CN108241819A - The recognition methods of pavement markers and device - Google Patents
The recognition methods of pavement markers and device Download PDFInfo
- Publication number
- CN108241819A CN108241819A CN201611201463.XA CN201611201463A CN108241819A CN 108241819 A CN108241819 A CN 108241819A CN 201611201463 A CN201611201463 A CN 201611201463A CN 108241819 A CN108241819 A CN 108241819A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- cloud data
- sample point
- road
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
Recognition methods and device this application provides a kind of pavement markers, method therein include：Along the lane line of road, from having been labeled as in the point cloud data of a pavement markers of the road, sample point cloud data is obtained；The sample point cloud data is ranked up along the direction of the road, the sample point cloud data after being sorted；Along the direction of road, the sample point cloud data after sequence is divided into sample point cloud data block by preset spacing distance；Each sample point cloud data block is counted, obtains the distributed data of the sample point cloud data；According to the distributed data of the sample point cloud data, judge whether the regularity of distribution of the sample point cloud data meets the regularity of distribution of the point cloud data of the pavement markers of preset property, if so, the pavement markers for being identified as the preset property by the pavement markers.The present invention makes full use of the point cloud distribution character of some pavement markers, can efficiently and accurately identify the pavement markers with cloud characteristic.
Description
Technical field
Recognition methods and device this application involves electronic map technique field more particularly to a kind of pavement markers.
Background technology
Pavement markers refer to use lines, arrow, word, object marking, protuberant guide post and delineator on the road surface of road
Deng the mark to traffic informations such as traffic participant transmission guiding, limitation, warnings, effect is control and guiding traffic.Road surface
Label is also referred to as pavement marker, road sign etc..
In the fields such as electronic map and automatic Pilot, need to identify all kinds of pavement markers.For example, making or updating electricity
During sub- map, need to obtain the pavement markers information on road, such as lane line, deceleration label, crossing, arrow,
Position accuracy demand is generally less than 10cm.
At present, there are mainly three types of the method for pavement markers identification substantially, each method respectively has feature.
The first is manual identified.The video data collected by collecting vehicle, finds road marking by hand in video data
Remember, then type.This method whole process needs manual operations, inefficiency, it is impossible to meet the processing requirement of mass data.
Second is pattern matching algorithm automatic identification.It is configured with the dimensioning of various pavement markers in a program in advance
The information such as very little, reflectivity, dot density judge the type of pavement markers, the class of output probability maximum by the method for Model Matching
Type.This method depends on the accuracy of model library due to pattern match result, and there are similar for the dimension information of some models
Property, such as the dimension information of slow down label and the models such as Chinese character, space confirmation line is similar, therefore recognition accuracy is not high.
The third is image matching algorithm automatic identification.Using the method for images match, image is converted the data into, then
It is identified with the model library of image.Then treatment effeciency is relatively low for some pavement markers for this method, for example, for by a large amount of points
The deceleration label image that cloud data are formed, due to each pixel of image to be handled, executing efficiency is relatively low.
Invention content
One of the technical issues of the application solves is to provide recognition methods and the device of a kind of pavement markers, by using point
The distribution character of cloud data efficiently carries out the identification of pavement markers.
According to the application one embodiment, a kind of recognition methods of pavement markers is provided, this method includes the following steps：Edge
The lane line of road from having been labeled as in the point cloud data of a pavement markers of the road, obtains sample point cloud data；It will
The sample point cloud data is ranked up along the direction of the road, the sample point cloud data after being sorted；Along the side of road
To the sample point cloud data after sequence is divided into sample point cloud data block by preset spacing distance；To each sample point cloud
Data block is counted, and obtains the distributed data of the sample point cloud data；According to the distributed data of the sample point cloud data,
Judge whether the regularity of distribution of the sample point cloud data meets the regularity of distribution of the point cloud data of the pavement markers of preset property,
If so, the pavement markers for being identified as the preset property by the pavement markers.
According to another embodiment of the application, a kind of identification device of pavement markers is provided, described device includes：It obtains single
Member for along the lane line of road, from having been labeled as in the point cloud data of a pavement markers of the road, obtains sample point
Cloud data；Sequencing unit, for the sample point cloud data to be ranked up along the direction of the road, the sample after being sorted
This point cloud data；Division unit, for along the direction of road, the sample point cloud data after sequence to be drawn by preset spacing distance
It is divided into sample point cloud data block；Statistic unit for being counted to each sample point cloud data block, obtains the sample point cloud
The distributed data of data；Recognition unit for the distributed data according to the sample point cloud data, judges the sample point cloud number
According to the regularity of distribution whether meet preset property pavement markers point cloud data the regularity of distribution, if so, by the road surface
The pavement markers for being identified as the preset property of label.
The embodiment of the present invention makes full use of the point cloud distribution character of some pavement markers, for having been labeled as pavement markers
Point cloud data is ranked up and counts, and obtains the regularity of distribution of point cloud data, and passes through the regularity of distribution and determine whether default spy
The pavement markers of property.The present invention program automatic identification without human intervention, so cost is very low, has very strong practicability.This
Invention makes full use of the point cloud distribution character of pavement markers, can efficiently and accurately identify the road marking with preset property
Note.
Although those of ordinary skill in the art will be appreciated that following detailed description carries out referenced in schematic embodiment, attached drawing,
But the application is not limited in these embodiments.
Description of the drawings
By reading the detailed description made to non-limiting example made with reference to the following drawings, the application's is other
Feature, objects and advantages will become more apparent upon：
Fig. 1 is the flow chart according to the recognition methods of the pavement markers of the embodiment of the present application；
Fig. 2 is to mark schematic diagram according to the transversal deceleration of the embodiment of the present application；
Fig. 3 is to mark schematic diagram according to the longitudinal deceleration of the embodiment of the present application；
Fig. 4 is the method flow diagram being ranked up to sample point cloud data along road direction according to the embodiment of the present application；
Fig. 5 is the sample point cloud area schematic obtained according to the embodiment of the present application；
Fig. 6 (a), Fig. 6 (b), Fig. 6 (c) are three kinds of pavement markers schematic diagrames to be identified of the embodiment of the present application；
Fig. 7 is the sliding window schematic diagram according to the embodiment of the present application；
Fig. 8 is the statistical information schematic diagram to tri- kinds of pavement markers to be identified of Fig. 6 according to the embodiment of the present application；
Fig. 9 is the structure diagram according to the pavement markers identification device of the embodiment of the present application.
Although those of ordinary skill in the art will be appreciated that following detailed description carries out referenced in schematic embodiment, attached drawing,
But the application is not limited in these embodiments.But scope of the present application is extensive, and is intended to be bound only by appended right
It is required that limit scope of the present application.
Specific embodiment
It should be mentioned that some exemplary embodiments are described as before exemplary embodiment is discussed in greater detail
The processing described as flow chart or method.Although operations are described as the processing of sequence by flow chart, therein to be permitted
Multioperation can be implemented concurrently, concomitantly or simultaneously.In addition, the sequence of operations can be rearranged.When it
The processing can be terminated when operation is completed, it is also possible to have the additional step being not included in attached drawing.The processing
It can correspond to method, function, regulation, subroutine, subprogram etc..
The computer equipment includes user equipment and the network equipment.Wherein, the user equipment includes but not limited to electricity
Brain, smart mobile phone, PDA etc.；The network equipment includes but not limited to single network server, multiple network servers form
Server group or the cloud being made of a large amount of computers or network server based on cloud computing (Cloud Computing), wherein,
Cloud computing is one kind of Distributed Calculation, a super virtual computer being made of the computer collection of a group loose couplings.Its
In, the computer equipment can isolated operation realize the application, also can access network and by with other calculating in network
The application is realized in the interactive operation of machine equipment.Wherein, the network residing for the computer equipment include but not limited to internet,
Wide area network, Metropolitan Area Network (MAN), LAN, VPN network etc..
It should be noted that the user equipment, the network equipment and network etc. are only for example, other are existing or from now on may be used
The computer equipment or network that can occur such as are applicable to the application, should also be included within the application protection domain, and to draw
It is incorporated herein with mode.
Method (some of them are illustrated by flow) discussed hereafter can be by hardware, software, firmware, centre
Part, microcode, hardware description language or its arbitrary combination are implemented.Implement when with software, firmware, middleware or microcode
When, to implement the program code of necessary task or code segment can be stored in machine or computer-readable medium and (for example deposit
Storage media) in.(one or more) processor can implement necessary task.
Specific structure and function details disclosed herein are only representative, and are for describing showing for the application
The purpose of example property embodiment.But the application can be implemented, and be not interpreted as by many alternative forms
It is limited only by the embodiments set forth herein.
Although it should be understood that may have been used term " first ", " second " etc. herein to describe each unit,
But these units should not be limited by these terms.The use of these items is only for by a unit and another unit
It distinguishes.For example, in the case of the range without departing substantially from exemplary embodiment, it is single that first unit can be referred to as second
Member, and similarly second unit can be referred to as first unit.Term "and/or" used herein above include one of them or
The arbitrary and all combination of more listed associated items.
It should be understood that when a unit is referred to as " connecting " or during " coupled " to another unit, can directly connect
It connects or is coupled to another unit or there may be temporary location.In contrast, when a unit is referred to as " directly connecting
Connect " or " direct-coupling " to another unit when, then there is no temporary locations.It should explain in a comparable manner and be used to retouch
State the relationship between unit other words (such as " between being in ... " compared to " between being directly in ... ", " and with ... it is adjacent
Closely " compared to " with ... be directly adjacent to " etc.).
Term used herein above is not intended to limit exemplary embodiment just for the sake of description specific embodiment.Unless
Context clearly refers else, otherwise singulative used herein above "one", " one " also attempt to include plural number.Should also
When understanding, term " comprising " and/or "comprising" used herein above provide stated feature, integer, step, operation,
The presence of unit and/or component, and do not preclude the presence or addition of other one or more features, integer, step, operation, unit,
Component and/or a combination thereof.
It should further be mentioned that in some replaces realization modes, the function/action being previously mentioned can be according to different from attached
The sequence indicated in figure occurs.For example, depending on involved function/action, the two width figures shown in succession actually may be used
Substantially simultaneously to perform or can perform in a reverse order sometimes.
The technical solution of the application is described in further detail below in conjunction with the accompanying drawings.
Fig. 1 is according to the flow chart of the recognition methods of the pavement markers of the embodiment of the present application, and this method mainly includes as follows
Step：
S101：Along the lane line of road, from having been labeled as in the point cloud data of a pavement markers of road, sample is obtained
Point cloud data；
Point cloud data (point cloud data) refers to the data recorded in dots by modes such as scannings, each
A point includes three-dimensional location coordinates, some may contain colouring information (RGB) or Reflection intensity information (Intensity).
After the completion of the point cloud data acquisition of roadside, a cloud can be marked in advance in a manner that artificial or machine marks
The classification of data.For example, the point cloud data of the point cloud data of the point cloud data of road, pavement markers, bridge, road affiliated facility
The point cloud data of such as signal lamp.What the present invention was handled is then the point cloud data labeled as pavement markers, is had on a road
Several pavement markers, each pavement markers can correspond to some point cloud datas.
S102：Sample point cloud data is ranked up along the direction of the road, the sample point cloud data after being sorted；
S103：Along the direction of road, the sample point cloud data after sequence is divided into sample point by preset spacing distance
Cloud data block；
S104：Each sample point cloud data block is counted, obtains the distributed data of sample point cloud data；
S105：According to the distributed data of sample point cloud data, it is pre- whether the regularity of distribution of judgement sample point cloud data meets
If the regularity of distribution of the point cloud data of the pavement markers of characteristic, if so, the road surface for being identified as preset property by pavement markers
Label.
To further understand the application, scheme is described in further detail below.
The embodiment of the present invention makes full use of the point cloud distribution character of some pavement markers, is advised for the repeatability of point cloud data
Rule identifies the pavement markers of specific feature.
Technical solution provided in an embodiment of the present invention is particularly suitable for the traffic marking road marking for having point cloud characteristic to some
Note is identified.The so-called traffic marking pavement markers with point cloud characteristic, refer to that the point cloud data by multiple repetitions is formed
The graticule that graticule is marked or is made of the point cloud data of multiple regularity presentations marks.These label be usually horizontal stripe shape and
Repeat, such as slow down label, crossing etc..For example, slow down label be normally placed in charge station square, exit ramp or
Other require the section of vehicle deceleration, can be divided into transversal deceleration label and longitudinal deceleration label.It is transversal deceleration mark referring to Fig. 2
Remember schematic diagram.Transversal deceleration label is generally white dashed line, and form has single dotted line, doublet and three dotted lines, perpendicular to driving
Direction is set.Referring to Fig. 3, schematic diagram is marked for longitudinal deceleration.Longitudinal deceleration is parallel to driveway line of demarcation labeled as one group
Diamond block dotted line in the initial position of driveway longitudinal deceleration label, sets the transition of one section (such as 30 meters), diamond block dotted line
It is broadened by narrow, width broadens from narrow and (e.g., fades to 30 centimetres from 10 centimetres).As it can be seen that whether transversal deceleration label and longitudinal direction
Slow down label, is all the point for including repetition or regularity in point cloud data.Therefore, transversal deceleration label or longitudinal deceleration label
All it is the typical pavement markers with point cloud characteristic.
It should be noted that, although only there is point cloud characteristic as typical using transversal deceleration label and longitudinal deceleration label
Pavement markers illustrate, it will be appreciated by those skilled in the art that without being limited thereto with the pavement markers of cloud characteristic,
The pavement markers with point cloud characteristic including other unlisted existing or following appearance.
To preceding method, each step describes in detail below.
Point cloud is the magnanimity point set that object space distribution and target surface characteristic are expressed under the same space referential, such as
It is huge that fruit all carries out point cloud data processing workload, inefficiency, and the embodiment of the present invention can be by only choosing road marking
The part point cloud data of note, to reduce pending data amount, so as to improve efficiency.
In the specific implementation, it is needed by step S101 first along the lane line of road, from a road for having been labeled as road
In the point cloud data of face label, sample point cloud data is obtained.
Under normal circumstances, choosing pavement markers has the point cloud data of characteristic feature as sample point cloud data.It is for example, right
It is marked in transversal deceleration, since transversal deceleration label is generally made of the several sections of dotted lines perpendicular to direction of traffic setting, can
According to the distribution character of transversal deceleration label point cloud data, to choose the center line both sides for being located at road in transversal deceleration label
Point cloud data in the range of one fixed width is as sample point cloud data.Accordingly, along the lane line of road, from having been labeled as road
In the point cloud data of one pavement markers, obtain sample point cloud data and following specific implementation may be used：
Obtain the center line of the lane line of road；
From having been labeled as in the point cloud data of a pavement markers of the road, the center line of the lane line is got
Distance no more than preset distance threshold point cloud data as sample point cloud data.
Wherein, if the track sum of road is odd number, the lane line of the preferred middle lane of lane line, if the vehicle of road
Road sum is even number, and preferably lane number is track sum divided by the lane line in 2 track, for state of the operator seat in left side
Family, lane number are numbered from left to right；For country of the operator seat on right side, lane number is numbered from right to left.It can be seen that
The center line in the track that the present invention obtains can be as the center line of road.
Then, in step s 102, sample point cloud data is ranked up along the direction of the road, after being sorted
Sample point cloud data.
Specifically, with reference to figure 4 and Fig. 5, it can realize that sample point cloud data sorts along road direction by following step：
S401：An optional sample point cloud data is used as with reference to point cloud data from sample point cloud data；
As shown in figure 5, the point cloud data in rectangle frame is sample point cloud data, step S401 is therefrom used as ginseng for optional one
Examine point cloud data.
S402：A source location is chosen from the center line of the lane line of road, wherein, source location is located at sample
Except this point cloud data overlay area；
S403：Each sample point cloud data is obtained to the distance of source location；
S404：Along the direction of road, if source location is located at the rear of reference point clouds data, by sample point cloud data
It sorts according to the ascending sequence of the distance to source location, the sample point cloud data after being sorted；
S405：Along the direction of road, if source location is located at the front of reference point clouds data, by sample point cloud data
It sorts according to the descending sequence of the distance to source location, the sample point cloud data after being sorted.
As shown in figure 5, center line arrow represents the direction of road, source location is located at the rear of reference point clouds data,
Therefore, it is necessary to step S404 by sample point cloud data according to the ascending sequence sequence of the distance to source location.Conversely,
Then it is ranked up according to step 405.
Next, in step s 103, along the direction of road, the sample point cloud data after sequence is pressed into preset spacer
From being divided into sample point cloud data block.
If identification horizontal stripe shape and the pavement markers repeated, it is preferred that can be by the sample point cloud number after sequence
According to according to preset spacing distance, division obtains four pieces or more of sample point cloud data block.
In the concrete realization, sequential scan is carried out to the sample point cloud data after sequence using sliding window, obtains sample
This point cloud data block.The mode that sliding window is taken to scan, it is therefore an objective to window be controlled to carry out the sample point cloud data after sequence
Sequential scan, among these including two layers of meaning, first, scanning sequency is certain (in a certain direction), second is that, pass through sliding window
Mouth carries out the sample point cloud data after sequence repeatedly segmentation blocked scan.Equidirectional multiple sample point cloud numbers are obtained as a result,
According to block, so as to for subsequent statistical and judge whether each sample point cloud data block there is certain repeatability/regularity to provide basis.
Sliding window (Sliding window) is a kind of flow control technique.In the embodiment of the present invention, based on sliding window
" multithreading " principle of mouth sets the length and step-length of sliding window, in a certain direction sequentially to the sample after the sequence
Point cloud data is scanned according to step-length.After being scanned, that is, obtain sample point cloud data block.
Then, in step S104, each sample point cloud data block is counted, obtains the distribution of sample point cloud data
Data.
Specifically, the statistics of sample point cloud data block can be realized by following step：
The number of step 1041, each sample point cloud data sample point cloud data in the block of statistics, presets if number is more than
Number threshold value, then by sample point cloud data block labeled as there is data, otherwise, labeled as blank；
Step 1042 traverses the label result of each sample point cloud data block, when continuously having data markers i-th
Secondary when occurring, record ith has the number LOne that data markers continuously occuriWith when continuous blank marks jth time occurs,
The number LZero that record jth time blank marks continuously occurj；Wherein, LOnei、LZerojPoint for the sample point cloud data
Cloth data, i and j are the natural number more than or equal to 1.
It, can be to these sample point cloud data blocks simultaneously to improve treatment effeciency since sample point cloud data block has polylith
Row performs abovementioned steps 1041, i.e., similar " multithreading " principle.Specifically, it can set equal with sample point cloud data number of blocks
Several data buffer zones, for storing each sample point cloud data data in the block respectively, then parallel execution of steps 1041 is fast
Speed completes the statistics to sample point cloud data block.For example, selected sample point cloud data is marked to be swept for transversal deceleration
It retouches, if sample point cloud data is 2 meters along the zone length that the direction of road covers, meanwhile, sliding window length is set as 0.1
Rice, window step length are 0.1 meter, then, sliding window needs to slide 20 times (2/0.1=20) so as to complete the sample to 2 meters of length
The scanning of this point cloud data, so as to obtain 20 pieces of sample point cloud data blocks.Correspondingly, 20 data buffer zones of setting carry out data
Storage, specifically, the array that length is 20 can be set to complete depositing for all sample point cloud data blocks as data buffer zone
Storage.
The specific implementation of above-mentioned steps 104 is introduced in citing.Assuming that sample point cloud data block shares 10 pieces, it is suitable
Sequence number is 1-10, number statistical result see table 1, wherein, 0 represents no data, and 1 represents and has data.To number statistical result
Traverse obtain the result is that：
Lone1=4, Lone2=2, LZero1=2, LZero2=2.
| Number | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
| Number | 0 | 0 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 1 |
Finally, in step S105, according to the distributed intelligence of sample point cloud data, the distribution of judgement sample point cloud data is advised
Whether rule meets the regularity of distribution of the point cloud data of the pavement markers of preset property.
For example, the pavement markers of preset property are horizontal stripe shape and the pavement markers of repetition.For example, crossing, zebra stripes
Etc. belonging to this pavement markers.The present invention's focuses on identifying horizontal stripe shape and the pavement markers repeated.
Specifically, the identification of the pavement markers of preset property can be accomplished in the following manner：
Judge each LOnei+1And LZeroj+1Whether equation below is met：
OneLength*Lself-adaption1<LOnei+1<OneLength*Lself-adaption2；
ZeroLength*Lself-adaption1<LZeroj+1<ZeroLength*Lself-adaption2；
If it being satisfied by, it is determined that pavement markers are the pavement markers for meeting preset horizontal stripe shape and repeating, wherein,
OneLength=LOne1, ZeroLength=LZero1, Lself-adaption1, Lself-adaption2 be it is preset from
Adaptation parameter, Lself-adaption1<Lself-adaption2.
In practical applications, Lself-adaption1=0.5, Lself-adaption1=1.5, the citing are only preferred
Embodiment is not the limitation to auto-adaptive parameter value of the present invention, and those skilled in the art can be according to actual needs to adaptive
Parameter carries out value.
Continue to use aforementioned citing, OneLength=LOne1=4, ZeroLength=LZero1=2, at this point, Lone2=2 is small
In OneLength*3/2, equal to OneLength/2；LZero2=2 are less than ZeroLength*3/2, more than ZeroLength/2,
Therefore, pavement markers shown in citing are not belonging to horizontal stripe shape and the pavement markers repeated.
As it can be seen that the embodiment of the present invention makes full use of the point cloud distribution character of some pavement markers, for having been labeled as road surface
The point cloud data of label is ranked up and counts, and obtains the regularity of distribution of point cloud data, and passes through the regularity of distribution and determine whether
The pavement markers of preset property.The present invention program automatic identification without human intervention, so cost is very low, has very strong reality
The property used.The present invention makes full use of the point cloud distribution character of some pavement markers, can efficiently and accurately identify with a cloud
The pavement markers of characteristic.
Below with the identification of a specific transversal deceleration label, the embodiment of the present invention is introduced.
Referring to Fig. 6, three kinds of pavement markers schematic diagrames to be identified are shown, wherein, Fig. 6 (a) and Fig. 6 (b) they are two kinds of roads
Slow down label, and Fig. 6 (c) is other optional pavement markers.The example being described below is identified these three pavement markers and right
Compare recognition result.
Identification to transversal deceleration label, mainly including following five steps.
Step 1 chooses point among pavement markers in the range of one fixed width as sample point cloud data, and along road direction
Sample point cloud data is ranked up.
The feature of distribution is repeated by multistage dotted line in view of transversal deceleration label, is not needed to using whole pavement markers
Point, it is only necessary to intermediate one piece of data, as shown in figure 5, be transversal deceleration label sample point cloud area schematic, the example
In, the point among road in certain area is chosen as sample point cloud data.Assuming that track line width is W, point Pi is apart from right side
The distance of lane line is Di, then the point Pi for meeting the following conditions can be used as sample point cloud data：
|W/2-Di|<l
Wherein, l is pre-set distance threshold, for example, take l=0.2, i.e., | W/2-Di|<0.2, then sample point cloud number
It is less than 0.2, that is, the width in the data cover of sample point cloud shown in Fig. 5 region is 0.4 according to the distance of the lane line to road,
That is, the point cloud in each 0.2 meter of center line by way of road or so is as sample point cloud data.
Step 2 is ranked up sample point cloud data along road direction.
Filtering out sample point cloud data PiAfterwards, the click-through line density after screening can be clustered, removal noise processed, example
Such as clustered using DBSCAN methods；Then it sorts to the point after cluster so that put and arranged along road direction.Finally, it defines
Point set is combined into P={ P1,P2,…,Pn}。
Step 3 is scanned the sample point cloud data after sequence using sliding window, obtains sample point cloud data block.
The present invention uses the method statistic point cloud distributed intelligence of sliding window.Referring to Fig. 7, show that a sliding window shows
It is intended to, which is divided into 6 pieces, i.e., if data can be scanned, which can once acquire 6 samples
Point cloud data block.In this example embodiment, window is slided along vehicle heading, and window grows 0.1 meter, 0.1 meter of sliding step.Window is set
Mouth buffer size is Dist (P1,Pn) * 2+1, wherein Dist 2 points of distances in center line projection of expression.
Step 4, based on sample point cloud data block, the quantity information of statistics point cloud.
Traverse sample point cloud data block, to each point Pi, belonging to buffering area under be designated as * 10 roundings of Dist (Pi, P1).
After the completion of traversal, the point quantity of each buffering area is recorded.It is that the statistical information of three pavement markers in Fig. 6 is illustrated referring to Fig. 8
Figure.In Fig. 8, abscissa is buffering area subscript, and ordinate is point quantity.From Fig. 8 statistical results it is found that road marking in Fig. 6 (a)
Note is with complete data statistics as a result, and in the rule that repeats, the pavement markers in Fig. 6 (b) have approximately half
Repeatability or regularity are not presented for the rule that data statistics as a result, also present repeats, the pavement markers in Fig. 6 (c).
Step 5, analysis statistical result, judge whether to meet periodic regularity.
The dot density of transversal deceleration label meets periodic feature, but specific cycle parameter is unknown, and various forms of horizontal strokes
It is inconsistent to the cycle parameter for the label (form has single dotted line, doublet and three dotted lines) that slows down, therefore global ginseng cannot be set
It counts to judge.The cycle parameter of statistical information is estimated in the present invention using adaptive method, then carrying out judgement to entirety is
It is no to meet periodic regularity.
Binary conversion treatment is carried out to buffering area, point quantity is modified to 1 higher than a certain amount of (such as 5), the amendment less than or equal to 5
It is 0.Buffering area is traversed, the length that first is continuously 1 is recorded, is denoted as OneLength；Continuous record first is 0
Length, be denoted as ZeroLength.
Secondary traversal is carried out to buffering area, the continuous length for being 1 or being 0 is recorded every time, is denoted as LOne and LZero respectively,
For arbitrary LOnei+1And LZeroj+1If all meet：
OneLength*Lself-adaption1<LOnei+1<OneLength*Lself-adaption2
ZeroLength*Lself-adaption1<LZeroj+1<ZeroLength*Lself-adaption2
Then meet periodic feature, be identified as transversal deceleration type, be otherwise other types.
Wherein, Lself-adaption1, Lself-adaption2 are the auto-adaptive parameters set previously according to experience,
Value defines adaptive space, takes Lself-adaption1<Lself-adaption2.For example, Lself-adaption1=
0.5th, Lself-adaption1=1.5.
In this example, by above-mentioned judgement, recognition result is to determine that Fig. 6 (a) and Fig. 6 (b) are correctly judged as that horizontal line subtracts
Speed label；It determines that Fig. 6 (c) slows down for non-transverse and marks or export UNKNOWN TYPE.
The embodiment of the present application provides a kind of identification device of pavement markers corresponding with the recognition methods of above-mentioned pavement markers,
The identification device structure diagram of the pavement markers is illustrated in figure 9, which mainly includes such as lower unit：
Acquiring unit 901, for along the lane line of road, from the point cloud for a pavement markers for having been labeled as the road
In data, sample point cloud data is obtained；
Sequencing unit 902, for the sample point cloud data to be ranked up along the direction of the road, after obtaining sequence
Sample point cloud data；
Division unit 903, for along the direction of road, the sample point cloud data after sequence to be drawn by preset spacing distance
It is divided into sample point cloud data block；
Statistic unit 904 for being counted to each sample point cloud data block, obtains point of the sample point cloud data
Cloth data；
Recognition unit 905 for the distributed data according to the sample point cloud data, judges the sample point cloud data
Whether the regularity of distribution meets the regularity of distribution of the point cloud data of the pavement markers of preset property, if so, by the pavement markers
The pavement markers for being identified as the preset property.
Preferably, the acquiring unit 901 is specifically used for：Obtain the center line of the lane line of road；From having been labeled as
In the point cloud data of a pavement markers for stating road, get the lane line center line distance no more than it is preset away from
Point cloud data from threshold value is as sample point cloud data.
Preferably, the sequencing unit 902 is specifically used for：The optional sample point cloud number from the sample point cloud data
According to as with reference to point cloud data；A source location, the target position are chosen from the center line of the lane line of the road
It puts and is a little located at except the sample point cloud data cover region；Each sample point cloud data is obtained to the source location
Distance；Along the direction of the road, if the source location is located at the rear of reference point clouds data, by the sample point cloud
Data sort according to the ascending sequence of the distance to the source location, the sample point cloud data after being sorted；Edge
If the source location is located at the front of reference point clouds data, the sample point cloud data is pressed for the direction of the road
Shine the descending sequence sequence of the distance of source location, the sample point cloud data after being sorted.
Preferably, the statistic unit 904 is specifically used for：Count each sample point cloud data sample point cloud data in the block
Number, if number is more than preset number threshold value, by the sample point cloud data block labeled as there is data, otherwise, label
For blank；The label result of each sample point cloud data is traversed, when continuously thering is data markers ith to occur, note
Record ith has the number LOne that data markers continuously occuriWith when continuous blank marks jth time occurs, record jth time is empty
The number LZero that white marker continuously occursj；The LOnei、LZerojFor the distributed data of the sample point cloud data, i and j are
Natural number more than or equal to 1.
Preferably, the pavement markers of the preset property for horizontal stripe shape and repeat pavement markers, the recognition unit 905
It is specifically used for：Judge each LOnei+1And LZeroj+1Whether equation below is met：
OneLength*Lself-adaption1<LOnei+1<OneLength**Lself-adaption2；
ZeroLength*Lself-adaption1<LZeroj+1<ZeroLength**Lself-adaption2；
If it being satisfied by, it is determined that the pavement markers are the pavement markers for meeting preset horizontal stripe shape and repeating,
Wherein, OneLength=LOne1, ZeroLength=LZero1, Lself-adaption1, Lself-adaption2 are default
Auto-adaptive parameter, Lself-adaption1<Lself-adaption2.
It should be noted that the application can be carried out in the assembly of software and/or software and hardware, for example, can adopt
With application-specific integrated circuit (ASIC), general purpose computer or any other realized similar to hardware device.In one embodiment
In, the software program of the application can perform to realize steps described above or function by processor.Similarly, the application
Software program can be stored in computer readable recording medium storing program for performing (including relevant data structure), for example, RAM memory,
Magnetic or optical driver or floppy disc and similar devices.In addition, hardware can be used to realize in some steps or function of the application, example
Such as, as coordinating with processor so as to perform the circuit of each step or function.
In addition, the part of the application can be applied to computer program product, such as computer program instructions, when its quilt
When computer performs, by the operation of the computer, it can call or provide according to the present processes and/or technical solution.
And the program instruction of the present processes is called, be possibly stored in fixed or moveable recording medium and/or is passed through
Broadcast or the data flow in other signal loaded mediums and be transmitted and/or be stored according to described program instruction operation
In the working storage of computer equipment.Here, including a device according to one embodiment of the application, which includes using
Memory in storage computer program instructions and processor for executing program instructions, wherein, when the computer program refers to
When order is performed by the processor, method and/or skill of the device operation based on aforementioned multiple embodiments according to the application are triggered
Art scheme.
It is obvious to a person skilled in the art that the application is not limited to the details of above-mentioned exemplary embodiment, Er Qie
In the case of without departing substantially from spirit herein or essential characteristic, the application can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and scope of the present application is by appended power
Profit requirement rather than above description limit, it is intended that all by what is fallen within the meaning and scope of the equivalent requirements of the claims
Variation is included in the application.Any reference numeral in claim should not be considered as to the involved claim of limitation.This
Outside, it is clear that one word of " comprising " is not excluded for other units or step, and odd number is not excluded for plural number.That is stated in system claims is multiple
Unit or device can also be realized by a unit or device by software or hardware.The first, the second grade words are used for table
Show title, and do not represent any particular order.
Claims (10)
1. a kind of recognition methods of pavement markers, which is characterized in that the method includes：
Along the lane line of road, from having been labeled as in the point cloud data of a pavement markers of the road, sample point cloud is obtained
Data；
The sample point cloud data is ranked up along the direction of the road, the sample point cloud data after being sorted；
Along the direction of road, the sample point cloud data after sequence is divided into sample point cloud data block by preset spacing distance；
Each sample point cloud data block is counted, obtains the distributed data of the sample point cloud data；
According to the distributed data of the sample point cloud data, judge the sample point cloud data the regularity of distribution whether meet it is default
The regularity of distribution of the point cloud data of the pavement markers of characteristic, if so, the pavement markers are identified as the preset property
Pavement markers.
2. the method as described in claim 1, which is characterized in that the lane line along road, from having been labeled as the road
A pavement markers point cloud data in, obtain sample point cloud data specifically include：
Obtain the center line of the lane line of road；
From having been labeled as in the point cloud data of a pavement markers of the road, get the lane line center line away from
From the point cloud data no more than preset distance threshold as sample point cloud data.
3. method as claimed in claim 2, which is characterized in that described by direction of the sample point cloud data along the road
It is ranked up and specifically includes：
An optional sample point cloud data is used as with reference to point cloud data from the sample point cloud data；
A source location is chosen from the center line of the lane line of the road, the source location is located at the sample
Except point cloud data overlay area；
Each sample point cloud data is obtained to the distance of the source location；
Along the direction of the road, if the source location is located at the rear of reference point clouds data, by the sample point cloud
Data sort according to the ascending sequence of the distance to the source location, the sample point cloud data after being sorted；
Along the direction of the road, if the source location is located at the front of reference point clouds data, by the sample point cloud
Data sort according to the descending sequence of the distance to source location, the sample point cloud data after being sorted.
4. the method as described in any one of claim 1-3 claims, which is characterized in that described to each sample point cloud number
It is counted according to block, the distributed data for obtaining the sample point cloud data specifically includes：
The number of each sample point cloud data sample point cloud data in the block is counted, if number is more than preset number threshold value,
By the sample point cloud data block labeled as there is data, otherwise, labeled as blank；
The label result of each sample point cloud data is traversed, when it is continuous there is data markers ith to occur when, record the
There is the number LOne that data markers continuously occur i timesiWith when continuous blank marks jth time occurs, jth time blank mark is recorded
Remember the number LZero continuously occurredj；
The LOnei、LZerojFor the distributed data of the sample point cloud data, i and j are the natural number more than or equal to 1.
5. method as claimed in claim 4, which is characterized in that the pavement markers of the preset property are horizontal stripe shape and repeat
Pavement markers, the distributed intelligence according to the sample point cloud data, judging the regularity of distribution of the sample point cloud data is
The regularity of distribution of the point cloud data of the no pavement markers for meeting preset property specifically includes：
Judge each LOnei+1And LZeroj+1Whether equation below is met：
OneLength*Lself-adaption1<LOnei<OneLength*Lself-adaption2；
ZeroLength*Lself-adaption1<LZeroi<ZeroLength*Lself-adaption2；
If it being satisfied by, it is determined that the pavement markers are the pavement markers for meeting preset horizontal stripe shape and repeating, wherein,
OneLength=LOne1, ZeroLength=LZero1；Lself-adaption1, Lself-adaption2 be it is preset from
Adaptation parameter, Lself-adaption1<Lself-adaption2.
6. a kind of identification device of pavement markers, which is characterized in that described device includes：
Acquiring unit, for along the lane line of road, from having been labeled as in the point cloud data of a pavement markers of the road,
Obtain sample point cloud data；
Sequencing unit, for the sample point cloud data to be ranked up along the direction of the road, the sample after being sorted
Point cloud data；
Division unit, for along the direction of road, the sample point cloud data after sequence to be divided into sample by preset spacing distance
This point cloud data block；
Statistic unit for being counted to each sample point cloud data block, obtains the distributed data of the sample point cloud data；
Recognition unit for the distributed data according to the sample point cloud data, judges the distribution rule of the sample point cloud data
Whether rule meets the regularity of distribution of the point cloud data of the pavement markers of preset property, if so, the identification by the pavement markers
Pavement markers for the preset property.
7. device as claimed in claim 6, which is characterized in that the acquiring unit is specifically used for：Obtain the lane line of road
Center line；From having been labeled as in the point cloud data of a pavement markers of the road, the center of the lane line is got
The distance of line is not more than the point cloud data of preset distance threshold as sample point cloud data.
8. device as claimed in claim 7, which is characterized in that the sequencing unit is specifically used for：From the sample point cloud number
An optional sample point cloud data is used as with reference to point cloud data in；One is chosen from the center line of the lane line of the road
Source location, the source location are located at except the sample point cloud data cover region；Obtain each sample point cloud
Data are to the distance of the source location；Along the direction of the road, if the source location is located at reference point clouds data
Rear, then the sample point cloud data according to the ascending sequence of the distance to the source location is sorted, obtained
Sample point cloud data after sequence；Along the direction of the road, if the source location is located at the front of reference point clouds data,
Then the sample point cloud data is sorted according to the descending sequence of the distance to source location, the sample after being sorted
Point cloud data.
9. the device as described in any one of claim 6-8 claims, which is characterized in that the statistic unit is specifically used
In：The number of each sample point cloud data sample point cloud data in the block is counted, it, will if number is more than preset number threshold value
The sample point cloud data block is labeled as there is data, otherwise, labeled as blank；To the label result of each sample point cloud data into
Row traversal, when continuously having data markers ith to occur, record ith has the number LOne that data markers continuously occuriWith
When continuous blank marks jth time occurs, the number LZero that jth time blank marks continuously occur is recordedj；The LOnei、
LZerojFor the distributed data of the sample point cloud data, i and j are the natural number more than or equal to 1.
10. device as claimed in claim 9, which is characterized in that the pavement markers of the preset property are horizontal stripe shape and repetition
Pavement markers, the recognition unit is specifically used for：
Judge each LOnei+1And LZeroj+1Whether equation below is met：
OneLength*Lself-adaption1<LOnei<OneLength*Lself-adaption2
ZeroLength*Lself-adaption1<LZeroi<ZeroLength*Lself-adaption2
If it being satisfied by, it is determined that the pavement markers are the pavement markers for meeting preset horizontal stripe shape and repeating, wherein,
OneLength=LOne1, ZeroLength=LZero1；Lself-adaption1, Lself-adaption2 be it is preset from
Adaptation parameter, Lself-adaption1<Lself-adaption2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611201463.XA CN108241819B (en) | 2016-12-23 | 2016-12-23 | Method and device for identifying pavement marker |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611201463.XA CN108241819B (en) | 2016-12-23 | 2016-12-23 | Method and device for identifying pavement marker |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108241819A true CN108241819A (en) | 2018-07-03 |
| CN108241819B CN108241819B (en) | 2020-07-28 |
Family
ID=62703214
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611201463.XA Active CN108241819B (en) | 2016-12-23 | 2016-12-23 | Method and device for identifying pavement marker |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108241819B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109343021A (en) * | 2018-12-03 | 2019-02-15 | 北京遥感设备研究所 | A kind of Plot coherence method |
| CN110458083A (en) * | 2019-08-05 | 2019-11-15 | 武汉中海庭数据技术有限公司 | A kind of lane line vectorization method, device and storage medium |
| CN111174777A (en) * | 2018-11-09 | 2020-05-19 | 阿里巴巴集团控股有限公司 | Positioning method and device and electronic equipment |
| CN112069899A (en) * | 2020-08-05 | 2020-12-11 | 深兰科技（上海）有限公司 | Road shoulder detection method and device and storage medium |
| CN112825196A (en) * | 2019-11-20 | 2021-05-21 | 阿里巴巴集团控股有限公司 | Road marker determining method and device, storage medium and editing platform |
| CN113570004A (en) * | 2021-09-24 | 2021-10-29 | 西南交通大学 | Riding hot spot area prediction method, device, equipment and readable storage medium |
| JP2022522385A (en) * | 2020-02-07 | 2022-04-19 | シェンチェン センスタイム テクノロジー カンパニー リミテッド | Road sign recognition methods, map generation methods, and related products |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103390169A (en) * | 2013-07-19 | 2013-11-13 | 武汉大学 | Sorting method of vehicle-mounted laser scanning point cloud data of urban ground objects |
| CN104197897A (en) * | 2014-04-25 | 2014-12-10 | 厦门大学 | Urban road marker automatic sorting method based on vehicle-mounted laser scanning point cloud |
| CN105069395A (en) * | 2015-05-17 | 2015-11-18 | 北京工业大学 | Road marking automatic identification method based on terrestrial three-dimensional laser scanning technology |
| CN105551082A (en) * | 2015-12-02 | 2016-05-04 | 百度在线网络技术（北京）有限公司 | Method and device of pavement identification on the basis of laser-point cloud |
-
2016
- 2016-12-23 CN CN201611201463.XA patent/CN108241819B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103390169A (en) * | 2013-07-19 | 2013-11-13 | 武汉大学 | Sorting method of vehicle-mounted laser scanning point cloud data of urban ground objects |
| CN104197897A (en) * | 2014-04-25 | 2014-12-10 | 厦门大学 | Urban road marker automatic sorting method based on vehicle-mounted laser scanning point cloud |
| CN105069395A (en) * | 2015-05-17 | 2015-11-18 | 北京工业大学 | Road marking automatic identification method based on terrestrial three-dimensional laser scanning technology |
| CN105551082A (en) * | 2015-12-02 | 2016-05-04 | 百度在线网络技术（北京）有限公司 | Method and device of pavement identification on the basis of laser-point cloud |
Non-Patent Citations (1)
| Title |
|---|
| TAKASHI OGAWA等: "Lane Recognition Using On-vehicle LIDAR", 《2006 IEEE INTELLIGENT VEHICLES SYMPOSIUM》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111174777A (en) * | 2018-11-09 | 2020-05-19 | 阿里巴巴集团控股有限公司 | Positioning method and device and electronic equipment |
| CN109343021A (en) * | 2018-12-03 | 2019-02-15 | 北京遥感设备研究所 | A kind of Plot coherence method |
| CN110458083A (en) * | 2019-08-05 | 2019-11-15 | 武汉中海庭数据技术有限公司 | A kind of lane line vectorization method, device and storage medium |
| CN112825196A (en) * | 2019-11-20 | 2021-05-21 | 阿里巴巴集团控股有限公司 | Road marker determining method and device, storage medium and editing platform |
| CN112825196B (en) * | 2019-11-20 | 2023-04-25 | 阿里巴巴集团控股有限公司 | Method and device for determining road sign, storage medium and editing platform |
| JP2022522385A (en) * | 2020-02-07 | 2022-04-19 | シェンチェン センスタイム テクノロジー カンパニー リミテッド | Road sign recognition methods, map generation methods, and related products |
| CN112069899A (en) * | 2020-08-05 | 2020-12-11 | 深兰科技（上海）有限公司 | Road shoulder detection method and device and storage medium |
| CN113570004A (en) * | 2021-09-24 | 2021-10-29 | 西南交通大学 | Riding hot spot area prediction method, device, equipment and readable storage medium |
| CN113570004B (en) * | 2021-09-24 | 2022-01-07 | 西南交通大学 | Riding hot spot area prediction method, device, equipment and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108241819B (en) | 2020-07-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108241819A (en) | The recognition methods of pavement markers and device | |
| CN111598030B (en) | Method and system for detecting and segmenting vehicle in aerial image | |
| CN104217427B (en) | Lane line localization method in a kind of Traffic Surveillance Video | |
| CN109711410A (en) | Three-dimensional object rapid segmentation and identification method, device and system | |
| CN108830171B (en) | Intelligent logistics warehouse guide line visual detection method based on deep learning | |
| CN110264468A (en) | Point cloud data mark, parted pattern determination, object detection method and relevant device | |
| CN112016605B (en) | Target detection method based on corner alignment and boundary matching of bounding box | |
| CN109740420A (en) | Vehicle illegal recognition methods and Related product | |
| CN104299260A (en) | Contact network three-dimensional reconstruction method based on SIFT and LBP point cloud registration | |
| CN105574063A (en) | Image retrieval method based on visual saliency | |
| CN106951830A (en) | A kind of many object marking methods of image scene constrained based on priori conditions | |
| CN103577875A (en) | CAD (computer-aided design) people counting method based on FAST (features from accelerated segment test) | |
| CN105574527A (en) | Quick object detection method based on local feature learning | |
| CN104282026A (en) | Distribution uniformity assessment method based on watershed algorithm and minimum spanning tree | |
| CN108399424A (en) | A kind of point cloud classifications method, intelligent terminal and storage medium | |
| CN112132014A (en) | Target re-identification method and system based on non-supervised pyramid similarity learning | |
| CN106650811A (en) | Hyperspectral mixed pixel classification method based on neighbor cooperation enhancement | |
| Singer et al. | Dales objects: A large scale benchmark dataset for instance segmentation in aerial lidar | |
| CN105574545A (en) | Environment image multi-view-angle meaning cutting method and device | |
| CN114372503A (en) | Cluster vehicle motion trail prediction method | |
| EP4174792A1 (en) | Method for scene understanding and semantic analysis of objects | |
| CN110335322A (en) | Roads recognition method and road Identification device based on image | |
| CN1252588C (en) | High spectrum remote sensing image combined weighting random sorting method | |
| CN101887522A (en) | Method for extracting remote sensing image road based on multipoint simultaneous and rapid advance | |
| CN109740554A (en) | A kind of road edge line recognition methods and system |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200506 Address after: 310052 room 508, floor 5, building 4, No. 699, Wangshang Road, Changhe street, Binjiang District, Hangzhou City, Zhejiang Province Applicant after: Alibaba (China) Co.,Ltd. Address before: 102200, No. 18, No., Changsheng Road, Changping District science and Technology Park, Beijing, China. 1-5 Applicant before: AUTONAVI SOFTWARE Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230511 Address after: 102299 floor 1-5, block B1, 18 Changsheng Road, science and Technology Park, Changping District, Beijing Patentee after: AUTONAVI SOFTWARE Co.,Ltd. Address before: 310052 room 508, 5th floor, building 4, No. 699 Wangshang Road, Changhe street, Binjiang District, Hangzhou City, Zhejiang Province Patentee before: Alibaba (China) Co.,Ltd. |