CN109359614A - A kind of plane recognition methods, device, equipment and the medium of laser point cloud - Google Patents
A kind of plane recognition methods, device, equipment and the medium of laser point cloud Download PDFInfo
- Publication number
- CN109359614A CN109359614A CN201811272869.6A CN201811272869A CN109359614A CN 109359614 A CN109359614 A CN 109359614A CN 201811272869 A CN201811272869 A CN 201811272869A CN 109359614 A CN109359614 A CN 109359614A
- Authority
- CN
- China
- Prior art keywords
- plane
- grid
- point cloud
- cloud data
- threshold value
- 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
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
- G01S17/93—Lidar systems specially adapted for specific applications for anti-collision purposes
- G01S17/931—Lidar systems specially adapted for specific applications for anti-collision purposes of land vehicles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/48—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
- G01S7/4802—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Electromagnetism (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The embodiment of the invention discloses plane recognition methods, device, equipment and the media of a kind of laser point cloud.Wherein, method includes: to obtain the collected frame laser point cloud data of vehicle collector；The laser point cloud data is subjected to grid division；For each grid, plane identification is carried out using plane recognizer and set point number threshold value, wherein set point number threshold value used by each grid is different according to the distance between grid and acquisition center difference；It is handled according to plane recognition result.The embodiment of the present invention, which solves, not to be treated according to the differentiation for carrying out plane identification the characteristics of laser point cloud data, the problem for causing the accuracy of recognition result lower, realize according to the characteristics of laser point cloud plane identification in dynamic given threshold, can be improved plane identification accuracy.
Description
Technical field
The present embodiments relate to the plane recognition methods of data processing technique more particularly to a kind of laser point cloud, device,
Equipment and medium.
Background technique
In the technical field that automatic Pilot and auxiliary drive, carrying out environment sensing using laser radar is a kind of quilt extensively
The technology of use.
In the prior art, laser beam is emitted by laser radar first, laser beam encounters not transparent barrier
When can be reflected back, by laser radar receiving device receive；By record transmitting receive between time delay, in conjunction with penetrating
The transmission speed of line calculates the distance between barrier and laser radar.It is detected, is able to carry out by a large amount of laser beam
The perception of ambient enviroment.Per second to can produce 1,300,000 points by taking 64 line laser radars as an example, each point can reflect corresponding distance.
The point cloud data that the one-time detection of laser radar obtains, can be described as a frame laser point cloud data.
After obtaining laser point cloud data, further progress data processing is needed, is carried out based on laser point cloud data all kinds of
Obstacle recognition, such as cart, trolley, pedestrian, thus the foundation as control vehicle driving.Carry out the advance of obstacle recognition
The identification of row plane is based process operation, and still, the characteristics of prior art is not bound with laser point cloud carries out the difference of plane identification
Alienation is treated, and causes the accuracy of recognition result lower.
Summary of the invention
The embodiment of the present invention provides plane recognition methods, device, equipment and the medium of a kind of laser point cloud, to combine laser
The characteristics of point cloud, carries out plane identification, improves the accuracy of identification.
In a first aspect, the embodiment of the invention provides a kind of plane recognition methods of laser point cloud, this method comprises:
Obtain the collected frame laser point cloud data of vehicle collector；
The laser point cloud data is subjected to grid division；
For each grid, plane identification is carried out using plane recognizer and set point number threshold value, wherein each grid
Used set point number threshold value according to the distance between grid and acquisition center difference and it is different；
It is handled according to plane recognition result.
Second aspect, the embodiment of the invention also provides a kind of plane identification device of laser point cloud, which includes:
Point cloud data obtains module, for obtaining the collected frame laser point cloud data of vehicle collector；
Grid division module, for the laser point cloud data to be carried out grid division；
Plane identification module carries out plane using plane recognizer and set point number threshold value for being directed to each grid
Identification, wherein set point number threshold value used by each grid is different according to the distance between grid and acquisition center difference；
Processing module, for being handled according to plane recognition result.
The third aspect, the embodiment of the invention also provides a kind of computer equipment, which includes:
One or more processors；
Storage device, for storing one or more programs；
When one or more of programs are executed by one or more of processors, so that one or more of processing
Device realizes the plane recognition methods of any laser point cloud in the embodiment of the present invention.
Fourth aspect, the embodiment of the invention also provides a kind of computer readable storage mediums, are stored thereon with computer
Program, which is characterized in that the flat of the laser point cloud as described in any in inventive embodiments is realized when the program is executed by processor
Face recognition methods.
The embodiment of the present invention is by obtaining laser point cloud data and the laser point cloud data being carried out grid division, in turn
In plane identification, for each grid, plane identification is carried out using plane recognizer and set point number threshold value, wherein each
Set point number threshold value used by a grid is different, and is then handled according to plane recognition result；Solves no root
It is treated according to the differentiation for carrying out plane identification the characteristics of laser point cloud data, the problem for causing the accuracy of recognition result lower,
Realize according to the characteristics of laser point cloud plane identification in dynamic given threshold, can be improved plane identification accuracy.
Detailed description of the invention
Fig. 1 is the flow chart of the plane recognition methods of the laser point cloud in the embodiment of the present invention one；
Fig. 2 a is the flow chart of the plane recognition methods of the laser point cloud in the embodiment of the present invention two；
Fig. 2 b is the effect picture that ground is filtered out in laser point cloud data in the embodiment of the present invention two；
Fig. 3 is the structural schematic diagram of the plane identification device of the laser point cloud in the embodiment of the present invention three；
Fig. 4 is the structural schematic diagram of the computer equipment in the embodiment of the present invention four.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention rather than limiting the invention.It also should be noted that in order to just
Only the parts related to the present invention are shown in description, attached drawing rather than entire infrastructure.
Embodiment one
Fig. 1 is the flow chart of the plane recognition methods for the laser point cloud that the embodiment of the present invention one provides, and the present embodiment can fit
The case where for being identified by laser point cloud data plane, this method can be realized by the plane identification device of laser point cloud, tool
Body can be implemented by the software and/or hardware in equipment, which can integrate in any progress laser point cloud data processing
Equipment in, optionally onboard control device.As shown in Figure 1, the plane recognition methods of laser point cloud specifically includes:
S110, the collected frame laser point cloud data of vehicle collector is obtained.
Wherein, laser point cloud data, which can be, carries Laser Scanning Equipment by vehicle (such as pilotless automobile), to be measured
The laser point cloud data acquired on amount road.Laser Scanning Equipment is generally positioned at body exterior, such as vehicle roof.Laser is swept
The point cloud data that the one-time detection of equipment obtains is retouched, can be described as a frame laser point cloud data, it can the optional wherein laser of a frame
Point cloud data carries out the analysis and identification of point cloud data.
S120, the laser point cloud data is subjected to grid division.
The scanning devices such as laser radar scanning road surface point be not completely flat, and be it is discontinuous, therefore, by laser
Point cloud data is divided into grid, and the size of each grid is identical.In one embodiment, it can be carried out according to Laser Scanning Equipment
The distance of scanning can set grid in the cube that length is 8 meters.The characteristics of due to laser point cloud itself, with vehicle it is
The heart, from closely gradually becoming sparse to remote point cloud data point.That is, putting cloud point quantity in the grid closer apart from laser light source
It is more.
S130, it is directed to each grid, plane identification is carried out using plane recognizer and set point number threshold value, wherein each
Set point number threshold value used by a grid according to the distance between grid and acquisition center difference and it is different.
Specifically, being directed to each grid, plane recognizer can be selected to carry out plane identification, in identification process, led to
Often need the limit value using set point number threshold value, as identification plane.Due to the degree of rarefication script of the data point in each grid
Difference, so set point number threshold value difference used by each grid is arranged in the embodiment of the present invention, according to grid and acquisition center
The distance between and change, to adapt to the sparse situation of data point in each grid, the accuracy for identifying plane is higher.
It is illustrated by taking RANSAC algorithm as an example.The identification of plane iteration is carried out using RANSAC algorithm, until the grid
In do not belong to plane remainder strong point quantity less than set point number threshold value when, stop plane identification.
RANSAC (Random Sample Consensus) stochastical sampling coherence method is by containing " office from one group
In the sample set of exterior point ", estimate mathematical model parameter using iterative manner (this paper model is space plane model).Specific mistake
Journey is to provide the mathematical model for needing to estimate first, and secondly random selection goes out a sample set from observation data sample, is led to
The model parameter that minimum variance estimate method calculates sample set is crossed, the side of all observation data samples and the model is then calculated
Difference, according to outside the sample point (inliers) and model in pre-set threshold value and variance statistic observation data in model
Point (outliers) not in the know, this process is repeated according to the number of iterations.Terminate in each iteration, according to current iteration number,
Model parameter, expected error rate and total sample number calculate the factor for judging that current iteration terminates, and decide whether to terminate iteration
Process.At the end of iterative process, model parameter is optimal estimation parameter value.
In the present embodiment, the end factor of current iteration is to be directed to different grids, set set point number threshold value.
S140, it is handled according to plane recognition result.
After carrying out plane identification, it can get several planes, including building surface, surface of vehicle, lawn and road
The planes such as face.Wherein, road surface data probably account for a quarter of total amount of data, and data processing is increased in obstacle recognition
Burden.It therefore, can be on the basis of plane identifies, from laser point cloud based on the subsequent demand identified to barrier etc.
The corresponding point cloud data in road surface is filtered out in data.
The technical solution of the present embodiment is drawn by obtaining laser point cloud data and the laser point cloud data being carried out grid
Point, and then in plane identification, for each grid, using RANSAC stochastical sampling consistency plane recognizer, and it is directed to
Different set point number threshold values, which is arranged, in different grids terminates the factor as iteration；It solves not according to laser point cloud data
The differentiation that feature carries out plane identification is treated, the problem for causing the accuracy of recognition result lower, is realized according to laser point cloud
The characteristics of data the dynamic given threshold in plane identification, can be improved the accuracy and road surface dividing method of plane identification
Robustness after dividing by road surface, can also improve the treatment effeciency of subsequent obstacle recognition.
Embodiment two
Fig. 2 a is a kind of flow chart of the plane recognition methods for laser point cloud that inventive embodiments two provide.The present embodiment with
In above-described embodiment based on each optinal plan, provides according to plane recognition result, filtered out from laser point cloud data
The process of the corresponding point cloud data in road surface.As shown in Figure 2 a, the plane identification side of the laser point cloud provided in the embodiment of the present invention
Method includes the following steps:
S210, the collected frame laser point cloud data of vehicle collector is obtained.
S220, the laser point cloud data is subjected to grid division.
S230, the quantity for obtaining data point in each grid, by the quantity multiplied by setting ratio, as the grid
Set point number threshold value.
The characteristics of due to laser point cloud itself, centered on vehicle collector, from closely gradually becoming dilute to remote point cloud point
It dredges, therefore, the set point number threshold value for carrying out plane identification setting for each grid can be determined according to grid positions.When at one
In grid, when being not belonging to the points of any one plane less than or equal to set point number threshold value, it can terminate in the grid to flat
The identification in face.
In one embodiment, it is assumed that the point cloud point quantity in grid is N (N is positive integer), can take the 10% of N to be
The set point number threshold value of grid.Wherein, 10% is setting ratio, and setting ratio can be according to actual laser point cloud data amount
It is set, the set point number threshold value and the point cloud point quantity proportion relation in grid for making each grid.
In another embodiment, it can be closed according to the distance between each grid and acquisition center based on setting inverse ratio
System determines the set point number threshold value of each grid.For example, when grid is 10 meters at a distance from acquisition center, set point number
Threshold value is 100；When grid is 20 meters at a distance from acquisition center, set point number threshold value is 90；When grid and acquire center
When distance is 20 meters, set point number threshold value is 80.Wherein, acquisition center is the position where vehicle collector.
S240, it is directed to each grid, plane identification is carried out using plane recognizer and set point number threshold value, wherein each
Set point number threshold value used by a grid according to the distance between grid and acquisition center difference and it is different.
Specifically, extracting plane using RANSAC method, iterative extraction plane remains according to the dynamic threshold of each grid
When remaining points are less than dynamic threshold, stop extracting plane.Such as after extracting plane it is remaining points be less than N*10% when, stop
Iteration, extracting plane terminates.
S250, according to plane recognition result, the corresponding point cloud data in road surface is filtered out from laser point cloud data.
The plane for belonging to road surface is determined in the plane identified, can pass through the normal vector of plane and the folder of vertical direction
Two aspects in angle and height are judged.
Specifically, road surface is usually horizontal plane or for the inclined-plane with certain angle, the folder of normal vector and vertical direction
Angle is equal or close to 90 degree, if the angle between the normal vector and vertical direction of any plane is less than setting angle threshold value,
Determine that the plane is non-road surface.
Usual Laser Scanning Equipment is placed on the top of laser point cloud data acquisition vehicle, and Laser Scanning Equipment is adopted for data
Vertical range between collection center, i.e. data acquisition center and road surface is the height that laser point cloud data acquires vehicle, if appointed
One plane and height threshold mismatch, it is determined that the plane is non-road surface, wherein the height threshold is according to the height of the vehicle
The average height setting of point cloud data point in degree and grid.
And then target road surface can be determined in conjunction with the road surface recognition result of angle threshold value and height threshold, and from laser point
The corresponding point cloud data in target road surface is filtered out in cloud data.What needs to be explained here is that not having to the judgement of above-mentioned two aspect
There is fixed sequencing.
Further, if any plane and height threshold mismatch, it is determined that the plane is that non-road surface includes:
According to the point cloud data in each grid, the average height value of each data point in grid is counted；According to known vehicle
Height value and the average height value determine that the dynamic height threshold value of each grid specifically calculates the known vehicle height
The ratio reduced value of value and the ratio reduced value of the average height value take the maximum value of the two, and the dynamic as the grid is high
Spend threshold value, wherein the conversion ratio of known vehicle height value and the conversion ratio of average height value can customize；For each
Grid, the plan range value of computation grid central point to each plane, if the plan range value is greater than the dynamic height threshold
Value, it is determined that the plane is non-road surface.
In one embodiment, the process screened to plane with specific reference to level is as follows: firstly, according to every
Point cloud data in a grid, the interior average height characteristic Z put of statistics grid, it is known that height of car H, then, settable plane
Dynamic height threshold value be max (Z/2, H/5), wherein 2 be average height value conversion ratio, 5 be vehicle height value conversion
Ratio.For each plane, computation grid central point to the space plane distance feature D extracted, if D is greater than dynamic height threshold
The space plane point that value is then extracted is not ground point.It excludes to be known as the point of roof facial planes, greenbelt top planes with this
It Wei not ground point.
In the present embodiment, the effect that ground is filtered out in laser point cloud data can refer to Fig. 2 b, in figure 2b, white point
Or the collection of point is combined into the point cloud point of non-road surface point, after deleting road surface point cloud point, it can be clearly seen that obstacle object point cloud and road surface
Between there are two lines of demarcation, two lines of demarcation intermediate region is road surface, there are also the obstacle object point cloud point of small volume on road surface,
In road surface both sides, obstacle object point cloud point is more concentrated, and barrier can be building, number etc..
The technical solution of the present embodiment is drawn by obtaining laser point cloud data and the laser point cloud data being carried out grid
Point, the quantity of data point in each grid is obtained, the set point number threshold by the quantity multiplied by setting ratio, as the grid
Value, or the set point of each grid can be determined based on setting inverse relation according to the distance between each grid and acquisition center
Number threshold value, and then in plane identification, for each grid, plane knowledge is carried out using plane recognizer and set point number threshold value
Not, the corresponding point cloud data in road surface is then filtered out from laser point cloud data；It solves not according to laser point cloud data
The differentiation that feature carries out plane identification is treated, the problem for causing the accuracy of recognition result lower, is realized according to laser point cloud
The characteristics of data the dynamic given threshold in plane identification, can be improved the accuracy and road surface dividing method of plane identification
Robustness after dividing by road surface, can also improve the treatment effeciency of subsequent obstacle recognition.
Embodiment three
Fig. 3 shows a kind of structural schematic diagram of the plane identification device of laser point cloud of the offer of the embodiment of the present disclosure three,
The embodiment of the present disclosure is applicable to the case where progress plane identification of the characteristics of based on laser point cloud data.
As shown in figure 3, in the embodiment of the present disclosure laser point cloud plane identification device, comprising: point cloud data obtain module
310, grid division module 320, plane identification module 330 and processing module 340.
Wherein, point cloud data obtains module 310, for obtaining the collected frame laser point cloud data of vehicle collector；
Grid division module 320, for the laser point cloud data to be carried out grid division；Plane identification module 330, for for every
A grid carries out plane identification using plane recognizer and set point number threshold value, wherein set point used by each grid
Number threshold values are different according to the distance between grid and acquisition center difference；Processing module 340, for according to plane recognition result
It is handled.
The technical solution of the present embodiment is drawn by obtaining laser point cloud data and the laser point cloud data being carried out grid
Point, and then in plane identification, for each grid, plane identification is carried out using plane recognizer and set point number threshold value,
Wherein, set point number threshold value used by each grid is different, and is then handled according to plane recognition result；It solves
It is not treated according to the differentiation for carrying out plane identification the characteristics of laser point cloud data, causes the accuracy of recognition result lower
Problem, realize according to the characteristics of laser point cloud plane identification in dynamic given threshold, can be improved plane identification accuracy
And the robustness of road surface dividing method, after dividing by road surface, the treatment effeciency of subsequent obstacle recognition can also be improved.
Further, processing module 340 is specifically used for:
According to plane recognition result, the corresponding point cloud data in road surface is filtered out from laser point cloud data.
Optionally, plane identification module 330 is specifically used for:
For each grid, the identification of plane iteration is carried out using RANSAC algorithm, until not belonging in the grid flat
When the quantity at the remainder strong point in face is less than the set point number threshold value, stop plane identification.
Further, the plane identification device of laser point cloud further includes points threshold setting module, for for each
Grid obtains the number of data point in each grid before carrying out plane identification using plane recognizer and set point number threshold value
Amount, the set point number threshold value by the quantity multiplied by setting ratio, as the grid；Or
According to the distance between each grid and acquisition center, setting for each grid is determined based on setting inverse relation
Fixed-point number threshold value.
Optionally, processing module 340 includes that the first non-road surface recognition unit, the second non-road surface recognition unit and processing are single
Member；Wherein,
First non-road surface recognition unit, for being less than setting when the angle between the normal vector and vertical direction of any plane
When angle threshold value, it is determined that the plane is non-road surface；
Second non-road surface recognition unit, for being mismatched when any plane and height threshold, it is determined that the plane is non-road
Face, wherein the height threshold is set according to the average height of point cloud data point in the height and grid of the vehicle；
Processing unit determines target road surface for combining the road surface recognition result of angle threshold value and height threshold, and from swashing
The corresponding point cloud data in target road surface is filtered out in light point cloud data.
Further, the second non-road surface recognition unit, comprising: average height counts subelement, dynamic height threshold value determines
Subelement and non-road surface identify subelement；Wherein,
Average height counts subelement, for counting each data point in grid according to the point cloud data in each grid
Average height value；
Dynamic height threshold value determines subelement, each for being determined according to known vehicle height value and the average height value
The dynamic height threshold value of grid；
Non- road surface identifies subelement, is used for for each grid, the plan range value of computation grid central point to each plane,
If the plan range value is greater than the dynamic height threshold value, it is determined that the plane is non-road surface.
Further, dynamic height threshold value determines that subelement is specifically used for calculating the ratio folding of the known vehicle height value
The ratio reduced value of calculation value and the average height value takes the maximum value of the two, the dynamic height threshold value as the grid.
The plane identification device of laser point cloud provided by the embodiment of the present invention can be performed any embodiment of that present invention and be mentioned
The plane recognition methods of the laser point cloud of confession has the corresponding functional module of execution method and beneficial effect.
Example IV
Fig. 4 is the structural schematic diagram of the computer equipment in the embodiment of the present invention four.Fig. 4, which is shown, to be suitable for being used to realizing this
The block diagram of the exemplary computer device 412 of invention embodiment.The computer equipment 412 that Fig. 4 is shown is only an example,
Should not function to the embodiment of the present invention and use scope bring any restrictions.The computer equipment 412 is preferably vehicle-mounted control
Equipment.
As shown in figure 4, computer equipment 412 is showed in the form of universal computing device.The component of computer equipment 412 can
To include but is not limited to: one or more processor or processing unit 416, system storage 428 connect not homologous ray group
The bus 418 of part (including system storage 428 and processing unit 416).
Bus 418 indicates one of a few class bus structures or a variety of, including memory bus or Memory Controller,
Peripheral bus, graphics acceleration port, processor or the local bus using any bus structures in a variety of bus structures.It lifts
For example, these architectures include but is not limited to industry standard architecture (ISA) bus, microchannel architecture (MAC)
Bus, enhanced isa bus, Video Electronics Standards Association (VESA) local bus and peripheral component interconnection (PCI) bus.
Computer equipment 412 typically comprises a variety of computer system readable media.These media can be it is any can
The usable medium accessed by computer equipment 412, including volatile and non-volatile media, moveable and immovable Jie
Matter.
System storage 428 may include the computer system readable media of form of volatile memory, such as deposit at random
Access to memory (RAM) 430 and/or cache memory 432.Computer equipment 412 may further include it is other it is removable/
Immovable, volatile/non-volatile computer system storage medium.Only as an example, storage system 434 can be used for reading
Write immovable, non-volatile magnetic media (Fig. 4 do not show, commonly referred to as " hard disk drive ").Although not shown in fig 4,
The disc driver for reading and writing to removable non-volatile magnetic disk (such as " floppy disk ") can be provided, and non-easy to moving
The CD drive that the property lost CD (such as CD-ROM, DVD-ROM or other optical mediums) is read and write.In these cases, each
Driver can be connected by one or more data media interfaces with bus 418.Memory 428 may include at least one
Program product, the program product have one group of (for example, at least one) program module, these program modules are configured to perform this
Invent the function of each embodiment.
Program/utility 440 with one group of (at least one) program module 442, can store in such as memory
In 428, such program module 442 includes but is not limited to operating system, one or more application program, other program modules
And program data, it may include the realization of network environment in each of these examples or certain combination.Program module 442
Usually execute the function and/or method in embodiment described in the invention.
Computer equipment 412 can also be with one or more external equipments 414 (such as keyboard, sensing equipment, display
424 etc.) it communicates, the equipment interacted with the computer equipment 412 communication can be also enabled a user to one or more, and/or
(such as network interface card is adjusted with any equipment for enabling the computer equipment 412 to be communicated with one or more of the other calculating equipment
Modulator-demodulator etc.) communication.This communication can be carried out by input/output (I/O) interface 422.Also, computer equipment
412 can also by network adapter 420 and one or more network (such as local area network (LAN), wide area network (WAN) and/or
Public network, such as internet) communication.As shown, network adapter 420 passes through its of bus 418 and computer equipment 412
The communication of its module.It should be understood that although not shown in fig 4, other hardware and/or soft can be used in conjunction with computer equipment 412
Part module, including but not limited to: microcode, device driver, redundant processing unit, external disk drive array, RAID system,
Tape drive and data backup storage system etc..
Processing unit 416 by the program that is stored in system storage 428 of operation, thereby executing various function application with
And data processing, such as realize the plane recognition methods of laser point cloud provided by the embodiment of the present invention, this method specifically includes that
Obtain the collected frame laser point cloud data of vehicle collector；
The laser point cloud data is subjected to grid division；
For each grid, plane identification is carried out using plane recognizer and set point number threshold value, wherein each grid
Used set point number threshold value according to the distance between grid and acquisition center difference and it is different；
It is handled according to plane recognition result.
Embodiment five
The embodiment of the present invention five additionally provides a kind of computer readable storage medium, is stored thereon with computer program, should
The plane recognition methods of the laser point cloud as provided by the embodiment of the present invention is realized when program is executed by processor, this method is main
Include:
Obtain the collected frame laser point cloud data of vehicle collector；
The laser point cloud data is subjected to grid division；
For each grid, plane identification is carried out using plane recognizer and set point number threshold value, wherein each grid
Used set point number threshold value according to the distance between grid and acquisition center difference and it is different；
It is handled according to plane recognition result.
The computer storage medium of the embodiment of the present invention, can be using any of one or more computer-readable media
Combination.Computer-readable medium can be computer-readable signal media or computer readable storage medium.It is computer-readable
Storage medium for example may be-but not limited to-the system of electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, device or
Device, or any above combination.The more specific example (non exhaustive list) of computer readable storage medium includes: tool
There are electrical connection, the portable computer diskette, hard disk, random access memory (RAM), read-only memory of one or more conducting wires
(ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-
ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.In this document, computer-readable storage
Medium can be any tangible medium for including or store program, which can be commanded execution system, device or device
Using or it is in connection.
Computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including but unlimited
In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can
Any computer-readable medium other than storage medium is read, which can send, propagates or transmit and be used for
By the use of instruction execution system, device or device or program in connection.
The program code for including on computer-readable medium can transmit with any suitable medium, including --- but it is unlimited
In wireless, electric wire, optical cable, RF etc. or above-mentioned any appropriate combination.
The computer for executing operation of the present invention can be write with one or more programming languages or combinations thereof
Program code, described program design language include object oriented program language-such as Java, Smalltalk, C++,
Further include conventional procedural programming language-such as " C " language or similar programming language.Program code can be with
It fully executes, partly execute on the user computer on the user computer, being executed as an independent software package, portion
Divide and partially executes or executed on a remote computer or server completely on the remote computer on the user computer.?
Be related in the situation of remote computer, remote computer can pass through the network of any kind --- including local area network (LAN) or
Wide area network (WAN)-be connected to subscriber computer, or, it may be connected to outer computer (such as mentioned using Internet service
It is connected for quotient by internet).
Note that the above is only a better embodiment of the present invention and the applied technical principle.It will be appreciated by those skilled in the art that
The invention is not limited to the specific embodiments described herein, be able to carry out for a person skilled in the art it is various it is apparent variation,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out by above embodiments to the present invention
It is described in further detail, but the present invention is not limited to the above embodiments only, without departing from the inventive concept, also
It may include more other equivalent embodiments, and the scope of the invention is determined by the scope of the appended claims.
Claims (10)
1. a kind of plane recognition methods of laser point cloud characterized by comprising
Obtain the collected frame laser point cloud data of vehicle collector；
The laser point cloud data is subjected to grid division；
For each grid, plane identification is carried out using plane recognizer and set point number threshold value, wherein each grid is adopted
Set point number threshold value according to the distance between grid and acquisition center difference and it is different；
It is handled according to plane recognition result.
2. the method according to claim 1, wherein according to plane recognition result carry out processing include:
According to plane recognition result, the corresponding point cloud data in road surface is filtered out from laser point cloud data.
3. according to the method described in claim 2, it is characterized in that, each grid is directed to, using plane recognizer and setting
Filtering threshold carries out plane identification
For each grid, the identification of plane iteration is carried out using RANSAC algorithm, until not belonging to plane in the grid
When the quantity at remainder strong point is less than the set point number threshold value, stop plane identification.
4. method according to claim 1 to 3, which is characterized in that each grid is directed to, using plane recognizer
Before the progress plane identification of set point number threshold value, further includes:
The quantity for obtaining data point in each grid, the set point number by the quantity multiplied by setting ratio, as the grid
Threshold value；Or
According to the distance between each grid and acquisition center, the set point of each grid is determined based on setting inverse relation
Number threshold value.
5. according to the method in claim 2 or 3, which is characterized in that according to plane recognition result, from laser point cloud data
Filtering out the corresponding point cloud data in road surface includes:
If the angle between the normal vector and vertical direction of any plane is less than setting angle threshold value, it is determined that the plane is non-
Road surface；
If any plane and height threshold mismatch, it is determined that the plane is non-road surface, wherein the height threshold is according to institute
State the average height setting of point cloud data point in the height and grid of vehicle；
In conjunction with the road surface recognition result of angle threshold value and height threshold, target road surface is determined, and filter from laser point cloud data
Fall the corresponding point cloud data in target road surface.
6. according to the method described in claim 5, it is characterized in that, if any plane and height threshold mismatch, it is determined that
The plane is that non-road surface includes:
According to the point cloud data in each grid, the average height value of each data point in grid is counted；
The dynamic height threshold value of each grid is determined according to known vehicle height value and the average height value；
For each grid, the plan range value of computation grid central point to each plane, if the plan range value is greater than institute
State dynamic height threshold value, it is determined that the plane is non-road surface.
7. according to the method described in claim 6, it is characterized in that, true according to known vehicle height value and the average height value
The dynamic height threshold value of each grid includes: calmly
Both the ratio reduced value of the known vehicle height value and the ratio reduced value of the average height value are calculated, take
Maximum value, the dynamic height threshold value as the grid.
8. a kind of plane identification device of laser point cloud characterized by comprising
Point cloud data obtains module, for obtaining the collected frame laser point cloud data of vehicle collector；
Grid division module, for the laser point cloud data to be carried out grid division；
Plane identification module carries out plane identification using plane recognizer and set point number threshold value for being directed to each grid,
Wherein, set point number threshold value used by each grid according to the distance between grid and acquisition center difference and it is different；
Processing module, for being handled according to plane recognition result.
9. a kind of computer equipment, which is characterized in that the computer equipment includes:
One or more processors；
Storage device, for storing one or more programs；
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
The now plane recognition methods of the laser point cloud as described in any in claim 1-7.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
The plane recognition methods of the laser point cloud as described in any in claim 1-7 is realized when execution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811272869.6A CN109359614B (en) | 2018-10-30 | 2018-10-30 | Laser point cloud plane identification method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811272869.6A CN109359614B (en) | 2018-10-30 | 2018-10-30 | Laser point cloud plane identification method, device, equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109359614A true CN109359614A (en) | 2019-02-19 |
| CN109359614B CN109359614B (en) | 2021-06-11 |
Family
ID=65347367
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811272869.6A Active CN109359614B (en) | 2018-10-30 | 2018-10-30 | Laser point cloud plane identification method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109359614B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110458805A (en) * | 2019-03-26 | 2019-11-15 | 华为技术有限公司 | Plane detection method, computing device and circuit system |
| CN110471074A (en) * | 2019-08-15 | 2019-11-19 | 岭纬科技（厦门）有限公司 | A method of improving planar laser point cloud range accuracy |
| CN110516564A (en) * | 2019-08-06 | 2019-11-29 | 深兰科技（上海）有限公司 | Pavement detection method and apparatus |
| CN110782472A (en) * | 2019-09-05 | 2020-02-11 | 腾讯科技（深圳）有限公司 | Point cloud ground point identification method and device |
| CN111027601A (en) * | 2019-11-25 | 2020-04-17 | 歌尔股份有限公司 | Plane detection method and device based on laser sensor |
| CN111611900A (en) * | 2020-05-15 | 2020-09-01 | 北京京东乾石科技有限公司 | Target point cloud identification method and device, electronic equipment and storage medium |
| CN112286178A (en) * | 2019-07-24 | 2021-01-29 | 本田技研工业株式会社 | Identification system, vehicle control system, identification method, and storage medium |
| CN112634181A (en) * | 2019-09-24 | 2021-04-09 | 北京百度网讯科技有限公司 | Method and apparatus for detecting ground point cloud points |
| CN112634487A (en) * | 2019-09-24 | 2021-04-09 | 北京百度网讯科技有限公司 | Method and apparatus for outputting information |
| CN113219446A (en) * | 2021-04-30 | 2021-08-06 | 森思泰克河北科技有限公司 | In-vehicle radar occupancy identification method and device and vehicle-mounted radar |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102855663A (en) * | 2012-05-04 | 2013-01-02 | 北京建筑工程学院 | Method for building CSG (Constructive Solid Geometry) model according to laser radar grid point cloud |
| CN104143194A (en) * | 2014-08-20 | 2014-11-12 | 清华大学 | Point cloud partition method and device |
| CN104156926A (en) * | 2014-08-19 | 2014-11-19 | 武汉海达数云技术有限公司 | Vehicle-mounted laser point cloud noise point removing methods in multiple scenes |
| CN104251662A (en) * | 2013-06-27 | 2014-12-31 | 杭州中科天维科技有限公司 | Ordered point cloud threshold adaptive noise suppression technology |
| CN104715482A (en) * | 2015-03-20 | 2015-06-17 | 四川大学 | Setting algorithm for calculating interior point threshold in fundamental matrix through RANSAC |
| CN105184852A (en) * | 2015-08-04 | 2015-12-23 | 百度在线网络技术（北京）有限公司 | Laser-point-cloud-based urban road identification method and apparatus |
| 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 |
| CN106408604A (en) * | 2016-09-22 | 2017-02-15 | 北京数字绿土科技有限公司 | Filtering method and device for point cloud data |
| CN106570823A (en) * | 2016-10-11 | 2017-04-19 | 山东科技大学 | Planar feature matching-based point cloud crude splicing method |
| CN107610174A (en) * | 2017-09-22 | 2018-01-19 | 深圳大学 | A kind of plane monitoring-network method and system based on depth information of robust |
| US20180276483A1 (en) * | 2016-08-22 | 2018-09-27 | Tencent Technology (Shenzhen) Company Limited | Road shoulder extraction |
-
2018
- 2018-10-30 CN CN201811272869.6A patent/CN109359614B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102855663A (en) * | 2012-05-04 | 2013-01-02 | 北京建筑工程学院 | Method for building CSG (Constructive Solid Geometry) model according to laser radar grid point cloud |
| CN104251662A (en) * | 2013-06-27 | 2014-12-31 | 杭州中科天维科技有限公司 | Ordered point cloud threshold adaptive noise suppression technology |
| CN104156926A (en) * | 2014-08-19 | 2014-11-19 | 武汉海达数云技术有限公司 | Vehicle-mounted laser point cloud noise point removing methods in multiple scenes |
| CN104143194A (en) * | 2014-08-20 | 2014-11-12 | 清华大学 | Point cloud partition method and device |
| CN104715482A (en) * | 2015-03-20 | 2015-06-17 | 四川大学 | Setting algorithm for calculating interior point threshold in fundamental matrix through RANSAC |
| CN105184852A (en) * | 2015-08-04 | 2015-12-23 | 百度在线网络技术（北京）有限公司 | Laser-point-cloud-based urban road identification method and apparatus |
| 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 |
| US20180276483A1 (en) * | 2016-08-22 | 2018-09-27 | Tencent Technology (Shenzhen) Company Limited | Road shoulder extraction |
| CN106408604A (en) * | 2016-09-22 | 2017-02-15 | 北京数字绿土科技有限公司 | Filtering method and device for point cloud data |
| CN106570823A (en) * | 2016-10-11 | 2017-04-19 | 山东科技大学 | Planar feature matching-based point cloud crude splicing method |
| CN107610174A (en) * | 2017-09-22 | 2018-01-19 | 深圳大学 | A kind of plane monitoring-network method and system based on depth information of robust |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110458805B (en) * | 2019-03-26 | 2022-05-13 | 华为技术有限公司 | Plane detection method, computing device and circuit system |
| CN110458805A (en) * | 2019-03-26 | 2019-11-15 | 华为技术有限公司 | Plane detection method, computing device and circuit system |
| CN112286178B (en) * | 2019-07-24 | 2023-12-01 | 本田技研工业株式会社 | Identification system, vehicle control system, identification method, and storage medium |
| CN112286178A (en) * | 2019-07-24 | 2021-01-29 | 本田技研工业株式会社 | Identification system, vehicle control system, identification method, and storage medium |
| CN110516564A (en) * | 2019-08-06 | 2019-11-29 | 深兰科技（上海）有限公司 | Pavement detection method and apparatus |
| CN110471074B (en) * | 2019-08-15 | 2021-04-27 | 岭纬科技（厦门）有限公司 | Method for improving planar laser point cloud distance measurement precision |
| CN110471074A (en) * | 2019-08-15 | 2019-11-19 | 岭纬科技（厦门）有限公司 | A method of improving planar laser point cloud range accuracy |
| CN110782472A (en) * | 2019-09-05 | 2020-02-11 | 腾讯科技（深圳）有限公司 | Point cloud ground point identification method and device |
| CN110782472B (en) * | 2019-09-05 | 2022-06-07 | 腾讯科技（深圳）有限公司 | Point cloud ground point identification method and device |
| CN112634487A (en) * | 2019-09-24 | 2021-04-09 | 北京百度网讯科技有限公司 | Method and apparatus for outputting information |
| CN112634181A (en) * | 2019-09-24 | 2021-04-09 | 北京百度网讯科技有限公司 | Method and apparatus for detecting ground point cloud points |
| CN111027601B (en) * | 2019-11-25 | 2023-10-17 | 歌尔股份有限公司 | Plane detection method and device based on laser sensor |
| CN111027601A (en) * | 2019-11-25 | 2020-04-17 | 歌尔股份有限公司 | Plane detection method and device based on laser sensor |
| CN111611900A (en) * | 2020-05-15 | 2020-09-01 | 北京京东乾石科技有限公司 | Target point cloud identification method and device, electronic equipment and storage medium |
| CN111611900B (en) * | 2020-05-15 | 2023-06-30 | 北京京东乾石科技有限公司 | Target point cloud identification method and device, electronic equipment and storage medium |
| CN113219446A (en) * | 2021-04-30 | 2021-08-06 | 森思泰克河北科技有限公司 | In-vehicle radar occupancy identification method and device and vehicle-mounted radar |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109359614B (en) | 2021-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109359614A (en) | A kind of plane recognition methods, device, equipment and the medium of laser point cloud | |
| US11415672B2 (en) | Method and apparatus for generating object detection box, device, storage medium, and vehicle | |
| US11024045B2 (en) | Method and apparatus for generating three-dimensional data, device, and storage medium | |
| CN110286387B (en) | Obstacle detection method and device applied to automatic driving system and storage medium | |
| CN109345596B (en) | Multi-sensor calibration method, device, computer equipment, medium and vehicle | |
| JP7112993B2 (en) | Laser Radar Internal Parameter Accuracy Verification Method and Its Apparatus, Equipment and Medium | |
| CN112634181B (en) | Method and device for detecting ground point cloud points | |
| EP3627180A1 (en) | Sensor calibration method and device, computer device, medium, and vehicle | |
| CN106951847B (en) | Obstacle detection method, apparatus, device and storage medium | |
| EP3624055B1 (en) | Ground detection method, apparatus, electronic device, vehicle and storage medium | |
| CN109059902A (en) | Relative pose determines method, apparatus, equipment and medium | |
| CN110389341B (en) | Charging pile identification method and device, robot and computer readable storage medium | |
| WO2021081808A1 (en) | Artificial neural network-based object detection system and method | |
| CN114387319B (en) | Point cloud registration method, device, equipment and storage medium | |
| WO2023040737A1 (en) | Target location determining method and apparatus, electronic device, and storage medium | |
| CN115880435A (en) | Image reconstruction method, model training method, device, electronic device and medium | |
| CN109827610B (en) | Method and device for verifying sensor fusion result | |
| CN113177980A (en) | Target object speed determination method and device for automatic driving and electronic equipment | |
| CN116205964B (en) | Point cloud downsampling method and device based on horizontal distance | |
| CN115866229A (en) | Method, apparatus, device and medium for converting view angle of multi-view image | |
| CN114549961B (en) | Target object detection method, device, equipment and storage medium | |
| CN113920273B (en) | Image processing method, device, electronic equipment and storage medium | |
| CN115937950A (en) | Multi-angle face data acquisition method, device, equipment and storage medium | |
| CN112258566B (en) | Pavement collection point identification method and device and server | |
| US20210207964A1 (en) | Verification Method And Device For Modeling Route, Unmanned Vehicle, And Storage Medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20211021 Address after: 105 / F, building 1, No. 10, Shangdi 10th Street, Haidian District, Beijing 100085 Patentee after: Apollo Intelligent Technology (Beijing) Co.,Ltd. Address before: 100085 Baidu Building, 10 Shangdi Tenth Street, Haidian District, Beijing Patentee before: BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) Co.,Ltd. |
|
| TR01 | Transfer of patent right |