CN105957145A - Road barrier identification method and device - Google Patents
Road barrier identification method and device Download PDFInfo
- Publication number
- CN105957145A CN105957145A CN201610282676.3A CN201610282676A CN105957145A CN 105957145 A CN105957145 A CN 105957145A CN 201610282676 A CN201610282676 A CN 201610282676A CN 105957145 A CN105957145 A CN 105957145A
- Authority
- CN
- China
- Prior art keywords
- barrier
- point cloud
- markup information
- laser point
- cube
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/05—Geographic models
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/08—Indexing scheme for image data processing or generation, in general involving all processing steps from image acquisition to 3D model generation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2210/00—Indexing scheme for image generation or computer graphics
- G06T2210/21—Collision detection, intersection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Remote Sensing (AREA)
- Computer Graphics (AREA)
- Multimedia (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a road barrier identification method and device. One specific embodiment of the invention comprises the steps of: obtaining laser point cloud of a barrier object on a road; based on the laser point cloud, obtaining mark information of the barrier object; and based on the mark information, generating training samples of a machine learning model. According to the invention, different type of mark information of the barrier objects are obtained based on the laser point cloud of the barrier objects, so that the sizes, positions and angles of the different kinds of barrier objects are accurately determined; and furthermore, based on the mark information of the barrier object, the training samples of the machine learning model for identifying the sizes, positions and angles of the different kinds of barriers using the laser point cloud of the barriers as input are generated, and the identification accuracy of the machine learning model is continuously improved.
Description
Technical field
The application relates to computer realm, is specifically related to field of image recognition, particularly relates to road
Obstacle recognition method and device.
Background technology
The computer technology extensively application on automobile makes automobile more and more intelligent.In garage
During sailing, it is possible to use Vehicle-borne Laser Scanning equipment obtains the laser spots of the barrier on road
Cloud, needs to utilize machine learning model with laser point cloud for input, size, the position to barrier
Put, angle is identified.
But, the laser point cloud of all barrier objects owing to getting mixes,
There will be machine learning model and correctly the feelings of overlapping barrier object cannot occur demarcation of location
Condition, and then cause the recognition result of mistake to be corrected in time, reduce machine learning model
Recognition accuracy.
Summary of the invention
This application provides road barrier recognition methods and device, be used for solving above-mentioned background skill
The technical problem that art part exists.
First aspect, this application provides road barrier recognition methods, and the method includes: obtain
Laser point cloud by way of the barrier object on road；Based on laser point cloud, obtain barrier object
Markup information, markup information includes: size, position, angle；Based on markup information, raw
Become the training sample of machine learning model, to utilize training sample that machine learning model is instructed
Practice, machine learning model be laser point cloud based on barrier object to the size of barrier object,
The model that position, angle are identified.
Second aspect, this application provides road barrier identification device, and this device includes: point
Cloud acquiring unit, is configured to obtain the laser point cloud of the barrier object on road；Information obtains
Take unit, be configured to, based on laser point cloud, obtain the markup information of barrier object, mark
Information includes: size, position, angle；Signal generating unit, is configured to based on markup information,
Generate the training sample of machine learning model, to utilize training sample that machine learning model is carried out
Training, machine learning model is big to barrier object of laser point cloud based on barrier object
The model that little, position, angle are identified.
The road barrier recognition methods of the application offer and device, obtain the barrier on road
The laser point cloud of object；Based on laser point cloud, obtain the markup information of barrier object；Based on
Markup information, generates the training sample of machine learning model.Achieve by based on barrier pair
The laser point cloud of elephant, obtains the markup information of different types of barrier object, thus exactly
Determine the size of different types of barrier object, position, angle, be based further on obstacle
The markup information of thing object, generates using the laser point cloud of barrier as input, to barrier
The training sample of the machine learning model that size, position, angle are identified, and then constantly
The recognition accuracy of hoisting machine learning model.
Accompanying drawing explanation
By reading retouching in detail with reference to made non-limiting example is made of the following drawings
Stating, other features, purpose and advantage will become more apparent upon:
Fig. 1 is that the application can apply to exemplary system architecture figure therein；
Fig. 2 shows the stream of an embodiment of the road barrier recognition methods according to the application
Cheng Tu；
Fig. 3 shows one exemplary conceptual diagram of road barrier recognition methods of the application；
Fig. 4 shows the knot of an embodiment of the road barrier identification device according to the application
Structure schematic diagram；
Fig. 5 is adapted for the computer for the terminal unit or server realizing the embodiment of the present application
The structural representation of system.
Detailed description of the invention
With embodiment, the application is described in further detail below in conjunction with the accompanying drawings.It is appreciated that
, specific embodiment described herein is used only for explaining related invention, rather than to this
Bright restriction.It also should be noted that, for the ease of describe, accompanying drawing illustrate only with
About the part that invention is relevant.
It should be noted that in the case of not conflicting, the embodiment in the application and embodiment
In feature can be mutually combined.Describe this below with reference to the accompanying drawings and in conjunction with the embodiments in detail
Application.
Fig. 1 shows road barrier recognition methods or the enforcement of device that can apply the application
The exemplary system architecture 100 of example.
As it is shown in figure 1, system architecture 100 can include terminal unit 101,102,103,
Network 104 and server 105.Network 104 is in order at terminal unit 101,102,103 and
The medium of transmission link is provided between server 105.Network 104 can include various connection class
Type, the most wired, wireless transmission link or fiber optic cables etc..
User can use terminal unit 101,102,103 by network 104 and server 105
Alternately, to receive or to send message etc..Can be provided with on terminal unit 101,102,103
Various communication applications, such as, laser point cloud processes class application, browser class is applied, searching class
Application, word processing class application etc..
Terminal unit 101,102,103 can have display screen and support network service
Various electronic equipments, include but not limited to smart mobile phone, panel computer, E-book reader,
(Moving Picture Experts Group Audio Layer III, dynamic image is special for MP3 player
Family compression standard audio frequency aspect 3), MP4 (Moving Picture Experts Group Audio
Layer IV, dynamic image expert's compression standard audio frequency aspect 4) player, portable meter on knee
Calculation machine and desk computer etc..
Server 105 can send swashing of barrier object to terminal unit 101,102,103
Luminous point cloud.It is labeled by 101,102,103 pairs of barrier objects of terminal unit, i.e. obtains
The markup informations such as the size of barrier object, position, angle.Terminal unit 101,102,103
The markup information of barrier object can be sent to server 105, so that server 105
On machine learning model barrier object being identified based on laser point cloud can utilize base
The training sample generated in markup information is trained, and hoisting machine learning model is to barrier
The size of object, position, the recognition accuracy of angle.
It should be understood that the number of terminal unit, network and the server in Fig. 1 is only signal
Property.According to realizing needs, can have any number of terminal unit, network and server.
Refer to Fig. 2, it illustrates of road barrier recognition methods according to the application
The flow process 200 of embodiment.It should be noted that the road barricade that the embodiment of the present application is provided
Thing recognition methods is typically performed by the terminal 101,102,103 in Fig. 1.
Step 201, the laser point cloud of the barrier object on acquisition road.
In the present embodiment, Vehicle-borne Laser Scanning equipment can be used to gather the barrier on road in advance
Hinder the laser point cloud of thing object.Such as, at vehicle (such as pilotless automobile) driving process
In, it is possible to use the Vehicle-borne Laser Scanning instrument on vehicle travels with default frequency acquisition collection vehicle
Road on the laser point cloud of barrier object, thus get the barrier object on road
Laser point cloud.
In some optional implementations of the present embodiment, barrier object includes: vehicle pair
As, pedestrian's object, traffic mark object.Vehicle Object can include but not limited to: bicycle,
Car, truck, minibus, bus.Traffic mark object can include but not limited to:
Lane line, traffic mark board, instruction graticule, traffic lights.
In some optional implementations of the present embodiment, obtain the barrier object on road
Laser point cloud include: obtain the laser point cloud of barrier object road from server.
In the present embodiment, the laser point cloud of barrier object can store on the server.Example
As, in vehicle (such as pilotless automobile) driving process, it is possible to use the car on vehicle
Carry barrier object on the road that laser scanner travels with default frequency acquisition collection vehicle
Laser point cloud, it is then possible to carry out the laser point cloud transmission of the magnanimity collected to server
Storage.When barrier object is labeled by needs, barrier can be obtained from Cloud Server
The laser point cloud of object.
Step 202, based on laser point cloud, obtains the markup information of barrier object.
In the present embodiment, at the barrier object obtained by step 201 on road, such as
Vehicle Object, pedestrian's object, traffic mark object laser point cloud after, can be based on vehicle
The laser point cloud of the barrier objects such as object, pedestrian's object, traffic mark object, obtains obstacle
The markup information of thing object.The markup information of barrier object can include but not limited to: size,
Position, angle.
In some optional implementations of the present embodiment, based on laser point cloud, obtain obstacle
The markup information of thing object includes: utilize the cube in Open Framework cloudcompare framework
Object, based on laser point cloud, obtains the markup information of barrier object.
In some optional implementations of the present embodiment, utilize Open Framework
Cube object in cloudcompare framework, based on laser point cloud, obtains barrier object
Markup information includes: create the cube object that barrier object is corresponding；Adjust cube object
Central point overlap with the central point of barrier object；Adjust the size of cube object, position,
Angle, so that cube object covers barrier object；Obtain the cube object after adjusting
Size, position, angle；Using size, position, angle as markup information.
In the present embodiment, Open Framework cloudcompare frame can be utilized in the following manner
Cube object in frame i.e. cube object, based on laser point cloud, obtains the mark of barrier object
Information: can first the Vehicle Object got, pedestrian's object, traffic mark object etc. be hindered
The laser point cloud hindering thing object imports cloudcompare framework.Cloudcompare framework is permissible
For processing laser point cloud.The laser point cloud of barrier object is being imported cloudcompare frame
After frame, the laser point cloud of barrier object may be displayed on the interface of cloudcompare framework
On.
In the present embodiment, can be in cloudcompare framework, for Vehicle Object, OK
The barrier objects such as people's object, pre-set various types of barrier object each self-corresponding vertical
Cube object i.e. cube object.The cube object that different types of barrier object is corresponding is permissible
Different colors is used to make a distinction.
In the present embodiment, can be based on display vehicle on the interface of cloudcompare framework
Object, pedestrian's object, the laser point cloud of traffic mark object, adjust Vehicle Object, pedestrian couple
As the sizes of each self-corresponding cube object of barrier object such as, traffic mark object, position,
Angle.It is then possible to according to the size of cube object after adjusting, position, angle, obtain
The size of barrier object, position, the angles such as Vehicle Object, pedestrian's object, traffic mark object
The markup information of the barrier objects such as degree.
In the present embodiment, can adjust that barrier object is corresponding in the following ways cube pair
The size of elephant, position, angle, obtain the markup information of barrier object: can first create
The cube object that different types of barrier object is corresponding.Then, at cloudcompare framework
Interface on add the cube object that different types of barrier is corresponding so that different types of barrier
The cube object hindering thing object corresponding shows on the interface of cloudcompare framework.Show
The laser point cloud of the barrier object on the interface of cloudcompare framework may be used for describing barrier
Hinder the profile of thing, can according to display laser point cloud on the interface of cloudcompare framework,
Determine the profile of different types of barrier object.Can be by the boundary at cloudcompare framework
The central point weight of the central point of the cube object on face and this cube object corresponding barrier object
Close.It is then possible to adjust the size of cube object, position, angle so that cube object
The barrier object of correspondence can be covered, such as, adjust profile and the barrier pair of cube object
The contour convergence of elephant.The cube object that different types of barrier object is corresponding can be adjusted respectively
Size, position, angle so that different types of cube object can cover the most right
The barrier object answered.
In the present embodiment, at the cube object that the barrier object adjusting different types is corresponding
Size, position, angle, cover after barrier object, can pass through further
Cloudcompare framework provide readings cube object interface acquisition adjusted after cube
The size of object, position, angle, i.e. the size of cube object, position, angle.Thus can
With utilize adjusted after the size of cube object, position, angle the barrier of correspondence is described
The size of object, position, angle.Can using the size of cube object, position, angle as
The markup information of barrier object, thus get the markup information of barrier object, i.e. complete
Mark to barrier object.
Step 203, based on markup information, generates the training sample of machine learning model.
In the present embodiment, by step 202 based on laser point cloud, obtain barrier object
Markup information after, can based on the markup information of barrier object, generate machine learning mould
The training sample of type.Such as, markup information is converted to the input vector of machine learning model,
Thus utilize training sample that machine learning model is trained.This machine learning model can be
Using the laser point cloud of barrier as input, the size of barrier, position, angle are known
Other machine learning model, such as degree of depth learning model.
In the present embodiment, by laser point cloud based on barrier object, obtain dissimilar
The size of barrier object, position, the markup information such as angle, thus accurately determine out not
With the size of barrier object of type, position, angle.It is then possible to based on barrier pair
The markup information of elephant, generates using the laser point cloud of barrier as input, to the size of barrier,
The training sample of the machine learning model that position, angle are identified.This training sample can be utilized
This is to using the laser point cloud of barrier as input, enters the size of barrier, position, angle
The machine learning model that row identifies is trained, and then the knowledge of constantly hoisting machine learning model
Other accuracy rate.
In some optional implementations of the present embodiment, based on markup information, generate machine
The training sample of learning model includes: send markup information to server, with base on the server
The instruction for the machine learning model being arranged on server is trained is generated in markup information
Practice sample.
In the present embodiment, using the laser point cloud of barrier as input, to the size of barrier,
The machine learning model that position, angle are identified can be arranged on the server.Can will mark
Note information sends to server, such that it is able to mark based on barrier object letter on the server
Breath, generates training sample, utilizes this training sample to the laser point cloud using barrier as input,
The machine learning model being identified the size of barrier, position, angle is trained, and enters
And constantly hoisting machine learning model to the size of barrier, position, angle identification accurate
Rate.
Refer to Fig. 3, the road barrier recognition methods one that it illustrates the application is exemplary
Schematic diagram.
In fig. 3 it is shown that client and server.Client can be arranged in terminal,
Client can be configured with cloudcompare framework.It is right that client can be configured with
Cube object in cloudcompare framework i.e. cube object carries out the cube object operated
Operation interface.Such as, client can be configured with and add the cube that different types of barrier is corresponding
The interface of object, the interface of editor's cube object, adjust cube object corresponding to barrier
The interface of size, adjust the interface of the position of cube object corresponding to barrier, adjust barrier
The interface of the angle of the cube object that object is corresponding.
The laser point cloud of the barrier object that server can store magnanimity is (the most original
Laser point cloud) and the markup information of barrier object.Can be provided with obstacle on server
The laser point cloud of thing as input, the machine that the size of barrier, position, angle are identified
Device learning model.
Client can get the laser point cloud of barrier object from server.It is then possible to
Barrier object is labeled by laser point cloud based on barrier object, i.e. obtains barrier pair
The markup informations such as the size of elephant, position, angle.Such as, client can receive user's input
Operational order, operational order can call cube Object Operations interface, corresponding to barrier
The size of cube object, position, angle are adjusted.Cloudcompare framework can be passed through
There is provided API obtain adjusted after the size of cube object, position, angle.
Such that it is able to utilize adjusted after the size of cube object, position, angle right to describe
The size of barrier object answered, position, angle.Can by the size of cube object, position,
Angle is as the markup information of barrier object, thus gets the markup information of barrier object,
I.e. complete the mark to barrier object.
Complete the mark to barrier object in client, i.e. get the mark of barrier object
After information, JSON (JavaScript Object Notation) data lattice can be used in client
The markup information of formula storage barrier object.User can be on the client to barrier object
The operations such as history markup information carries out checking, editor.
Meanwhile, the markup information of barrier object can be sent to server by client, thus
Can markup information based on barrier object on the server, generate training sample.Can be in order to
With this training sample to using the laser point cloud of barrier as input, size, the position to barrier
Put, machine learning model that angle is identified is trained, and then constantly hoisting machine
Practise the recognition accuracy of model.
With further reference to Fig. 4, as to the realization of method shown in above-mentioned each figure, the application provides
A kind of embodiment of road barrier identification device, this device embodiment with shown in Fig. 2
Embodiment of the method corresponding, this device specifically can apply in various electronic equipment.
As shown in Figure 4, the road barrier identification device 400 of the present embodiment includes: some cloud obtains
Take unit 401, information acquisition unit 402, signal generating unit 403.Wherein, some cloud acquiring unit
401 are configured to obtain the laser point cloud of the barrier object on road；Information acquisition unit 402
Being configured to, based on laser point cloud, obtain the markup information of barrier object, markup information includes:
Size, position, angle；Signal generating unit 403 is configured to, based on markup information, generate machine
The training sample of learning model, to utilize training sample that machine learning model is trained, machine
Device learning model be laser point cloud based on barrier object to the size of barrier object, position,
The model that angle is identified.
In some optional implementations of the present embodiment, information acquisition unit 402 includes:
Markup information obtains subelement (not shown), is configured to utilize Open Framework cloudcompare
Cube object in framework, based on described laser point cloud, obtains the markup information of barrier object.
In some optional implementations of the present embodiment, barrier object includes: vehicle pair
As, pedestrian's object, traffic mark object.
In some optional implementations of the present embodiment, markup information obtains subelement and enters one
Step is configured to: create the cube object that barrier object is corresponding；Adjust cube object
Central point overlaps with the central point of barrier object；Adjust the size of cube object, position,
Angle, so that cube object covers barrier object；Obtain the cube object after adjusting
Size, position, angle；Using described size, position, angle as described markup information.
In some optional implementations of the present embodiment, some cloud acquiring unit 401 includes:
Laser point cloud obtains subelement (not shown), is configured to obtain the obstacle road from server
The laser point cloud of thing object.
In some optional implementations of the present embodiment, signal generating unit 403 includes: send
Subelement (not shown), is configured to send markup information, with base on the server to server
Generate for the described machine learning model being arranged on server is trained in markup information
Training sample.
Fig. 5 shows the meter be suitable to for the terminal unit or server realizing the embodiment of the present application
The structural representation of calculation machine system.
As it is shown in figure 5, computer system 500 includes CPU (CPU) 501, its
Can be according to the program being stored in read only memory (ROM) 502 or from storage part 508
It is loaded into the program in random access storage device (RAM) 503 and performs various suitable action
And process.In RAM503, also storage has system 500 to operate required various program sums
According to.CPU701, ROM 502 and RAM503 is connected with each other by bus 504.Input/
Output (I/O) interface 505 is also connected to bus 504.
It is connected to I/O interface 505: include the importation 506 of keyboard, mouse etc. with lower component；
Including such as cathode ray tube (CRT), liquid crystal display (LCD) etc. and speaker etc.
Output part 507；Storage part 508 including hard disk etc.；And include such as LAN card,
The communications portion 509 of the NIC of modem etc..Communications portion 509 is via such as
The network of the Internet performs communication process.Driver 510 is connected to I/O interface also according to needs
505.Detachable media 511, such as disk, CD, magneto-optic disk, semiconductor memory etc.,
Be arranged on as required in driver 510, in order to the computer program read from it according to
Needs are mounted into storage part 508.
Especially, according to embodiment of the disclosure, the process described above with reference to flow chart is permissible
It is implemented as computer software programs.Such as, embodiment of the disclosure and include a kind of computer journey
Sequence product, it includes the computer program being tangibly embodied on machine readable media, described meter
Calculation machine program comprises the program code for performing the method shown in flow chart.In such enforcement
In example, this computer program can be downloaded and installed from network by communications portion 509,
And/or be mounted from detachable media 511.
Flow chart in accompanying drawing and block diagram, it is illustrated that according to the various embodiment of the application system,
Architectural framework in the cards, function and the operation of method and computer program product.This point
On, each square frame in flow chart or block diagram can represent a module, program segment or code
A part, a part for described module, program segment or code comprise one or more for
Realize the executable instruction of the logic function of regulation.It should also be noted that at some as replacement
In realization, the function marked in square frame can also be sent out to be different from the order marked in accompanying drawing
Raw.Such as, two square frames succeedingly represented can essentially perform substantially in parallel, they
Sometimes can also perform in the opposite order, this is depending on involved function.It is also noted that
It is, the square frame in each square frame in block diagram and/or flow chart and block diagram and/or flow chart
Combination, can realize by the special hardware based system of the function or operation that perform regulation,
Or can realize with the combination of specialized hardware with computer instruction.
As on the other hand, present invention also provides a kind of nonvolatile computer storage media,
This nonvolatile computer storage media can be described in above-described embodiment included in device
Nonvolatile computer storage media；Can also be individualism, be unkitted allocate in terminal non-
Volatile computer storage medium.Above-mentioned nonvolatile computer storage media storage have one or
The multiple program of person, when one or more program is performed by an equipment so that described
Equipment: the laser point cloud of the barrier object on acquisition road；Based on described laser point cloud, obtain
Taking the markup information of barrier object, described markup information includes: size, position, angle；
Based on described markup information, generate the training sample of machine learning model, to utilize described training
Machine learning model is trained by sample, and described machine learning model is based on barrier object
Laser point cloud model that the size of barrier object, position, angle are identified.
Above description is only the preferred embodiment of the application and saying institute's application technology principle
Bright.It will be appreciated by those skilled in the art that invention scope involved in the application, do not limit
In the technical scheme of the particular combination of above-mentioned technical characteristic, also should contain simultaneously without departing from
In the case of described inventive concept, above-mentioned technical characteristic or its equivalent feature carry out combination in any
And other technical scheme formed.Such as features described above and (but not limited to) disclosed herein
The technical characteristic with similar functions is replaced mutually and the technical scheme that formed.
Claims (12)
1. a road barrier recognition methods, it is characterised in that described method includes:
The laser point cloud of the barrier object on acquisition road；
Based on described laser point cloud, obtain the markup information of barrier object, described markup information
Including: size, position, angle；
Based on described markup information, generate the training sample of machine learning model, described to utilize
Machine learning model is trained by training sample, and described machine learning model is based on barrier
The model that the size of barrier object, position, angle are identified by the laser point cloud of object.
Method the most according to claim 1, it is characterised in that described based on described laser
Point cloud, the markup information obtaining barrier object includes:
Utilize the cube object in Open Framework cloudcompare framework based on described laser spots
Cloud, obtains the markup information of barrier object.
Method the most according to claim 2, it is characterised in that barrier object includes:
Vehicle Object, pedestrian's object, traffic mark object.
Method the most according to claim 3, it is characterised in that utilize Open Framework
Cube object in cloudcompare framework, based on described laser point cloud, obtains barrier pair
The markup information of elephant includes:
Create the cube object that barrier object is corresponding；
The central point adjusting cube object overlaps with the central point of barrier object；
Adjust the size of cube object, position, angle, so that cube object covers obstacle
Thing object；
Obtain the size of cube object after adjusting, position, angle；
Using described size, position, angle as described markup information.
Method the most according to claim 4, it is characterised in that obtain the obstacle on road
The laser point cloud of thing object includes:
The laser point cloud of the barrier object from server acquisition road.
Method the most according to claim 5, it is characterised in that based on described markup information,
The training sample generating machine learning model includes:
Described markup information is sent, with raw based on described markup information on the server to server
Become the training sample for the described machine learning model being arranged on server is trained.
7. a road barrier object recognition equipment, it is characterised in that described device includes:
Point cloud acquiring unit, is configured to obtain the laser point cloud of the barrier object on road；
Information acquisition unit, is configured to, based on described laser point cloud, obtain barrier object
Markup information, described markup information includes: size, position, angle；
Signal generating unit, is configured to, based on described markup information, generate the instruction of machine learning model
Practice sample, to utilize described training sample that machine learning model is trained, described engineering
Practising model is laser point cloud based on barrier object to the size of barrier object, position, angle
The model that degree is identified.
Device the most according to claim 7, it is characterised in that information acquisition unit includes:
Markup information obtains subelement, is configured to utilize Open Framework cloudcompare framework
In cube object based on described laser point cloud, obtain the markup information of barrier object.
Device the most according to claim 8, it is characterised in that barrier object includes:
Vehicle Object, pedestrian's object, traffic mark object.
Device the most according to claim 9, it is characterised in that markup information obtains son
Unit is configured to further: create the cube object that barrier object is corresponding；Adjustment cube
The central point of body object overlaps with the central point of barrier object；Adjust cube object size,
Position, angle, so that cube object covers barrier object；Obtain the cube after adjusting
The size of object, position, angle；Using described size, position, angle as described mark letter
Breath.
11. devices according to claim 10, it is characterised in that some cloud acquiring unit bag
Include laser point cloud and obtain subelement, be configured to obtain the barrier object road from server
Laser point cloud.
12. devices according to claim 11, it is characterised in that described signal generating unit bag
Include:
Send subelement, be configured to send described markup information, with at server to server
On generate for the described machine learning model being arranged on server based on described markup information
The training sample being trained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610282676.3A CN105957145A (en) | 2016-04-29 | 2016-04-29 | Road barrier identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610282676.3A CN105957145A (en) | 2016-04-29 | 2016-04-29 | Road barrier identification method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105957145A true CN105957145A (en) | 2016-09-21 |
Family
ID=56914997
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610282676.3A Pending CN105957145A (en) | 2016-04-29 | 2016-04-29 | Road barrier identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105957145A (en) |
Cited By (35)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106485233A (en) * | 2016-10-21 | 2017-03-08 | 深圳地平线机器人科技有限公司 | Drivable region detection method, device and electronic equipment |
| CN106503653A (en) * | 2016-10-21 | 2017-03-15 | 深圳地平线机器人科技有限公司 | Area marking method, device and electronic equipment |
| CN106778548A (en) * | 2016-11-30 | 2017-05-31 | 百度在线网络技术（北京）有限公司 | Method and apparatus for detecting barrier |
| CN106845412A (en) * | 2017-01-20 | 2017-06-13 | 百度在线网络技术（北京）有限公司 | Obstacle recognition method and device, computer equipment and computer-readable recording medium |
| CN106951847A (en) * | 2017-03-13 | 2017-07-14 | 百度在线网络技术（北京）有限公司 | Obstacle detection method, device, equipment and storage medium |
| CN107871129A (en) * | 2016-09-27 | 2018-04-03 | 北京百度网讯科技有限公司 | Method and apparatus for handling cloud data |
| CN107907886A (en) * | 2017-11-07 | 2018-04-13 | 广东欧珀移动通信有限公司 | Travel conditions recognition methods, device, storage medium and terminal device |
| CN108280886A (en) * | 2018-01-25 | 2018-07-13 | 北京小马智行科技有限公司 | Laser point cloud mask method, device and readable storage medium storing program for executing |
| CN108596223A (en) * | 2018-04-11 | 2018-09-28 | 珠海博明视觉科技有限公司 | A method of automatically generating object data set |
| CN108630015A (en) * | 2018-05-21 | 2018-10-09 | 浙江吉利汽车研究院有限公司 | A kind of driving warning method, device and electronic equipment |
| CN108734780A (en) * | 2017-04-17 | 2018-11-02 | 百度在线网络技术（北京）有限公司 | Method, apparatus and equipment for generating map |
| CN109101861A (en) * | 2017-06-20 | 2018-12-28 | 百度在线网络技术（北京）有限公司 | Obstacle identity recognition methods, device, equipment and storage medium |
| CN109215135A (en) * | 2018-09-07 | 2019-01-15 | 百度在线网络技术（北京）有限公司 | A kind of Obstacle Position emulation mode, device and terminal based on statistics |
| CN109344804A (en) * | 2018-10-30 | 2019-02-15 | 百度在线网络技术（北京）有限公司 | A kind of recognition methods of laser point cloud data, device, equipment and medium |
| CN109376664A (en) * | 2018-10-29 | 2019-02-22 | 百度在线网络技术（北京）有限公司 | Machine learning training method, device, server and medium |
| CN109444915A (en) * | 2018-09-11 | 2019-03-08 | 成都优艾维智能科技有限责任公司 | A kind of danger zone pre-judging method based on laser radar data |
| CN109711285A (en) * | 2018-12-11 | 2019-05-03 | 百度在线网络技术（北京）有限公司 | Training, test method and the device of identification model |
| CN109709961A (en) * | 2018-12-28 | 2019-05-03 | 百度在线网络技术（北京）有限公司 | Road barricade object detecting method, device and autonomous driving vehicle |
| CN109858374A (en) * | 2018-12-31 | 2019-06-07 | 武汉中海庭数据技术有限公司 | Arrow class graticule extraction method and device in high-precision cartography |
| CN109934141A (en) * | 2019-03-01 | 2019-06-25 | 北京百度网讯科技有限公司 | Method and apparatus for labeled data |
| CN110379157A (en) * | 2019-06-04 | 2019-10-25 | 深圳市速腾聚创科技有限公司 | Road blind area monitoring method, system, device, equipment and storage medium |
| CN110807439A (en) * | 2019-11-12 | 2020-02-18 | 银河水滴科技（北京）有限公司 | Method and device for detecting obstacle |
| CN110858415A (en) * | 2018-08-24 | 2020-03-03 | 北京图森未来科技有限公司 | Method and device for labeling object in 3D point cloud data |
| CN110866504A (en) * | 2019-11-20 | 2020-03-06 | 北京百度网讯科技有限公司 | Method, device and equipment for acquiring marked data |
| EP3621040A1 (en) * | 2018-09-07 | 2020-03-11 | Baidu Online Network Technology (Beijing) Co., Ltd. | Data augmentation method, device and terminal |
| CN112116804A (en) * | 2019-06-19 | 2020-12-22 | 北京地平线机器人技术研发有限公司 | Vehicle state quantity information determination method and device |
| CN112269379A (en) * | 2020-10-14 | 2021-01-26 | 北京石头世纪科技股份有限公司 | Obstacle identification information feedback method |
| US10984588B2 (en) | 2018-09-07 | 2021-04-20 | Baidu Online Network Technology (Beijing) Co., Ltd | Obstacle distribution simulation method and device based on multiple models, and storage medium |
| US11047673B2 (en) | 2018-09-11 | 2021-06-29 | Baidu Online Network Technology (Beijing) Co., Ltd | Method, device, apparatus and storage medium for detecting a height of an obstacle |
| US11126875B2 (en) | 2018-09-13 | 2021-09-21 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method and device of multi-focal sensing of an obstacle and non-volatile computer-readable storage medium |
| WO2021213241A1 (en) * | 2020-04-20 | 2021-10-28 | 上海商汤临港智能科技有限公司 | Target detection method and apparatus, and electronic device, storage medium and program |
| US11276243B2 (en) | 2018-09-07 | 2022-03-15 | Baidu Online Network Technology (Beijing) Co., Ltd. | Traffic simulation method, device and storage medium |
| US11307302B2 (en) | 2018-09-07 | 2022-04-19 | Baidu Online Network Technology (Beijing) Co., Ltd | Method and device for estimating an absolute velocity of an obstacle, and non-volatile computer-readable storage medium |
| US11718318B2 (en) | 2019-02-22 | 2023-08-08 | Apollo Intelligent Driving (Beijing) Technology Co., Ltd. | Method and apparatus for planning speed of autonomous vehicle, and storage medium |
| US11780463B2 (en) | 2019-02-19 | 2023-10-10 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method, apparatus and server for real-time learning of travelling strategy of driverless vehicle |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20080100859A (en) * | 2007-02-14 | 2008-11-21 | 한국건설기술연구원 | Portable roadway detector evaluation system and method for processing the same |
| CN102779280A (en) * | 2012-06-19 | 2012-11-14 | 武汉大学 | Traffic information extraction method based on laser sensor |
| 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 |
| CN104766058A (en) * | 2015-03-31 | 2015-07-08 | 百度在线网络技术（北京）有限公司 | Method and device for obtaining lane line |
| CN105116418A (en) * | 2015-09-08 | 2015-12-02 | 北京铁路局北京科学技术研究所 | Obstacle detection method and apparatus |
| CN105469098A (en) * | 2015-11-20 | 2016-04-06 | 中北大学 | Precise LINDAR data ground object classification method based on adaptive characteristic weight synthesis |
-
2016
- 2016-04-29 CN CN201610282676.3A patent/CN105957145A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20080100859A (en) * | 2007-02-14 | 2008-11-21 | 한국건설기술연구원 | Portable roadway detector evaluation system and method for processing the same |
| CN102779280A (en) * | 2012-06-19 | 2012-11-14 | 武汉大学 | Traffic information extraction method based on laser sensor |
| 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 |
| CN104766058A (en) * | 2015-03-31 | 2015-07-08 | 百度在线网络技术（北京）有限公司 | Method and device for obtaining lane line |
| CN105116418A (en) * | 2015-09-08 | 2015-12-02 | 北京铁路局北京科学技术研究所 | Obstacle detection method and apparatus |
| CN105469098A (en) * | 2015-11-20 | 2016-04-06 | 中北大学 | Precise LINDAR data ground object classification method based on adaptive characteristic weight synthesis |
Cited By (51)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107871129B (en) * | 2016-09-27 | 2019-05-10 | 北京百度网讯科技有限公司 | Method and apparatus for handling point cloud data |
| CN107871129A (en) * | 2016-09-27 | 2018-04-03 | 北京百度网讯科技有限公司 | Method and apparatus for handling cloud data |
| CN106485233A (en) * | 2016-10-21 | 2017-03-08 | 深圳地平线机器人科技有限公司 | Drivable region detection method, device and electronic equipment |
| CN106503653A (en) * | 2016-10-21 | 2017-03-15 | 深圳地平线机器人科技有限公司 | Area marking method, device and electronic equipment |
| CN106485233B (en) * | 2016-10-21 | 2020-01-17 | 深圳地平线机器人科技有限公司 | Method and device for detecting travelable area and electronic equipment |
| CN106778548A (en) * | 2016-11-30 | 2017-05-31 | 百度在线网络技术（北京）有限公司 | Method and apparatus for detecting barrier |
| CN106778548B (en) * | 2016-11-30 | 2021-04-09 | 百度在线网络技术（北京）有限公司 | Method and apparatus for detecting obstacles |
| CN106845412A (en) * | 2017-01-20 | 2017-06-13 | 百度在线网络技术（北京）有限公司 | Obstacle recognition method and device, computer equipment and computer-readable recording medium |
| CN106845412B (en) * | 2017-01-20 | 2020-07-10 | 百度在线网络技术（北京）有限公司 | Obstacle identification method and device, computer equipment and readable medium |
| US10803326B2 (en) | 2017-03-13 | 2020-10-13 | Baidu Online Network Technology (Beijing) Co., Ltd. | Obstacle detecting method and apparatus, device and storage medium |
| CN106951847B (en) * | 2017-03-13 | 2020-09-29 | 百度在线网络技术（北京）有限公司 | Obstacle detection method, apparatus, device and storage medium |
| CN106951847A (en) * | 2017-03-13 | 2017-07-14 | 百度在线网络技术（北京）有限公司 | Obstacle detection method, device, equipment and storage medium |
| CN108734780A (en) * | 2017-04-17 | 2018-11-02 | 百度在线网络技术（北京）有限公司 | Method, apparatus and equipment for generating map |
| CN109101861A (en) * | 2017-06-20 | 2018-12-28 | 百度在线网络技术（北京）有限公司 | Obstacle identity recognition methods, device, equipment and storage medium |
| US11017244B2 (en) | 2017-06-20 | 2021-05-25 | Baidu Online Network Technology (Beijing) Co., Ltd. | Obstacle type recognizing method and apparatus, device and storage medium |
| CN107907886A (en) * | 2017-11-07 | 2018-04-13 | 广东欧珀移动通信有限公司 | Travel conditions recognition methods, device, storage medium and terminal device |
| CN108280886A (en) * | 2018-01-25 | 2018-07-13 | 北京小马智行科技有限公司 | Laser point cloud mask method, device and readable storage medium storing program for executing |
| CN108596223A (en) * | 2018-04-11 | 2018-09-28 | 珠海博明视觉科技有限公司 | A method of automatically generating object data set |
| CN108630015A (en) * | 2018-05-21 | 2018-10-09 | 浙江吉利汽车研究院有限公司 | A kind of driving warning method, device and electronic equipment |
| CN110858415A (en) * | 2018-08-24 | 2020-03-03 | 北京图森未来科技有限公司 | Method and device for labeling object in 3D point cloud data |
| CN109215135A (en) * | 2018-09-07 | 2019-01-15 | 百度在线网络技术（北京）有限公司 | A kind of Obstacle Position emulation mode, device and terminal based on statistics |
| US11205289B2 (en) | 2018-09-07 | 2021-12-21 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method, device and terminal for data augmentation |
| CN109215135B (en) * | 2018-09-07 | 2019-11-08 | 百度在线网络技术（北京）有限公司 | A kind of Obstacle Position emulation mode, device and terminal based on statistics |
| US11354461B2 (en) | 2018-09-07 | 2022-06-07 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method and device for simulating a distribution of obstacles |
| US11307302B2 (en) | 2018-09-07 | 2022-04-19 | Baidu Online Network Technology (Beijing) Co., Ltd | Method and device for estimating an absolute velocity of an obstacle, and non-volatile computer-readable storage medium |
| US11276243B2 (en) | 2018-09-07 | 2022-03-15 | Baidu Online Network Technology (Beijing) Co., Ltd. | Traffic simulation method, device and storage medium |
| US10984588B2 (en) | 2018-09-07 | 2021-04-20 | Baidu Online Network Technology (Beijing) Co., Ltd | Obstacle distribution simulation method and device based on multiple models, and storage medium |
| EP3621040A1 (en) * | 2018-09-07 | 2020-03-11 | Baidu Online Network Technology (Beijing) Co., Ltd. | Data augmentation method, device and terminal |
| US11047673B2 (en) | 2018-09-11 | 2021-06-29 | Baidu Online Network Technology (Beijing) Co., Ltd | Method, device, apparatus and storage medium for detecting a height of an obstacle |
| CN109444915A (en) * | 2018-09-11 | 2019-03-08 | 成都优艾维智能科技有限责任公司 | A kind of danger zone pre-judging method based on laser radar data |
| CN109444915B (en) * | 2018-09-11 | 2020-11-03 | 成都优艾维智能科技有限责任公司 | Dangerous area pre-judging method based on laser radar data |
| US11519715B2 (en) | 2018-09-11 | 2022-12-06 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method, device, apparatus and storage medium for detecting a height of an obstacle |
| US11126875B2 (en) | 2018-09-13 | 2021-09-21 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method and device of multi-focal sensing of an obstacle and non-volatile computer-readable storage medium |
| CN109376664A (en) * | 2018-10-29 | 2019-02-22 | 百度在线网络技术（北京）有限公司 | Machine learning training method, device, server and medium |
| CN109344804A (en) * | 2018-10-30 | 2019-02-15 | 百度在线网络技术（北京）有限公司 | A kind of recognition methods of laser point cloud data, device, equipment and medium |
| CN109711285A (en) * | 2018-12-11 | 2019-05-03 | 百度在线网络技术（北京）有限公司 | Training, test method and the device of identification model |
| CN109709961A (en) * | 2018-12-28 | 2019-05-03 | 百度在线网络技术（北京）有限公司 | Road barricade object detecting method, device and autonomous driving vehicle |
| CN109709961B (en) * | 2018-12-28 | 2021-12-07 | 百度在线网络技术（北京）有限公司 | Road obstacle detection method and device and automatic driving automobile |
| CN109858374A (en) * | 2018-12-31 | 2019-06-07 | 武汉中海庭数据技术有限公司 | Arrow class graticule extraction method and device in high-precision cartography |
| US11780463B2 (en) | 2019-02-19 | 2023-10-10 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method, apparatus and server for real-time learning of travelling strategy of driverless vehicle |
| US11718318B2 (en) | 2019-02-22 | 2023-08-08 | Apollo Intelligent Driving (Beijing) Technology Co., Ltd. | Method and apparatus for planning speed of autonomous vehicle, and storage medium |
| CN109934141B (en) * | 2019-03-01 | 2021-05-04 | 北京百度网讯科技有限公司 | Method and device for marking data |
| CN109934141A (en) * | 2019-03-01 | 2019-06-25 | 北京百度网讯科技有限公司 | Method and apparatus for labeled data |
| CN110379157A (en) * | 2019-06-04 | 2019-10-25 | 深圳市速腾聚创科技有限公司 | Road blind area monitoring method, system, device, equipment and storage medium |
| CN112116804A (en) * | 2019-06-19 | 2020-12-22 | 北京地平线机器人技术研发有限公司 | Vehicle state quantity information determination method and device |
| CN110807439A (en) * | 2019-11-12 | 2020-02-18 | 银河水滴科技（北京）有限公司 | Method and device for detecting obstacle |
| CN110866504A (en) * | 2019-11-20 | 2020-03-06 | 北京百度网讯科技有限公司 | Method, device and equipment for acquiring marked data |
| CN110866504B (en) * | 2019-11-20 | 2023-10-17 | 北京百度网讯科技有限公司 | Method, device and equipment for acquiring annotation data |
| WO2021213241A1 (en) * | 2020-04-20 | 2021-10-28 | 上海商汤临港智能科技有限公司 | Target detection method and apparatus, and electronic device, storage medium and program |
| CN112269379A (en) * | 2020-10-14 | 2021-01-26 | 北京石头世纪科技股份有限公司 | Obstacle identification information feedback method |
| CN112269379B (en) * | 2020-10-14 | 2024-02-27 | 北京石头创新科技有限公司 | Obstacle identification information feedback method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105957145A (en) | Road barrier identification method and device | |
| CN108121984B (en) | Character recognition method and device | |
| CN107025642B (en) | Vehicle's contour detection method and device based on point cloud data | |
| CN106997466A (en) | Method and apparatus for detecting road | |
| CN105740402A (en) | Method and device for acquiring semantic labels of digital images | |
| US11003926B2 (en) | Method and apparatus for recognizing boundary of traffic sign | |
| CN108197623A (en) | For detecting the method and apparatus of target | |
| CN105973245A (en) | Method and device for updating online map by using unmanned vehicle | |
| CN107545000A (en) | The information-pushing method and device of knowledge based collection of illustrative plates | |
| CN110163208B (en) | Scene character detection method and system based on deep learning | |
| CN107894237A (en) | Method and apparatus for showing navigation information | |
| CN108052920A (en) | For the method and apparatus of output information | |
| CN107941226A (en) | Method and apparatus for the direction guide line for generating vehicle | |
| CN106781507A (en) | Method and apparatus for generating traffic information and test road conditions identifying system | |
| CN112396032B (en) | Writing detection method and device, storage medium and electronic equipment | |
| CN107830869A (en) | Information output method and device for vehicle | |
| CN108133197A (en) | For generating the method and apparatus of information | |
| CN108170751A (en) | For handling the method and apparatus of image | |
| CN107527074A (en) | Image processing method and device for vehicle | |
| CN109903308A (en) | For obtaining the method and device of information | |
| CN108280489A (en) | A kind of sample data generation method and device | |
| CN103854020A (en) | Character recognition method and device | |
| CN110097600A (en) | The method and device of traffic mark board for identification | |
| CN110321854A (en) | Method and apparatus for detected target object | |
| CN107084728A (en) | Method and apparatus for detecting numerical map |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160921 |
|
| RJ01 | Rejection of invention patent application after publication |