stationary obstruction improves the efficiency that pseudo- lane line device for eliminating determines obstacle information.
208, obstacle information is divided into left side barrier by the orientation in disturbance of analysis object information by pseudo- lane line device for eliminating Hinder object information and right side obstacle information.
In the embodiment of the present invention, since vehicle travels between road left margin and road right margin, and road is left Boundary and road right margin usually can all have the stationary obstructions such as guardrail, greenbelt or building, and therefore, it is necessary to acquisition To stationary obstruction classify relative to the orientation in the current driving direction of vehicle, stationary obstruction information can be divided into Two classes, one kind are the left side obstacle informations of road left margin, and another kind of is the right side obstacle information of road right margin.It will be quiet Only obstacle information is divided into left side obstacle information and right side obstacle information, the road left margin that subsequent fitting can be made to generate Curvilinear equation and road right margin curvilinear equation are more accurate.
209, pseudo- lane line device for eliminating respectively believes left side obstacle information and right side barrier using least square method Breath is fitted, generate corresponding with left side obstacle information road left margin curvilinear equation and with right side obstacle information pair The road right margin curvilinear equation answered.
In the embodiment of the present invention, least square three can be carried out to left side obstacle information and right side obstacle information respectively Rank multinomial fitting, the road left margin curvilinear equation and road right margin curvilinear equation of generation are ok are as follows:
Y=C3*x3+C2*x2+C1*x+C0
Wherein, y is abscissa, and x is ordinate, C0、C1、C2And C3It is the polynomial coefficient, and passes through minimum Two, which multiply three rank multinomials, is fitted to obtain, in addition, sharing road left margin curvilinear equation and the shared C of road right margin curvilinear equation1、 C2And C3These three coefficients.
210, pseudo- lane line device for eliminating calculates life according to road left margin curvilinear equation and road right margin curvilinear equation At the boundary information of road.
In the embodiment of the present invention, implement above-mentioned step 207~step 210, road left margin curvilinear equation can be generated With road right margin curvilinear equation, and then the boundary information of road is calculated, so that the boundary for the road being calculated Information is more accurate.
211, the lane line being located at except the boundary of road in lane line is determined as pseudo- lane by pseudo- lane line device for eliminating Line rejects pseudo- lane line from lane line.
In the method depicted in fig. 2, lane line can be obtained by the road image of road where vehicle, and passed through Obstacle information around road determines the boundary information of road, and the pseudo- lane line being not in road boundary can be picked It removes, ensure that the lane line identified is true lane line, to promote the accuracy of Lane detection.In addition, implementing The controllability of visual sensor can be improved in method described in Fig. 2, improves the diversity of road image content.In addition, Implement method described in Fig. 2, can make lane line present in road image that can identify as far as possible, improve The accuracy rate of Lane detection.In addition, implementing method described in Fig. 2, it can simplify and identification is carried out to each road image Lane line operation, improve to road image processing speed.In addition, implementing method described in Fig. 2, this can use It estimates traveling curve and accurately identifies pseudo- lane line.In addition, implementing method described in Fig. 2, road left margin can be generated Curvilinear equation and road right margin curvilinear equation, and then the boundary information of road is calculated, so that the road being calculated The boundary information on road is more accurate.
Embodiment three
Referring to Fig. 3, Fig. 3 is the flow diagram of another pseudo- lane line elimination method disclosed by the embodiments of the present invention. As shown in figure 3, the puppet lane line elimination method may comprise steps of:
301, the road image of road where pseudo- lane line device for eliminating obtains vehicle, and lane is identified from road image Line.
302, pseudo- lane line device for eliminating is fitted the lane line in road image, generates vehicle corresponding with lane line Diatom curvilinear equation, wherein the corresponding lane line curvilinear equation of a lane line.
In the embodiment of the present invention, the corresponding lane line curvilinear equation of lane line is also quasi- by three rank multinomial of least square Symphysis at, the formula that lane line curvilinear equation can be used with road left margin curvilinear equation and road right margin curvilinear equation Equally, further, since lane line, road left margin and road right margin are all to be obtained with the vehicle on present road for foundation It takes, therefore, lane line curvilinear equation can share C with road left margin curvilinear equation and road right margin curvilinear equation1、C2 And C3These three coefficients.
In the embodiment of the present invention, implement above-mentioned step 302, can be generated according to each lane line computation identified each The corresponding lane line curvilinear equation of a lane line, so that the subsequent pseudo- lane line identified from lane line is more accurate.
303, pseudo- lane line device for eliminating obtains the obstacle information of the stationary obstruction of road, and by believing barrier Breath is fitted the boundary information for generating road, and boundary information includes at least the boundary of road.
304, pseudo- lane line device for eliminating is corresponding in conjunction with the boundary information of road and each lane line using vehicle as foundation Lane line curvilinear equation generate comprehensive road information.
It (can be road left margin curvilinear equation and the road right side by the boundary information of road in the embodiment of the present invention Boundary curve equation) and the corresponding lane line curvilinear equation of each lane line be fused in a unified analytical standard, should Analytical standard can be using the vehicle as foundation, because of the boundary information of road and the corresponding lane line curve side of each lane line Journey is all based on what the vehicle obtained, which can be integrated into comprehensive road information.
305, pseudo- lane line device for eliminating determines the pseudo- lane line except the boundary in road from comprehensive road information Curvilinear equation, and the corresponding pseudo- lane line of pseudo- lane line curvilinear equation is rejected from lane line.
In the embodiment of the present invention, there can be the pseudo- vehicle except one or more boundaries in road in comprehensive road information Diatom curvilinear equation, for pseudo- lane line curvilinear equation the quantity embodiment of the present invention without limitation.Because of a lane line pair A lane line curvilinear equation is answered, and pseudo- lane line curvilinear equation can be a kind of lane line curvilinear equation, so a pseudo- vehicle Diatom curvilinear equation can correspond to a pseudo- lane line.It is pseudo- when the quantity of the pseudo- lane line curvilinear equation identified is one Lane line device for eliminating can reject the corresponding pseudo- lane line of the puppet lane line curvilinear equation from lane line；When identifying The quantity of pseudo- lane line curvilinear equation when being multiple, pseudo- lane line device for eliminating can reject each pseudo- lane from lane line The corresponding pseudo- lane line of line curvilinear equation.
In the embodiment of the present invention, implement above-mentioned step 304~step 305, by the boundary information and lane line of road Curvilinear equation is integrated into comprehensive road information, pseudo- lane line can be identified in the same standard, to make pseudo- lane line It identifies easier.
Referring to Figure 4 together, Fig. 4 is a kind of schematic diagram of comprehensive road information.Wherein, using the center of vehicle as origin O Plane right-angle coordinate is established, the driving direction of vehicle is the positive direction of the x-axis of plane right-angle coordinate, parallel with x-axis for y Axis, and the positive direction of y-axis is the left side of the driving direction of vehicle, the negative direction of y-axis is the right side of the driving direction of vehicle.Fig. 4 In also include several obstacle informations, positioned at estimate traveling curve on the left of obstacle information be left side obstacle information, position It is right side obstacle information in estimating the obstacle information on the right side of traveling curve, according to several left sides, obstacle information can be intended At road left margin curvilinear equation, curve L is road left margin equation corresponding curve in plane right-angle coordinate for symphysis, Point A be curve L and y-axis intersection point, and the coordinate of point A for (0, a)；According to several right sides, obstacle information can be fitted generation Road right margin curvilinear equation, curve R are road right margin equation corresponding curves in plane right-angle coordinate, and point D is The intersection point of curve R and y-axis, and the coordinate of point D is (0, d).Curve Z1, curve Z2And curve Z3It is lane line curvilinear equation, Point B is curve Z1With the intersection point of y-axis, and the coordinate of point B be (0, b)；Point C is curve Z2Coordinate with the intersection point of y-axis, and point C is (0, c)；Point E is curve Z3With the intersection point of y-axis, and the coordinate of point E be (0, e).Point A is corresponding with the abscissa of point D on the y axis Section (a, d) is the width of real roads, and each lane line curvilinear equation can have intersection point with y-axis, only when lane line curve When the abscissa of equation and the intersection point of y-axis is between section (a, d), just think that the lane line curvilinear equation is that true lane line is bent Line equation, conversely, when the abscissa of lane line curvilinear equation and the intersection point of y-axis is not between section (a, d), then it is assumed that the vehicle Diatom curvilinear equation is pseudo- lane line curvilinear equation.In Fig. 4, the abscissa c of the abscissa b and point C of point B are at section (a, d) Between, therefore, the curve Z where point B1Corresponding lane line curvilinear equation and the curve Z where point C2Corresponding lane line curve Equation is true lane line curvilinear equation；And the abscissa e of point E is between section (a, d), and e < d, therefore, where point E Curve Z3Corresponding lane line curvilinear equation puppet lane line curvilinear equation, it is believed that curve Z3Corresponding lane line is located at road The right of road right margin is needed curve Z3Corresponding puppet lane line curvilinear equation is rejected from lane line curvilinear equation.
It should be noted that the quantity of lane line curvilinear equation can not be 3 in the schematic diagram of comprehensive road information, it is right The quantity of lane line curvilinear equation, the embodiment of the present invention is without limitation.
306, pseudo- lane line device for eliminating is bent by the lane line in comprehensive road information in addition to pseudo- lane line curvilinear equation Line equation is determined as true lane line curvilinear equation.
In the embodiment of the present invention, determining true lane line curvilinear equation can be sent to vehicle by pseudo- lane line device for eliminating Control centre, so that vehicle control center makes the decision of vehicle driving according to true lane line curvilinear equation, to guarantee The accuracy of vehicle driving decision.
307, the display output that pseudo- lane line device for eliminating is arranged by vehicle shows true lane line curvilinear equation pair The true lane line answered.
In the embodiment of the present invention, the quantity of true lane line curvilinear equation can be one or more, in this regard, the present invention is real Apply example without limitation.Because of the corresponding lane line curvilinear equation of lane line, and true lane line curvilinear equation can also be with It is a kind of lane line curvilinear equation, so a true lane line curvilinear equation can correspond to a true lane line.Work as identification When the quantity of true lane line curvilinear equation out is one, pseudo- lane line device for eliminating can export display by display should The really corresponding true lane line of lane line curvilinear equation；When the quantity of the true lane line curvilinear equation identified is more When a, pseudo- lane line device for eliminating can be exported by display shows the corresponding puppet of each true lane line curvilinear equation Lane line.
In the embodiment of the present invention, vehicle can be exported display by way of the analog image by generating present road True lane line, the analog image can be two-dimensional analog image, or three-dimensional simulation image, in this regard, the present invention is implemented Example is without limitation.
In the embodiment of the present invention, implement above-mentioned step 306~step 307, after can eliminating pseudo- lane line Lane line output display, so that the received lane line information of the driver of vehicle is accurate, and then improves the safety of driving Property.
In the method depicted in fig. 3, lane line can be obtained by the road image of road where vehicle, and passed through Obstacle information around road determines the boundary information of road, and the pseudo- lane line being not in road boundary can be picked It removes, ensure that the lane line identified is true lane line, to promote the accuracy of Lane detection.In addition, implementing Method described in Fig. 3 can make the pseudo- lane line identified from lane line more accurate.In addition, described in implementing Fig. 3 Method can identify pseudo- lane line, to keep the identification of pseudo- lane line easier in the same standard.In addition, implementing Fig. 3 Described method, it is ensured that the accuracy of vehicle driving decision.In addition, method described in implementing Fig. 3, can will pick In addition to after pseudo- lane line lane line export display so that the received lane line information of the driver of vehicle be accurately, into And improve the safety of driving.
Example IV
Referring to Fig. 5, Fig. 5 is a kind of structural schematic diagram of pseudo- lane line device for eliminating disclosed by the embodiments of the present invention.Such as Shown in Fig. 5, which may include:
Recognition unit 501 for the road image of road where obtaining vehicle, and identifies lane line from road image.
First fitting unit 502, the obstacle information of the stationary obstruction for obtaining road, and by believing barrier Breath is fitted the boundary information for generating road, and boundary information includes at least the boundary of road.
As an alternative embodiment, the first fitting unit 502 can be also used for:
Obtain the barrier of road；
The speed of barrier is detected by velocity radar；
The barrier that speed in barrier is zero is determined as stationary obstruction.
Wherein, implement this embodiment, can know to avoid other vehicles and/or pedestrian for moving about the vehicle etc. Not Wei barrier, reduce fitting generate road boundary information error.
Culling unit 503 is located at the fitting life of the first fitting unit 502 in the lane line for identifying recognition unit 501 At road boundary except lane line be determined as pseudo- lane line, pseudo- lane line is rejected from lane line.
As an alternative embodiment, culling unit 503 will be located at the lane except the boundary of road in lane line Line is determined as pseudo- lane line, and the mode that pseudo- lane line is rejected from lane line is specifically as follows:
Lane line set is generated in conjunction with the lane line identified；
The lane line in the lane line set is successively traversed, the lane line except the boundary for being located at road is labeled as pseudo- vehicle Diatom, and the lane line within the boundary for being located at road is labeled as true lane line；
It rejects in lane line set and is marked as the lane line of pseudo- lane line, to keep the lane line in lane line set equal For true lane line.
Wherein, implement this embodiment, can determine that each lane line was traversed, and lane line is carried out true Puppet label, so as to the deletion puppet lane line of batch, improves the speed of service of pseudo- lane line device for eliminating.
In pseudo- lane line device for eliminating shown in Fig. 5, lane can be obtained by the road image of road where vehicle Line, and determine by the obstacle information around road the boundary information of road, and can will be not in road boundary Pseudo- lane line is rejected, and ensure that the lane line identified is true lane line, to promote the accuracy of Lane detection. In addition, in pseudo- lane line device for eliminating shown in Fig. 5, it can be to avoid other vehicles and/or row for moving about the vehicle People etc. is identified as barrier, reduces the error that fitting generates the boundary information of road.In addition, pseudo- lane line shown in Fig. 5 In device for eliminating, can batch deletion puppet lane line, improve the speed of service of pseudo- lane line device for eliminating.
Embodiment five
Referring to Fig. 6, Fig. 6 is the structural schematic diagram of another pseudo- lane line device for eliminating disclosed by the embodiments of the present invention. Wherein, pseudo- lane line device for eliminating shown in fig. 6 is that pseudo- lane line device for eliminating as shown in Figure 5 optimizes.With Puppet lane line device for eliminating shown in fig. 5 is compared, and the recognition unit 501 of puppet lane line device for eliminating shown in fig. 6 may include:
First obtains subelement 5011, the road for road where obtaining vehicle using the visual sensor being arranged on vehicle Road image.
Subelement 5012 is extracted, carries out semantic feature knowledge for obtaining the road image that subelement 5011 obtains to first Not, the semantic feature in road image is extracted.
Judgment sub-unit 5013 whether there is and lane line for judging to extract in the semantic feature that subelement 5012 extracts Matched semantic feature.
It identifies subelement 5014, when the result for judging in judgment sub-unit 5013 is is, passes through image recognition technology Identification first obtains the lane line in the road image that subelement 5011 obtains.
In the embodiment of the present invention, can confirmation road image in there are lane line after again go identification road image in Lane line improves to simplify the operation for being carried out the lane line of identification to each road image to road image processing Speed.
As an alternative embodiment, identification subelement 5014 is identified in road image by image recognition technology The mode of lane line is specifically as follows:
By image Segmentation Technology by the Road image segmentation be several subgraphs；
The image, semantic feature of several subgraphs is obtained using deep learning algorithm；
The determining and matched target image semantic feature of lane line semantic feature from image, semantic feature；
Pass through the corresponding lane line of image recognition technology recognition target image semantic feature, a target image semantic feature A corresponding lane line.
Wherein, implement this embodiment, can use image Segmentation Technology, deep learning algorithm and image recognition skill Art identifies lane line present in road image jointly, so that lane line present in road image can be known as far as possible It does not come out, improves the accuracy rate of Lane detection.
As an alternative embodiment, the first fitting unit 502 of puppet lane line device for eliminating shown in fig. 6 can be with Include:
Second obtains subelement 5021, for obtaining the obstacle of stationary obstruction using the radar sensor being arranged on vehicle Object information, obstacle information include the orientation of estimating traveling curve of the stationary obstruction relative to vehicle；
Subelement 5022 is analyzed, for obtaining the side in the obstacle information that subelement 5021 obtains by analysis second Obstacle information is divided into left side obstacle information and right side obstacle information by position；
It is fitted subelement 5023, is hindered for analyzing obtained left side to analysis subelement 5022 respectively using least square method Hinder object information and right side obstacle information to be fitted, generates road left margin curvilinear equation corresponding with left side obstacle information And road right margin curvilinear equation corresponding with right side obstacle information；
Computation subunit 5024, the road left margin curvilinear equation and road for being generated according to fitting subelement 5023 are right Boundary curve equation calculates the boundary information for generating road.
Wherein, implement this embodiment, road left margin curvilinear equation and road right margin curvilinear equation can be generated, And then the boundary information of road is calculated, so that the boundary information for the road being calculated is more accurate.
As an alternative embodiment, puppet lane line device for eliminating shown in fig. 6 can also include:
Acquiring unit 504, for quiet using the radar sensor acquisition being arranged on vehicle in the second acquisition subelement 5021 Only before the obstacle information of barrier, the running condition information of vehicle is obtained, running condition information includes at least the speed of vehicle The corner and yaw velocity of degree, runner；
Unit 505 is estimated, the running condition information for being obtained according to acquiring unit 504 estimates the driving trace of vehicle, And traveling curve is estimated by what driving trace was determined as vehicle, and triggering second obtains subelement 5021 and executes using on vehicle The radar sensor of setting obtains the obstacle information of stationary obstruction.
Wherein, implement this embodiment, the driving trace of vehicle is estimated out according to the current driving status of vehicle, in turn Vehicle is calculated estimates traveling curve, estimates traveling curve so that pseudo- lane line device for eliminating can use this and accurately knows It Chu not pseudo- lane line.
In pseudo- lane line device for eliminating shown in Fig. 6, lane can be obtained by the road image of road where vehicle Line, and determine by the obstacle information around road the boundary information of road, and can will be not in road boundary Pseudo- lane line is rejected, and ensure that the lane line identified is true lane line, to promote the accuracy of Lane detection. In addition, in pseudo- lane line device for eliminating shown in Fig. 6, the lane line that identification is carried out to each road image can simplify Operation improves the speed to road image processing.In addition, road can be made in pseudo- lane line device for eliminating shown in Fig. 6 Lane line present in image can identify as far as possible, improve the accuracy rate of Lane detection.In addition, in Fig. 6 Shown in puppet lane line device for eliminating, the boundary information for the road being calculated can be made more accurate.In addition, in Fig. 6 Shown in puppet lane line device for eliminating, pseudo- lane line device for eliminating can be made to can use this and estimate traveling curve and accurately known It Chu not pseudo- lane line.
Embodiment six
Referring to Fig. 7, Fig. 7 is the structural schematic diagram of another pseudo- lane line device for eliminating disclosed by the embodiments of the present invention. Wherein, pseudo- lane line device for eliminating shown in Fig. 7 is that pseudo- lane line device for eliminating as shown in Figure 6 optimizes.With Puppet lane line device for eliminating shown in fig. 6 is compared, and puppet lane line device for eliminating shown in Fig. 7 can also include:
Second fitting unit 506, for the road image of road where obtaining vehicle in recognition unit 501, and from road In image identify lane line after, and culling unit 503 will in lane line positioned at road boundary except lane line it is true It is set to before pseudo- lane line, the lane line in road image is fitted, generates lane line curve side corresponding with lane line Journey, wherein the corresponding lane line curvilinear equation of a lane line.
In the embodiment of the present invention, it is bent the corresponding lane line of each lane line can be generated according to the lane line computation identified Line equation, so that the subsequent pseudo- lane line identified from lane line is more accurate.
As an alternative embodiment, the culling unit 503 of puppet lane line device for eliminating shown in Fig. 7 can wrap It includes:
Subelement 5031 is generated, for being believed in conjunction with the boundary of the road of the first fitting unit 502 generation using vehicle as foundation The corresponding lane line curvilinear equation of lane line that breath and the second fitting unit 506 generate generates comprehensive road information；
Subelement 5032 is rejected, for determining from the synthesis road information for generating the generation of subelement 5031 in road Pseudo- lane line curvilinear equation except boundary, and the corresponding pseudo- lane line of pseudo- lane line curvilinear equation is rejected from lane line.
Wherein, implement this embodiment, the boundary information of road and lane line curvilinear equation are integrated into comprehensive road In the information of road, pseudo- lane line can be identified in the same standard, to keep the identification of pseudo- lane line easier.
As an alternative embodiment, puppet lane line device for eliminating shown in Fig. 7 can also include:
Determination unit 507, for reject subelement 5032 from comprehensive road information and determine the boundary in road it Outer pseudo- lane line curvilinear equation, and after rejecting the corresponding pseudo- lane line of pseudo- lane line curvilinear equation in lane line, it will give birth to In the synthesis road information generated at subelement 5031 in addition to rejecting the pseudo- lane line curvilinear equation that subelement 5032 generates Lane line curvilinear equation is determined as true lane line curvilinear equation；
Output unit 508, the true lane that the display output display determination unit 507 for being arranged by vehicle determines The corresponding true lane line of line curvilinear equation.
Wherein, implement this embodiment, the output display of the lane line after pseudo- lane line can will be eliminated, so that vehicle The received lane line information of driver be accurately, and then to improve the safety of driving.
In pseudo- lane line device for eliminating shown in Fig. 7, lane can be obtained by the road image of road where vehicle Line, and determine by the obstacle information around road the boundary information of road, and can will be not in road boundary Pseudo- lane line is rejected, and ensure that the lane line identified is true lane line, to promote the accuracy of Lane detection. In addition, in pseudo- lane line device for eliminating shown in Fig. 7 each lane line pair can be generated according to the lane line computation identified The lane line curvilinear equation answered, so that the subsequent pseudo- lane line identified from lane line is more accurate.In addition, shown in Fig. 7 Pseudo- lane line device for eliminating in, pseudo- lane line can be identified in the same standard, to make the identification of pseudo- lane line more It is easy.In addition, the received lane line information of the driver of vehicle can be made to be in pseudo- lane line device for eliminating shown in Fig. 7 Accurately, and then the safety of driving is improved.
Embodiment seven
Referring to Fig. 8, Fig. 8 is the structural schematic diagram of another pseudo- lane line device for eliminating disclosed by the embodiments of the present invention. As shown in figure 8, the puppet lane line device for eliminating may include:
It is stored with the memory 801 of executable program code；
The processor 802 coupled with memory 801；
Wherein, processor 802 calls the executable program code stored in memory 801, executes the above each method and implements Some or all of method in example step.
A kind of computer readable storage medium is also disclosed in the embodiment of the present invention, wherein computer-readable recording medium storage Program code, wherein program code includes for executing some or all of the method in above each method embodiment step Instruction.
A kind of computer program product is also disclosed in the embodiment of the present invention, wherein when computer program product on computers When operation, so that computer executes some or all of the method in such as above each method embodiment step.
The embodiment of the present invention is also disclosed a kind of using distribution platform, wherein using distribution platform for issuing computer journey Sequence product, wherein when computer program product is run on computers, so that computer executes such as the above each method embodiment In some or all of method step.
It should be understood that " embodiment of the present invention " that specification is mentioned in the whole text mean special characteristic related with embodiment, Structure or characteristic is included at least one embodiment of the present invention.Therefore, " the present invention reality occurred everywhere in the whole instruction Apply in example " not necessarily refer to identical embodiment.In addition, these a particular feature, structure, or characteristics can be with any suitable side Formula combines in one or more embodiments.Those skilled in the art should also know that embodiment described in this description is equal Belong to alternative embodiment, related actions and modules are not necessarily necessary for the present invention.
In various embodiments of the present invention, it should be appreciated that magnitude of the sequence numbers of the above procedures are not meant to execute suitable Successively, the execution sequence of each process should be determined by its function and internal logic the certainty of sequence, without coping with the embodiment of the present invention Implementation process constitutes any restriction.
In embodiment provided by the present invention, it should be appreciated that " B corresponding with A " indicates that B is associated with A, can be with according to A Determine B.It is also to be understood that determine that B is not meant to determine B only according to A according to A, it can also be according to A and/or other information Determine B.
Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of above-described embodiment is can It is completed with instructing relevant hardware by program, which can be stored in a computer readable storage medium, storage Medium include read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), programmable read only memory (Programmable Read-only Memory, PROM), erasable programmable is read-only deposits Reservoir (Erasable Programmable Read Only Memory, EPROM), disposable programmable read-only memory (One- Time Programmable Read-Only Memory, OTPROM), the electronics formula of erasing can make carbon copies read-only memory (Electrically-Erasable Programmable Read-Only Memory, EEPROM), CD-ROM (Compact Disc Read-Only Memory, CD-ROM) or other disc memories, magnetic disk storage, magnetic tape storage or can For carrying or any other computer-readable medium of storing data.
Above-mentioned unit as illustrated by the separation member may or may not be physically separated, aobvious as unit The component shown may or may not be physical unit, can be in one place, or may be distributed over multiple nets On network unit.Some or all of units can be selected to achieve the purpose of the solution of this embodiment according to the actual needs.
In addition, each functional unit in various embodiments of the present invention can integrate in one processing unit, it is also possible to Each unit physically exists alone, and can also be integrated in one unit with two or more units.Above-mentioned integrated unit Both it can take the form of hardware realization, can also realize in the form of software functional units.
If above-mentioned integrated unit is realized in the form of SFU software functional unit and when sold or used as an independent product, It can store in a retrievable memory of computer.Based on this understanding, technical solution of the present invention substantially or Person says all or part of of the part that contributes to existing technology or the technical solution, can be in the form of software products It embodies, which is stored in a memory, including several requests are with so that a computer is set Standby (can be personal computer, server or network equipment etc., specifically can be the processor in computer equipment) executes Some or all of each embodiment above method of the invention step.
A kind of pseudo- lane line elimination method disclosed by the embodiments of the present invention and device are described in detail above, herein In apply that a specific example illustrates the principle and implementation of the invention, the explanation of above example is only intended to sides Assistant solves method and its core concept of the invention；At the same time, for those skilled in the art, think of according to the present invention Think, there will be changes in the specific implementation manner and application range, in conclusion the content of the present specification should not be construed as pair Limitation of the invention.

Claims (12)

1. a kind of puppet lane line elimination method, which is characterized in that the described method includes:
The road image of road where obtaining vehicle, and lane line is identified from the road image；
The obstacle information of the stationary obstruction of the road is obtained, and generates institute by being fitted to the obstacle information The boundary information of road is stated, the boundary information includes at least the boundary of the road；
The lane line being located at except the boundary of the road in the lane line is determined as pseudo- lane line, from the lane line Reject the pseudo- lane line.
2. the method according to claim 1, wherein it is described obtain vehicle where road road image, and from Lane line is identified in the road image, comprising:
The road image of road where obtaining the vehicle using the visual sensor being arranged on vehicle；
Semantic feature identification is carried out to the road image, extracts the semantic feature in the road image；
Judge to whether there is and the matched semantic feature of lane line in the semantic feature；
If it does, identifying the lane line in the road image by image recognition technology.
3. method according to claim 1 or 2, which is characterized in that the barrier of the stationary obstruction for obtaining the road Hinder object information, and generate the boundary information of the road by being fitted to the obstacle information, comprising:
The obstacle information of stationary obstruction, the obstacle information packet are obtained using the radar sensor being arranged on the vehicle Include the orientation of estimating traveling curve of the stationary obstruction relative to the vehicle；
By analyzing the orientation in the obstacle information, the obstacle information is divided into left side obstacle information and the right side Side obstacle information；
The left side obstacle information and the right side obstacle information are fitted respectively using least square method, generate with The corresponding road left margin curvilinear equation of left side obstacle information and road corresponding with the right side obstacle information Right margin curvilinear equation；
The boundary for generating the road is calculated according to the road left margin curvilinear equation and the road right margin curvilinear equation Information.
4. according to the method described in claim 3, it is characterized in that, described obtained using the radar sensor being arranged on the vehicle Before taking the obstacle information of stationary obstruction, the method also includes:
Obtain the running condition information of the vehicle, the running condition information includes at least the speed of the vehicle, runner Corner and yaw velocity；
The driving trace of the vehicle is estimated according to the running condition information, and the driving trace is determined as the vehicle Estimate traveling curve.
5. method according to any one of claims 1 to 4, which is characterized in that the road for obtaining vehicle place road Image, and from the road image identify lane line after and it is described will in the lane line be located at the road side Lane line except boundary is determined as before pseudo- lane line, the method also includes:
The lane line in the road image is fitted, lane line curve side corresponding with the lane line is generated Journey, wherein the corresponding lane line curvilinear equation of a lane line；
It is described that the lane line being located at except the boundary of the road in the lane line is determined as pseudo- lane line, from the lane The pseudo- lane line is rejected in line, comprising:
Using the vehicle as foundation, in conjunction with the corresponding lane line curve of the boundary information of the road and the lane line Equation generates comprehensive road information；
From the pseudo- lane line curvilinear equation determined in the comprehensive road information except the boundary in the road, and from described The corresponding pseudo- lane line of the pseudo- lane line curvilinear equation is rejected in lane line.
6. according to the method described in claim 5, it is characterized in that, described determine from the comprehensive road information in described Pseudo- lane line curvilinear equation except the boundary of road, and reject the pseudo- lane line curvilinear equation from the lane line and correspond to Pseudo- lane line after, the method also includes:
The lane line curvilinear equation in the comprehensive road information in addition to the pseudo- lane line curvilinear equation is determined as True lane line curvilinear equation；
The display output being arranged by the vehicle shows the corresponding true lane line of the true lane line curvilinear equation.
7. a kind of puppet lane line device for eliminating characterized by comprising
Recognition unit for the road image of road where obtaining vehicle, and identifies lane line from the road image；
First fitting unit, the obstacle information of the stationary obstruction for obtaining the road, and by the barrier Information is fitted the boundary information for generating the road, and the boundary information includes at least the boundary of the road；
Culling unit, for the lane line being located at except the boundary of the road in the lane line to be determined as pseudo- lane line, The pseudo- lane line is rejected from the lane line.
8. puppet lane line device for eliminating according to claim 7, which is characterized in that the recognition unit includes:
First obtains subelement, the mileage chart for road where obtaining the vehicle using the visual sensor being arranged on vehicle Picture；
Subelement is extracted, for carrying out semantic feature identification to the road image, the semanteme extracted in the road image is special Sign；
Judgment sub-unit whether there is and the matched semantic feature of lane line in the semantic feature for judging；
Identify subelement, when result for judging in the judgment sub-unit is is, identified by image recognition technology described in Lane line in road image.
9. puppet lane line device for eliminating according to claim 7 or 8, which is characterized in that first fitting unit includes:
Second obtains subelement, and the barrier for being obtained stationary obstruction using the radar sensor being arranged on the vehicle is believed Breath, the obstacle information include the orientation of estimating traveling curve of the stationary obstruction relative to the vehicle；
Subelement is analyzed, for by analyzing the orientation in the obstacle information, the obstacle information to be divided into a left side Side obstacle information and right side obstacle information；
It is fitted subelement, for utilizing least square method respectively to the left side obstacle information and the right side obstacle information Be fitted, generate corresponding with left side obstacle information road left margin curvilinear equation and with the right side barrier The corresponding road right margin curvilinear equation of information；
Computation subunit is generated for being calculated according to the road left margin curvilinear equation and the road right margin curvilinear equation The boundary information of the road.
10. puppet lane line device for eliminating according to claim 9, which is characterized in that the puppet lane line device for eliminating is also Include:
Acquiring unit, for obtaining static barrier using the radar sensor being arranged on the vehicle in the second acquisition subelement Before the obstacle information for hindering object, the running condition information of the vehicle is obtained, the running condition information includes at least described The speed of vehicle, the corner of runner and yaw velocity；
Unit is estimated, for estimating the driving trace of the vehicle according to the running condition information, and by the driving trace Be determined as the vehicle estimates traveling curve, and utilizes the vehicle described in triggering the second acquisition subelement execution The radar sensor of upper setting obtains the obstacle information of stationary obstruction.
11. according to the described in any item pseudo- lane line device for eliminating of claim 7~10, which is characterized in that the puppet lane line Device for eliminating further include:
Second fitting unit, for the road image of road where obtaining vehicle in the recognition unit, and from the mileage chart As in identify lane line after, and the culling unit will in the lane line positioned at the road boundary except vehicle Diatom is determined as before pseudo- lane line, is fitted to the lane line in the road image, generates and the lane line Corresponding lane line curvilinear equation, wherein the corresponding lane line curvilinear equation of a lane line；
The culling unit includes:
Subelement is generated, for being corresponded in conjunction with the boundary information of the road and the lane line using the vehicle as foundation The lane line curvilinear equation generate comprehensive road information；
Subelement is rejected, for bent from the pseudo- lane line except the boundary in the road determining in the comprehensive road information Line equation, and the corresponding pseudo- lane line of the pseudo- lane line curvilinear equation is rejected from the lane line.
12. puppet lane line device for eliminating according to claim 11, which is characterized in that the puppet lane line device for eliminating is also Include:
Determination unit, for it is described rejecting subelement from the comprehensive road information determine in the road boundary it Outer pseudo- lane line curvilinear equation, and reject from the lane line the corresponding pseudo- lane line of the pseudo- lane line curvilinear equation it Afterwards, the lane line curvilinear equation in the comprehensive road information in addition to the pseudo- lane line curvilinear equation is determined as very Real vehicle diatom curvilinear equation；
Output unit, the display output for being arranged by the vehicle show that the true lane line curvilinear equation is corresponding True lane line.
CN201810914211.4A 2018-08-10 2018-08-10 A kind of puppet lane line elimination method and device Pending CN109271857A (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201810914211.4A CN109271857A (en) 2018-08-10 2018-08-10 A kind of puppet lane line elimination method and device
PCT/CN2019/093423 WO2020029706A1 (en) 2018-08-10 2019-06-28 Dummy lane line elimination method and apparatus

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810914211.4A CN109271857A (en) 2018-08-10 2018-08-10 A kind of puppet lane line elimination method and device

Publications (1)

Publication Number Publication Date
CN109271857A true CN109271857A (en) 2019-01-25

Family

ID=65153711

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810914211.4A Pending CN109271857A (en) 2018-08-10 2018-08-10 A kind of puppet lane line elimination method and device

Country Status (2)

Country Link
CN (1) CN109271857A (en)
WO (1) WO2020029706A1 (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110414487A (en) * 2019-08-16 2019-11-05 东软睿驰汽车技术（沈阳）有限公司 Method and device for identifying lane line
CN110413942A (en) * 2019-06-04 2019-11-05 联创汽车电子有限公司 Lane line equation screening technique and its screening module
WO2020029706A1 (en) * 2018-08-10 2020-02-13 广州小鹏汽车科技有限公司 Dummy lane line elimination method and apparatus
CN111797701A (en) * 2020-06-10 2020-10-20 东莞正扬电子机械有限公司 Road obstacle sensing method and system for vehicle multi-sensor fusion system
CN111814724A (en) * 2020-07-20 2020-10-23 北京百度网讯科技有限公司 Method, device, equipment and storage medium for identifying number of lanes
CN112132109A (en) * 2020-10-10 2020-12-25 北京百度网讯科技有限公司 Lane line processing and lane positioning method, device, equipment and storage medium
CN112441022A (en) * 2019-09-02 2021-03-05 华为技术有限公司 Lane center line determining method and device

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR3123745A1 (en) * 2021-06-07 2022-12-09 Psa Automobiles Sa Method for assisting the driving of a vehicle traveling on a traffic lane delimited by a first and a second line marked on the ground.
CN114399588B (en) * 2021-12-20 2022-11-11 禾多科技（北京）有限公司 Three-dimensional lane line generation method and device, electronic device and computer readable medium

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020061123A1 (en) * 2000-11-22 2002-05-23 Nissan Motor Co., Ltd. Apparatus and method for detecting road white line for automotive vehicle
CN102591332A (en) * 2011-01-13 2012-07-18 同济大学 Device and method for local path planning of pilotless automobile
CN103177246A (en) * 2013-03-26 2013-06-26 北京理工大学 Dual-model lane line identification method based on dynamic area division
CN104766058A (en) * 2015-03-31 2015-07-08 百度在线网络技术（北京）有限公司 Method and device for obtaining lane line
CN106570446A (en) * 2015-10-12 2017-04-19 腾讯科技（深圳）有限公司 Lane line extraction method and device
CN107389084A (en) * 2017-06-09 2017-11-24 深圳市速腾聚创科技有限公司 Planning driving path planing method and storage medium
CN107730520A (en) * 2017-09-22 2018-02-23 智车优行科技（北京）有限公司 Method for detecting lane lines and system
CN108021844A (en) * 2016-10-31 2018-05-11 高德软件有限公司 A kind of road edge recognition methods and device
CN108267747A (en) * 2017-01-03 2018-07-10 中交宇科（北京）空间信息技术有限公司 Road feature extraction method and apparatus based on laser point cloud

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105667518B (en) * 2016-02-25 2018-07-24 福州华鹰重工机械有限公司 The method and device of lane detection
KR101864066B1 (en) * 2017-01-11 2018-07-05 숭실대학교산학협력단 Lane marking detection device, Lane departure determination device, Lane marking detection method and Lane departure determination method
CN109271857A (en) * 2018-08-10 2019-01-25 广州小鹏汽车科技有限公司 A kind of puppet lane line elimination method and device

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020061123A1 (en) * 2000-11-22 2002-05-23 Nissan Motor Co., Ltd. Apparatus and method for detecting road white line for automotive vehicle
CN102591332A (en) * 2011-01-13 2012-07-18 同济大学 Device and method for local path planning of pilotless automobile
CN103177246A (en) * 2013-03-26 2013-06-26 北京理工大学 Dual-model lane line identification method based on dynamic area division
CN104766058A (en) * 2015-03-31 2015-07-08 百度在线网络技术（北京）有限公司 Method and device for obtaining lane line
CN106570446A (en) * 2015-10-12 2017-04-19 腾讯科技（深圳）有限公司 Lane line extraction method and device
CN108021844A (en) * 2016-10-31 2018-05-11 高德软件有限公司 A kind of road edge recognition methods and device
CN108267747A (en) * 2017-01-03 2018-07-10 中交宇科（北京）空间信息技术有限公司 Road feature extraction method and apparatus based on laser point cloud
CN107389084A (en) * 2017-06-09 2017-11-24 深圳市速腾聚创科技有限公司 Planning driving path planing method and storage medium
CN107730520A (en) * 2017-09-22 2018-02-23 智车优行科技（北京）有限公司 Method for detecting lane lines and system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
李松泽: "基于深度学习的车道线检测系统的设计与实现", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020029706A1 (en) * 2018-08-10 2020-02-13 广州小鹏汽车科技有限公司 Dummy lane line elimination method and apparatus
CN110413942A (en) * 2019-06-04 2019-11-05 联创汽车电子有限公司 Lane line equation screening technique and its screening module
CN110413942B (en) * 2019-06-04 2023-08-08 上海汽车工业(集团)总公司 Lane line equation screening method and screening module thereof
CN110414487A (en) * 2019-08-16 2019-11-05 东软睿驰汽车技术（沈阳）有限公司 Method and device for identifying lane line
CN112441022A (en) * 2019-09-02 2021-03-05 华为技术有限公司 Lane center line determining method and device
CN111797701A (en) * 2020-06-10 2020-10-20 东莞正扬电子机械有限公司 Road obstacle sensing method and system for vehicle multi-sensor fusion system
CN111797701B (en) * 2020-06-10 2024-05-24 广东正扬传感科技股份有限公司 Road obstacle sensing method and system for vehicle multi-sensor fusion system
CN111814724A (en) * 2020-07-20 2020-10-23 北京百度网讯科技有限公司 Method, device, equipment and storage medium for identifying number of lanes
CN111814724B (en) * 2020-07-20 2023-07-04 北京百度网讯科技有限公司 Lane number identification method, device, equipment and storage medium
CN112132109A (en) * 2020-10-10 2020-12-25 北京百度网讯科技有限公司 Lane line processing and lane positioning method, device, equipment and storage medium

Also Published As

Publication number Publication date
WO2020029706A1 (en) 2020-02-13

Similar Documents

Publication Publication Date Title
CN109271857A (en) A kind of puppet lane line elimination method and device
Song et al. Real-time obstacles detection and status classification for collision warning in a vehicle active safety system
US11593950B2 (en) System and method for movement detection
CN108647646B (en) Low-beam radar-based short obstacle optimized detection method and device
Laugier et al. Probabilistic analysis of dynamic scenes and collision risks assessment to improve driving safety
EP2372310B1 (en) Image processing system and position measurement system
Nguyen et al. Stereo-camera-based urban environment perception using occupancy grid and object tracking
US20180348374A1 (en) Range-View Lidar-Based Object Detection
US8452103B2 (en) Scene matching reference data generation system and position measurement system
Wu et al. Applying a functional neurofuzzy network to real-time lane detection and front-vehicle distance measurement
CN110362077A (en) Automatic driving vehicle urgent danger prevention decision system, method and medium
Shim et al. An autonomous driving system for unknown environments using a unified map
CN107646114A (en) Method for estimating track
CN111771207A (en) Enhanced vehicle tracking
CN106503653A (en) Area marking method, device and electronic equipment
CN109583416B (en) Pseudo lane line identification method and system
AlZu'bi et al. Data fusion in autonomous vehicles research, literature tracing from imaginary idea to smart surrounding community
US20210149408A1 (en) Generating Depth From Camera Images and Known Depth Data Using Neural Networks
JP2020052695A (en) Object detection apparatus, object detection method, and computer program for object detection
KR102337352B1 (en) Method and device for providing advanced pedestrian assistance system to protect pedestrian preoccupied with smartphone
CN114061581A (en) Ranking agents in proximity to autonomous vehicles by mutual importance
CN116830164A (en) LiDAR decorrelated object detection system and method
Liu et al. Dynamic scene’s laser localization by NeuroIV-based moving objects detection and LiDAR points evaluation
López et al. Computer vision in vehicle technology: Land, sea, and air
CN115158309A (en) Factor trajectory prediction using context sensitive fusion

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20190125

RJ01 Rejection of invention patent application after publication