CN106570454A - Pedestrian traffic parameter extraction method based on mobile laser scanning - Google Patents
Pedestrian traffic parameter extraction method based on mobile laser scanning Download PDFInfo
- Publication number
- CN106570454A CN106570454A CN201610883070.5A CN201610883070A CN106570454A CN 106570454 A CN106570454 A CN 106570454A CN 201610883070 A CN201610883070 A CN 201610883070A CN 106570454 A CN106570454 A CN 106570454A
- Authority
- CN
- China
- Prior art keywords
- pedestrian
- data
- target
- traffic parameter
- vehicle
- 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
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
Abstract
The invention belongs to a pedestrian traffic parameter extraction method based on mobile laser scanning, and belongs to the technical field of microscopic traffic parameter obtaining. The method comprises the steps: collecting surrounding point cloud of a vehicle through a three-dimensional laser scanner installed on the vehicle; obtaining different types of data objects through data segmentation, clustering, and detection; recognizing a pedestrian, and extracting fixed objects on a road; carrying out the matching of fixed ground objects in two data frames; calculating the movement distance of an autonomous vehicle in the two frames; calculating the speed of a vehicle; and finally carrying out the statistics of traffic parameters of a pedestrian flow. The method is specifically characterized in that a VLP-16 small-size laser radar from the Velodyne company is used for collecting the dynamic data of pedestrians on a road and the surroundings, and the cost is low; for the correlation of target data and the extraction of traffic parameters, the method employs a nearest neighbor method with reflection information in order to obtain the accurate matching information in the multiframe data, and solves a problem of target matching well.
Description
Technical field
The present invention relates to microcosmic traffic parameter acquiring technical field, show in particular a kind of row based on mobile laser scanning
People's traffic parameter extracting method.
Background technology
Because sense of traffic is thin, the vehicle accident that non-motor vehicle, pedestrian's reason are caused has become China's vehicle accident
Pith.Therefore, it is that driver's development seems particularly heavy to non-motor vehicle, the initiative recognition of pedestrian and active safety early warning
Will.Laser scanner technique has obtained research with application in many aspects such as City Modeling, vegetational analysises.The present invention is based on
The line laser scanners of Velodyne 16, detect various inhomogeneities such as motor vehicles, non-motor vehicle, pedestrian, the atural object of vehicle periphery
The data of type, by data segmentation, classification, respectively obtain different classes of data object.And by taking pedestrian's object as an example, with reference to road
Circuit-switched data, analyzes the traffic behavior of the class object, obtains corresponding microscopic traffic flow parameter.
Obtain microscopic traffic flow parameter to the decision-making of Autonomous Vehicles and vehicle accident avoid have great significance.And it is effective
Ground identifies pedestrian and non motorized vehicle using the data that sensor is obtained, and the microscopic traffic flow parameter for extracting target is research
One of difficult point and the obstruction of unmanned Autonomous Vehicles popularization.One reason for this is that the sensor obtained needed for precise information is very high
It is expensive, and need Multi-sensor Fusion.
The present invention utilize 16 relatively inexpensive line laser radars, complete under single sensor pedestrian target recognition and
The identification of passageway tree, and then the speed of vehicle and the speed of pedestrian are calculated, and then analyze the traffic behavior of pedestrian.Using patrilineal line of descent with only one son in each generation sense
Device can just complete the difficult point and significance that the target is the present invention.
In terms of Target detection and identification, various different methods are had according to different sensor used.Main flow now
Sensor has monocular camera, many mesh cameras and laser radar.According to all of sensor, there is the target based on monocular vision to examine
Survey, the target detection based on stereoscopic vision and the target detection combined based on laser radar and vision.Due to of the invention main
It is based on the target detection of laser radar, so main below introduce based on the target detection of laser radar.
In the target detection based on laser radar, the radar for using has single line radar, multi-thread radar and three-dimensional omnidirectional thunder
Reach.Wherein single line radar and multi-thread radar can only do the simple functions such as detection of obstacles, and three-dimensional omnidirectional radar can obtain richer
Rich, more comprehensively and more accurately environmental information, is widely used in now in the research of unmanned vehicle.
Institutes Of Technology Of Tianjin sand moral roc et al. have studied vehicle laser velocity-measuring system, and using laser the survey to automobile is realized
Speed.Do based on sharp because laser radar serves important function, Central South University Zhou Zhi et al. in obstacle detection and identification
The intelligent vehicle active safety algorithm and model research of optical radar, using laser radar data front vehicles are recognized, and obtain vehicle
Speed information, judge whether vehicle in a safe condition with reference to the knowledge in terms of kinesiology.Liu university of the National University of Defense technology
Et al. done the research of multi-line laser radar country obstacle detection, adopted using four line laser radars on Autonomous Vehicles in research
Collection data.Liu great Xue et al. realizes the classification of data in scene；Factor to affecting radar ranging accuracy is analyzed, and is given
The filtering method of radar data is gone out；With relative altitude, the gradient and dot density are completed in country and hindered as judgment condition
Hinder the identification of thing.
In real-time detection and identification, the Cheng Jian of Zhejiang University realizes real-time mesh using the line laser radars of velodyne 64
Mark detection.For target classification, pedestrian is judged according to simple geometry feature in the barrier of non-car.In pedestrian detection side
Face, Navarro-Serment et al. propose to be based on high-precision three-dimensional laser radar, by impact point cloud mass according to two lower limbs and body point
Into three parts, point cloud feature, such as covariance matrix, inertial tensor matrix etc. are then extracted in each section, reach standard
True pedestrian detection.3D Lidar are come pedestrian detection for Luciano Spinello et al. propositions, are layered according to cluster tile height,
Geometric properties are extracted in each layer, statistical nature, finally by machine learning method, obtains grader.Kiyosumi kidono
Et al. propose two kinds of novelties feature：A kind of length and width of the hierarchy slicing projection for being characterized in that cluster block representing pedestrian Lun Temple,
Pedestrian's body width of differing heights level is different, and with the feature description of hierarchy slicing the outline of pedestrian can be preferably described；
Another kind is characterized in that the distribution of the point cloud reflex strength of the tested cluster block of statistics, comes stable finally by SVM training graders
Pedestrian detection.
Scholar has done many researchs with regard to pedestrian detection and vehicle driving safety based on laser radar, it is proposed that Hen Duofu
Close using the algorithm of scene, there are many reference values to the present invention.But many laser treatment data methods are all based on specific
Laser radar data, the present invention need to based on Velodyne companies VLP-16 laser radar datas, propose in specific field
The solution that scape is used.
The content of the invention
Goal of the invention
Present invention is primarily targeted at providing a kind of pedestrian traffic parameter extracting method based on mobile laser scanning, adopt
Realize the traffic flow parameter analysis of pedestrian with VLP with low cost, solve currently without realize between adjacent two frame shade tree it
Between association and the further method of the traffic flow parameter such as calculating speed.
Technical scheme
A kind of pedestrian traffic parameter extracting method based on mobile laser scanning, it is characterised in that include：Target detection,
VELOCITY EXTRACTION and traffic parameter statistics, wherein target detection are divided into sets up grating map, target label and object filtering, concrete step
It is rapid as follows：
S1. grating map is set up：Based on the three-dimensional laser data for collecting, determine that method is true by the barrier based on distance
The scope and size of fixed grid lattice；
S2. target label：Target label is divided into pedestrian and trees, using pedestrian's relation mark row poor for ground level
People, using trees are straight and upright and the characteristics of highly significant, according to target discrepancy in elevation labelling trees；
S3. object filtering：After completing grating map mark, by morphological dilation and the cluster side of zone marker
Method, expands to target grid, after completing to expand and cluster, destination object here is screened, and rejecting does not meet the object of feature, this
In screening be mainly to pedestrian's Object Operations；
S4. after completing target acquisition screening, data association is carried out to target, data association part is believed using reflection is added
The nearest neighbor method of breath completes data association, subsequently extracts the speed of data acquisition vehicle；
S5. data acquisition vehicle and pedestrian have been obtained after target detection and target data association this sequence of operations
Speed, counts traffic parameter of the pedestrian relative to collection vehicle.
Preferably, S2 is specific as follows：After dyspoiesis thing grating map, obstacle tag out, after labelling is complete, first
The data of barrier are labeled as in traversal barrier grating map, these data are calculated and judged, if meeting pedestrian
Feature be just labeled as pedestrian, if the feature for meeting shade tree is just labeled as trees, at two when labelling pedestrian and trees
Labelling on barrier grating map.
Preferably, S3 is specific as follows：Using the ratio of width to height constraints, from the point cloud of labelling pedestrian is identified, it is to avoid
The interference of other barriers of the roadside such as street lamp, refuse receptacle.
Preferably, S4 data association concrete steps：(1) spy for extracting trees target (2) the calculating target that detection is completed
Levy point (3) carries out next frame association using the nearest neighbor algorithm for adding reflective information.
Preferably, S5 is specific as follows：Using the principle that shade tree is motionless, according to zero relative shade tree, pedestrian
Change calculate car speed and pedestrian's speed, so as to calculate the microscopic traffic flow parameter of pedestrian.
Concrete advantage is as follows：
1) pedestrian and surrounding are gathered in campus road using the VLP-16 Compact Laser Radars of Velodyne companies
Dynamic data, it is cheap；
2) for the detection of pedestrian and shade tree, the process of laser point cloud data for convenience, using grating map labelling
Barrier, labelling pedestrian and shade tree on the basis of barrier.Finally by expansion and cluster, extract complete pedestrian and row
Road tree laser point cloud data；
3) for the association and traffic parameter extraction of target data, in order to be able to obtain correct matching letter in many frame data
Breath, the present invention solves the problems, such as well object matching using the nearest neighbor method for adding reflective information.Using kinesiology formula,
The information such as the speed of vehicle and pedestrian are obtained, microcosmic traffic parameter is counted.
Figure of description
Fig. 1 is the trees based on grating map and pedestrian's labelling flow chart；
Fig. 2 is pedestrian's laser point cloud shape graph at 10 meters of distances of VLP16 sensors acquisition；
Fig. 3 is pedestrian's laser point cloud shape graph at 5 meters of distances of VLP16 sensors acquisition；
Fig. 4 is pedestrian's typical shape of VLP16 collections；
Fig. 5 is the shade tree typical shape of VLP16 collections.
Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present application, the technical scheme in the embodiment of the present application is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only some embodiments of the present application, rather than the embodiment of whole.It is based on
Embodiment in the application, it is every other that those of ordinary skill in the art are obtained under the premise of creative work is not made
Embodiment, belongs to the scope of the application protection.
The features such as laser radar data has high precision and big data volume.VLP-16 laser radars produce more than 120,000 each second
Point cloud.Therefore unsuitable for the directly process on initial data, in terms of target detection, data are divided using grating map fine
Solution this problem.Grid detection has simple and quick, and stability is preferable.As shown in figure 1, the present invention is used based on grid
The object detection method of lattice map, is that each object sets up a grating map, facilitates follow-up identification and relevant parameter extraction.
Based on the object detection method of grating map, using discrete two-dimensional grid environment is divided.With feature based
Target detection is compared, and grating map need not assume the geometric model of target, can detect polytype target, such as pedestrian, from
Driving, automobile etc..The present invention will be described in detail the labelling of the ground map generalization of barrier grid and destination object, primarily determine that grid ground
The object properties of cell in figure.The Complete three-dimensional cloud data of destination object is obtained followed by expansion and cluster, finally
Carry out the screening of destination object.
S1. grating map is set up
Grating map by calculating each cell in the probit comprising object expressing surrounding.Often
The state of individual cell is " blank " or " occupying " one of both, or is fallen between, and is unknown state.Set up grating map
Key problem, that is, calculate the posterior probability of map using data-oriented.I.e. how using data-oriented estimation unit lattice
Quantity of state, different grating maps can set different estimation scheme, tell different destination objects.
Continuous space is divided into equally distributed grid by grating map.Grating map the most frequently used at present is horizontal layout
Figure, that is, represent three-dimensional world with two-dimensional section figure.The present invention projects to three-dimensional point cloud in two-dimensional grid map, reduces
The quantity and amount of calculation of cell, it is single accuracy have been affected simultaneously.If generating three-dimensional grating map, grid list
The quantity of unit can increase several times.And the laser radar that the present invention is used is 16 line radars, when target compares from zero
When remote, point cloud is very sparse, generates three-dimensional grating map and does not improve a lot to improving degree of accuracy.Therefore using two dimension
Grid map.
In view of the visual range of laser radar, 20 meters after setting grating map size as Herba Plantaginis and car, 10 meters of both sides, grid
Lattice size is 20cm*20cm, that is, set up the grating map of a 200*100.
Example：Barrier grating map
The angle of vehicle is set out, and artificial abortion, non-motor vehicle and trees are all one kind of barrier.Barrier map is first generated,
Subsequently according to the signature grid of destination object.Grating map is by three-dimensional point discretization, according to the data of each cell
Estimation unit trellis state.But when target is dispersed to multiple cells, because data lose integrity, target recognition will
Can be very inaccurate.The present invention proposes the zone marker method a kind of grating map by barrier, improves the complete of cluster
Property.
Grating map is generated to barrier, it is necessary first to determine that what is only barrier, subsequently could determining unit lattice category
Property.One of barrier is mainly characterized by the projection for having certain altitude than ground, therefore can be with the point cloud in Traversal Unit lattice, such as
The density and height of fruit dot cloud meets certain requirements, so that it may think there is barrier in this Unit Cell.In Traversal Unit lattice
Three-dimensional point cloud, statistics obtains Max_Z, Min_Z and number of scanning lines n in cell, obtains relative maximum height Δ Z.If Δ Z is more than
Threshold xi and number of scanning lines are more than threshold value λ, and with regard to cell barrier point is labeled as.It is this that obstacle is determined according to relative altitude and density
The method of point, can effectively reduce erroneous judgement, but it is also possible that increasing the risk failed to judge.Because angle interval between scan line
It is fixed, scanning property is more sparse in the more remote target of laser origin, and possible distant place barrier only has little several to sweep
Retouch line scanning to arrive.
S2. grid object attribute labelling
Using labelling pedestrian and shade tree in multi-thread laser scanning data, and provide the experience ginseng of pedestrian, shade tree labelling
Number.
Example：
Pedestrian's labelling
Before labelling pedestrian's cell attribute, it is to be understood that the three-dimensional point cloud geometric shape of pedestrian.Select respectively pedestrian with
Cloud data when lasing central horizontal range is 10m and 5m is analyzed.Pedestrian as seen from Figure 2, when pedestrian from
During launching centre horizontal range 10m.There was only about 6 scan lines with pedestrian.It is humanoid that can find out dimly from cloud data,
Therefore we only detect the pedestrian within 10m.As shown in figure 3, the shape of people can be clearly seen that during horizontal range 5m, on the person
There are about 10 scan lines.In the same manner pedestrian can be marked according to the geometric properties of the bar number of scan line and pedestrian.
Because laser radar is erected on roof, and laser radar vertical field of view angle only has 30 °, so laser radar is near
Can there is a data acquisition blind area in place.When pedestrian from vehicle too close to when pedestrian's body scan cannot be arrived, as shown in Figure 3.
From fig. 4 it can be seen that when pedestrian from far point horizontal range be 3.5 meters when, pedestrian only has more than half body to be swashed
Photoscanner scanning is arrived, and now scan line compares comparatively dense.Now pedestrian's geometric properties of the geometric properties and 10m of pedestrian and 5m
It is different, it is impossible to continue to use feature criterion above.According to the erection situation of laser radar, when horizontal range is less than 4 meters,
Complete pedestrian's point cloud cannot be scanned.
Laser radar blind area is avoided, the complete pedestrian of scanning in policy tag 4m a to 10m is first worked out.According to above institute
The barrier grating map generating principle stated, is that a tightened up specific threshold standard is set up in the generation of pedestrian's grating map.Tool
Body way be traversal barrier grating map, then to being labeled as the cell of barrier in all cloud datas travel through once,
Its discrepancy in elevation and number of scanning lines are counted, cell attribute is re-flagged according to the threshold value for resetting.Enter the contrast of excessive frame data
Test, is given good threshold value standard is showed in experimental data here：Maximum discrepancy in elevation Δ Z is less than 2 meters more than 0.9 meter, scan line
Number is less than 12 more than or equal to 5.
In order to be able to be tagged to that incomplete pedestrian is scanned in laser radar blind area, another strategy detection trip need to be specified
People's target.It can be seen in figure 3 that scanning the pedestrian's shape come only has half body, therefore discrepancy in elevation standard can be relaxed, if
It is scheduled in 4 meters of horizontal range, target maximum discrepancy in elevation Δ Z is less than 1.5 meters more than 0.4 meter, and scan line is less than 11 more than or equal to 6.
Shade tree labelling
According to the method for pedestrian detection, it is equally applicable to detect shade tree, only needs to change threshold value.In given threshold
Before it should be understood that the cloud data of trees.The cloud data of trees is extracted by hand first, using Matlab Software on Drawing point
Cloud, as shown in Figure 5.
From fig. 5, it is seen that trunk height is both greater than 2 meters, and there is multi-strip scanning line to sweep to.In the same manner according to trees
Geometric properties change marking-threshold.The labelling strategies of shade tree are：Maximum discrepancy in elevation Δ Z is more than 2 meters, and scan line is more than 9.Due to
Abundanter apart from the nearlyer tree information of vehicle in order to calculate data acquisition vehicle speed when shade tree detects, setting here is compared
Strict threshold value, retains remote missing inspection with a distance from the near woods permission of vehicle.
The obstacle tag parameter of table 1
The pedestrian's flag parameters of table 2
The shade tree flag parameters of table 3
S3. object filtering
Using the ratio of width to height constraints, from the point cloud of labelling pedestrian is identified, it is to avoid the roadside such as street lamp, refuse receptacle other
The interference of barrier.
Pedestrian target is screened
By sequence of operations above, although destination object is tentatively marked, there is substantial amounts of flase drop, its
Middle most of flase drop object is trees, it is therefore desirable to which target is further screened, and reservation meets the object of feature, rejects flase drop
Object.
The present invention is shone again based on depth-width ratio feature and selects the pedestrian target to be obtained.According to《Chinese adult human body
Size (GB10000-88)》, normal person's depth-width ratio is about between 4 to 5.In view of eclipse phenomena, therefore depth-width ratio threshold value is set
Put between 3 to 7.After adding depth-width ratio constraint, only individual other flase drop point and missing inspection point.
According to experimental data, the depth-width ratio of trees is generally more than 9 in data, therefore can pick in pedestrian target well
Except trees target.And in trees detection, tree features are fairly obvious, and false drop rate is than relatively low.
Pedestrian and surface constraints
Can obtain, when pedestrian is in 4m and 10m, being generally possible to scanning to complete pedestrian's shape data according to above analysis,
Pedestrian always walks on the ground.The target similar to pedestrian's geometric properties can be caused because blocking according to this feature rejecting.
Target in 4m cannot can only collect pedestrian's half body using this screening constraints when near apart from collection vehicle
Picture, so should now retain " suspension " skyborne target.
S4. target data association and traffic parameter are extracted
Data acquisition vehicle and pedestrian are movable bodies, and vehicle and pedestrian can be relatively fixed atural object movement.In order to obtain car
And pedestrian's velocity information, it is necessary to before and after frame is carried out to fixed atural object and pedestrian this two classes data and is associated, calculate vehicle and row
The traffic parameters such as the speed of people.The process of this two type games target includes the data association and fortune of frame before and after the step of two keys
The estimation of dynamic rail mark.For moving object, sensor just can not possibly can obtain its kinestate according to an only frame data, number
Need to recognize moving object and make condition adjudgement according to continuous multiple frames sensing data according to processing module.
Through being tracked to target within a period of time to target data association, then extracted according to kinesiology
The speed consecutive variations figure of data acquisition vehicle, by the relativeness of data acquisition vehicle and pedestrian the traffic row of artificial abortion is counted
For.
Data association concrete steps：
(1) extract and detect that the trees target (2) for completing calculates clarification of objective point (3) and uses the most adjacent of addition reflective information
Nearly method carries out next frame association
S5 obtains artificial abortion's traffic parameter
Using the principle that shade tree is motionless, according to the change of the relative shade tree of zero, pedestrian calculate car speed and
Pedestrian's speed, so as to calculate the microscopic traffic flow parameter of pedestrian.
S5.1 coordinate systems and kinematic principle
Laser data is collected after conversion, is the Z axis Vertical Launch plane with launching centre as zero.Coordinate
The origin of system is the launching centre of laser radar all the time, as the kinetic coordinate system origin of vehicle also can be with motion.
Only a frame data can not obtain the movable information of this moving target of coordinate origin, need and two frame data in front and back
It is associated the direction of motion for calculating zero.With ground as motion reference system, zero is still relative to regularly
Thing motion.With instrument as motion reference system, fixed atural object is relative to instrument motion.Therefore fixed atural object is in Qian Hou two frame
The change of coordinate is exactly the change of zero motion.
S5.2 calculates car speed
All frames are all carried out after data association, and using kinematic principle the instantaneous velocity at each moment is obtained.
Can see, automobile is to have been started up when start recording data having one section to add according to velocity variations are calculated
The stage of speed, subsequent period speed has several minor swings, and the stage reduces speed.Extremely meet with the situation of gathered data, into reality
Automobile starts to accelerate behind road inspection road, and road midway runs into artificial abortion so being slowed down, and has finally arrived straight way end automobile and has started
Slow down.
S5.3 pedestrian traffic parameters
According to previously described target detection, pedestrian target can be detected, calculate the data association of two frames according to before and after, together
Sample can also calculate the speed of pedestrian.After completing these operations, the traffic parameter of artificial abortion can be counted with that.Every 10 frames
The microcosmic traffic parameter of statistics artificial abortion, parameter list such as table 4 below institute.
The artificial abortion's traffic parameter of table 4
Claims (5)
1. it is a kind of based on the pedestrian traffic parameter extracting method for moving laser scanning, it is characterised in that to include：Target detection, speed
Degree extraction and traffic parameter statistics, wherein target detection are divided into sets up grating map, target label and object filtering, concrete steps
It is as follows：
S1. grating map is set up：Based on the three-dimensional laser data for collecting, determine that method determines grid by the barrier based on distance
The scope and size of lattice；
S2. target label：Target label is divided into pedestrian and trees, using the pedestrian relation mark pedestrian poor for ground level,
Using trees are straight and upright and the characteristics of highly significant, according to target discrepancy in elevation labelling trees；
S3. object filtering：It is right by morphological dilation and the clustering method of zone marker after completing grating map mark
Target grid expands, and after completing to expand and cluster, destination object here is screened, and rejecting does not meet the object of feature, here
Screening is mainly to pedestrian's Object Operations；
S4. after completing target acquisition screening, data association, data association part, using addition reflective information are carried out to target
Nearest neighbor method completes data association, subsequently extracts the speed of data acquisition vehicle；
S5. the speed of data acquisition vehicle and pedestrian has been obtained after target detection and target data association this sequence of operations
Degree, counts traffic parameter of the pedestrian relative to collection vehicle.
2. a kind of based on the pedestrian traffic parameter extracting method for moving laser scanning according to right wants 1, S2 is specific as follows：
After dyspoiesis thing grating map, after labelling is complete, first travels through and barrier is labeled as in barrier grating map obstacle tag out
Hinder the data of thing, these data are calculated and judged, pedestrian is labeled as if the feature for meeting pedestrian, if meeting
The feature of shade tree is just labeled as trees, when labelling pedestrian and trees on two barrier grating maps labelling.
3. a kind of based on the pedestrian traffic parameter extracting method for moving laser scanning according to right wants 1 or 2, its feature exists
In S3 is specific as follows：Using the ratio of width to height constraints, pedestrian is identified from the point cloud of labelling, it is to avoid the road such as street lamp, refuse receptacle
The interference of other other barriers.
4. a kind of based on the pedestrian traffic parameter extracting method for moving laser scanning according to right wants 3, it is characterised in that
S4 data association concrete steps：(1) extract and detect that the trees target (2) for completing calculates clarification of objective point (3) and uses addition anti-
Penetrating the nearest neighbor algorithm of information carries out next frame association.
5. a kind of based on the pedestrian traffic parameter extracting method for moving laser scanning according to right wants 4, it is characterised in that
S5 is specific as follows：Using the principle that shade tree is motionless, car speed is calculated according to the change of the relative shade tree of zero, pedestrian
With pedestrian's speed, so as to calculate the microscopic traffic flow parameter of pedestrian.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610883070.5A CN106570454B (en) | 2016-10-10 | 2016-10-10 | Pedestrian traffic parameter extracting method based on mobile laser scanning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610883070.5A CN106570454B (en) | 2016-10-10 | 2016-10-10 | Pedestrian traffic parameter extracting method based on mobile laser scanning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106570454A true CN106570454A (en) | 2017-04-19 |
| CN106570454B CN106570454B (en) | 2019-06-11 |
Family
ID=58532696
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610883070.5A Active CN106570454B (en) | 2016-10-10 | 2016-10-10 | Pedestrian traffic parameter extracting method based on mobile laser scanning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106570454B (en) |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108501954A (en) * | 2018-04-03 | 2018-09-07 | 北京瑞特森传感科技有限公司 | A kind of gesture identification method, device, automobile and storage medium |
| CN108717540A (en) * | 2018-08-03 | 2018-10-30 | 浙江梧斯源通信科技股份有限公司 | The method and device of pedestrian and vehicle are distinguished based on 2D laser radars |
| CN108763730A (en) * | 2018-05-24 | 2018-11-06 | 浙江农林大学 | Shade tree screening technique, system, terminal and medium based on thermal comfort index |
| CN108803602A (en) * | 2018-06-01 | 2018-11-13 | 浙江亚特电器有限公司 | Barrier self-learning method and new barrier self-learning method |
| CN108931773A (en) * | 2017-05-17 | 2018-12-04 | 通用汽车环球科技运作有限责任公司 | Automobile-used sextuple point cloud system |
| CN108985254A (en) * | 2018-08-01 | 2018-12-11 | 上海主线科技有限公司 | A kind of band based on laser hangs tag vehicle tracking |
| WO2019037129A1 (en) * | 2017-08-25 | 2019-02-28 | Beijing Didi Infinity Technology And Development Co., Ltd. | Methods and systems for detecting environmental information of a vehicle |
| CN109407115A (en) * | 2018-12-25 | 2019-03-01 | 中山大学 | A kind of road surface extraction system and its extracting method based on laser radar |
| CN109521756A (en) * | 2017-09-18 | 2019-03-26 | 百度在线网络技术（北京）有限公司 | Barrier motion information generation method and device for automatic driving vehicle |
| CN109753982A (en) * | 2017-11-07 | 2019-05-14 | 北京京东尚科信息技术有限公司 | Obstacle point detecting method, device and computer readable storage medium |
| CN109785632A (en) * | 2019-03-14 | 2019-05-21 | 济南浪潮高新科技投资发展有限公司 | A kind of magnitude of traffic flow statistical method and device |
| CN110161526A (en) * | 2019-05-24 | 2019-08-23 | 河南辉煌科技股份有限公司 | A kind of circuitry obstacle object recognition methods based on three-dimensional imaging |
| CN110309240A (en) * | 2018-03-14 | 2019-10-08 | 北京京东尚科信息技术有限公司 | The method and apparatus for removing dynamic object |
| CN110470308A (en) * | 2019-09-06 | 2019-11-19 | 北京云迹科技有限公司 | A kind of obstacle avoidance system and method |
| CN111007534A (en) * | 2019-11-19 | 2020-04-14 | 武汉光庭科技有限公司 | Obstacle detection method and system using sixteen-line laser radar |
| CN111033316A (en) * | 2017-08-18 | 2020-04-17 | 株式会社小糸制作所 | Recognition sensor, method for controlling recognition sensor, automobile, vehicle lamp, object recognition system, and object recognition method |
| CN111259722A (en) * | 2018-11-30 | 2020-06-09 | 株式会社小糸制作所 | In-vehicle object recognition system, automobile, vehicle lamp, classifier learning method, and arithmetic processing device |
| CN112102151A (en) * | 2020-07-27 | 2020-12-18 | 广州视源电子科技股份有限公司 | Grid map generation method and device, mobile intelligent device and storage medium |
| CN113030890A (en) * | 2019-12-24 | 2021-06-25 | 深圳市大富科技股份有限公司 | Target identification method and device based on vehicle-mounted radar |
| CN113343835A (en) * | 2021-06-02 | 2021-09-03 | 合肥泰瑞数创科技有限公司 | Object identification method and system suitable for emergency rescue and storage medium |
| CN113358112A (en) * | 2021-06-03 | 2021-09-07 | 北京超星未来科技有限公司 | Map construction method and laser inertia odometer |
| CN113859228A (en) * | 2020-06-30 | 2021-12-31 | 上海商汤智能科技有限公司 | Vehicle control method and device, electronic equipment and storage medium |
| WO2022017133A1 (en) * | 2020-07-22 | 2022-01-27 | 商汤集团有限公司 | Method and apparatus for processing point cloud data |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010157118A (en) * | 2008-12-26 | 2010-07-15 | Denso It Laboratory Inc | Pattern identification device and learning method for the same and computer program |
| CN102779280A (en) * | 2012-06-19 | 2012-11-14 | 武汉大学 | Traffic information extraction method based on laser sensor |
| CN103871234A (en) * | 2012-12-10 | 2014-06-18 | 中兴通讯股份有限公司 | Grid mapping growth-based traffic network division method and configuration server |
| CN104281746A (en) * | 2014-09-28 | 2015-01-14 | 同济大学 | Method for generating traffic safety road characteristic graphs on basis of point-cloud |
| CN106022460A (en) * | 2016-05-25 | 2016-10-12 | 重庆市勘测院 | Crowd density real-time monitoring method based on laser radar |
-
2016
- 2016-10-10 CN CN201610883070.5A patent/CN106570454B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010157118A (en) * | 2008-12-26 | 2010-07-15 | Denso It Laboratory Inc | Pattern identification device and learning method for the same and computer program |
| CN102779280A (en) * | 2012-06-19 | 2012-11-14 | 武汉大学 | Traffic information extraction method based on laser sensor |
| CN103871234A (en) * | 2012-12-10 | 2014-06-18 | 中兴通讯股份有限公司 | Grid mapping growth-based traffic network division method and configuration server |
| CN104281746A (en) * | 2014-09-28 | 2015-01-14 | 同济大学 | Method for generating traffic safety road characteristic graphs on basis of point-cloud |
| CN106022460A (en) * | 2016-05-25 | 2016-10-12 | 重庆市勘测院 | Crowd density real-time monitoring method based on laser radar |
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108931773A (en) * | 2017-05-17 | 2018-12-04 | 通用汽车环球科技运作有限责任公司 | Automobile-used sextuple point cloud system |
| CN111033316A (en) * | 2017-08-18 | 2020-04-17 | 株式会社小糸制作所 | Recognition sensor, method for controlling recognition sensor, automobile, vehicle lamp, object recognition system, and object recognition method |
| CN111033316B (en) * | 2017-08-18 | 2024-04-23 | 株式会社小糸制作所 | Identification sensor, control method therefor, automobile, vehicle lamp, object identification system, and object identification method |
| WO2019037129A1 (en) * | 2017-08-25 | 2019-02-28 | Beijing Didi Infinity Technology And Development Co., Ltd. | Methods and systems for detecting environmental information of a vehicle |
| US10824880B2 (en) | 2017-08-25 | 2020-11-03 | Beijing Voyager Technology Co., Ltd. | Methods and systems for detecting environmental information of a vehicle |
| CN109521756A (en) * | 2017-09-18 | 2019-03-26 | 百度在线网络技术（北京）有限公司 | Barrier motion information generation method and device for automatic driving vehicle |
| CN109521756B (en) * | 2017-09-18 | 2022-03-08 | 阿波罗智能技术(北京)有限公司 | Obstacle motion information generation method and apparatus for unmanned vehicle |
| CN109753982A (en) * | 2017-11-07 | 2019-05-14 | 北京京东尚科信息技术有限公司 | Obstacle point detecting method, device and computer readable storage medium |
| CN109753982B (en) * | 2017-11-07 | 2021-09-03 | 北京京东乾石科技有限公司 | Obstacle point detection method, obstacle point detection device, and computer-readable storage medium |
| CN110309240A (en) * | 2018-03-14 | 2019-10-08 | 北京京东尚科信息技术有限公司 | The method and apparatus for removing dynamic object |
| CN108501954A (en) * | 2018-04-03 | 2018-09-07 | 北京瑞特森传感科技有限公司 | A kind of gesture identification method, device, automobile and storage medium |
| CN108763730A (en) * | 2018-05-24 | 2018-11-06 | 浙江农林大学 | Shade tree screening technique, system, terminal and medium based on thermal comfort index |
| CN108803602A (en) * | 2018-06-01 | 2018-11-13 | 浙江亚特电器有限公司 | Barrier self-learning method and new barrier self-learning method |
| CN108985254A (en) * | 2018-08-01 | 2018-12-11 | 上海主线科技有限公司 | A kind of band based on laser hangs tag vehicle tracking |
| CN108717540B (en) * | 2018-08-03 | 2024-02-06 | 浙江梧斯源通信科技股份有限公司 | Method and device for distinguishing pedestrians and vehicles based on 2D laser radar |
| CN108717540A (en) * | 2018-08-03 | 2018-10-30 | 浙江梧斯源通信科技股份有限公司 | The method and device of pedestrian and vehicle are distinguished based on 2D laser radars |
| CN111259722A (en) * | 2018-11-30 | 2020-06-09 | 株式会社小糸制作所 | In-vehicle object recognition system, automobile, vehicle lamp, classifier learning method, and arithmetic processing device |
| CN109407115A (en) * | 2018-12-25 | 2019-03-01 | 中山大学 | A kind of road surface extraction system and its extracting method based on laser radar |
| CN109407115B (en) * | 2018-12-25 | 2022-12-27 | 中山大学 | Laser radar-based pavement extraction system and extraction method thereof |
| CN109785632B (en) * | 2019-03-14 | 2021-05-04 | 浪潮集团有限公司 | Traffic flow statistical method and device |
| CN109785632A (en) * | 2019-03-14 | 2019-05-21 | 济南浪潮高新科技投资发展有限公司 | A kind of magnitude of traffic flow statistical method and device |
| CN110161526A (en) * | 2019-05-24 | 2019-08-23 | 河南辉煌科技股份有限公司 | A kind of circuitry obstacle object recognition methods based on three-dimensional imaging |
| CN110470308A (en) * | 2019-09-06 | 2019-11-19 | 北京云迹科技有限公司 | A kind of obstacle avoidance system and method |
| CN111007534A (en) * | 2019-11-19 | 2020-04-14 | 武汉光庭科技有限公司 | Obstacle detection method and system using sixteen-line laser radar |
| CN113030890B (en) * | 2019-12-24 | 2023-11-21 | 大富科技(安徽)股份有限公司 | Target identification method and device based on vehicle-mounted radar |
| CN113030890A (en) * | 2019-12-24 | 2021-06-25 | 深圳市大富科技股份有限公司 | Target identification method and device based on vehicle-mounted radar |
| CN113859228A (en) * | 2020-06-30 | 2021-12-31 | 上海商汤智能科技有限公司 | Vehicle control method and device, electronic equipment and storage medium |
| WO2022017133A1 (en) * | 2020-07-22 | 2022-01-27 | 商汤集团有限公司 | Method and apparatus for processing point cloud data |
| CN112102151A (en) * | 2020-07-27 | 2020-12-18 | 广州视源电子科技股份有限公司 | Grid map generation method and device, mobile intelligent device and storage medium |
| CN113343835A (en) * | 2021-06-02 | 2021-09-03 | 合肥泰瑞数创科技有限公司 | Object identification method and system suitable for emergency rescue and storage medium |
| CN113358112A (en) * | 2021-06-03 | 2021-09-07 | 北京超星未来科技有限公司 | Map construction method and laser inertia odometer |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106570454B (en) | 2019-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106570454B (en) | Pedestrian traffic parameter extracting method based on mobile laser scanning | |
| CN102779280B (en) | Traffic information extraction method based on laser sensor | |
| Han et al. | Research on road environmental sense method of intelligent vehicle based on tracking check | |
| Serna et al. | Paris-rue-Madame database: A 3D mobile laser scanner dataset for benchmarking urban detection, segmentation and classification methods | |
| WO2022141911A1 (en) | Roadside sensing unit-based method for quick recognition of dynamic target point cloud and point cloud segmentation | |
| CN101975951B (en) | Field environment barrier detection method fusing distance and image information | |
| WO2018133851A1 (en) | Point cloud data processing method and apparatus, and computer storage medium | |
| CN110244321A (en) | A kind of road based on three-dimensional laser radar can traffic areas detection method | |
| CN113313154A (en) | Integrated multi-sensor integrated automatic driving intelligent sensing device | |
| CN106842231A (en) | A kind of road edge identification and tracking | |
| CN106199558A (en) | Barrier method for quick | |
| CN112389440B (en) | Vehicle driving risk prediction method in off-road environment based on vehicle-road action mechanism | |
| CN106683530A (en) | Computerized judging system and method based on three-dimensional laser vision and high-precision lane model | |
| CN110320504A (en) | A kind of unstructured road detection method based on laser radar point cloud statistics geometrical model | |
| CN109583415A (en) | A kind of traffic lights detection and recognition methods merged based on laser radar with video camera | |
| CN103902976A (en) | Pedestrian detection method based on infrared image | |
| CN110065494A (en) | A kind of vehicle collision avoidance method based on wheel detection | |
| Wang et al. | Automatic road extraction from mobile laser scanning data | |
| CN113345237A (en) | Lane-changing identification and prediction method, system, equipment and storage medium for extracting vehicle track by using roadside laser radar data | |
| CN108717540A (en) | The method and device of pedestrian and vehicle are distinguished based on 2D laser radars | |
| CN112379674B (en) | Automatic driving equipment and system | |
| Knoeppel et al. | Robust vehicle detection at large distance using low resolution cameras | |
| CN115605777A (en) | Dynamic target point cloud rapid identification and point cloud segmentation method based on road side sensing unit | |
| CN101425128A (en) | People stream detecting method based on laser sensor depth image | |
| CN109241855A (en) | Intelligent vehicle based on stereoscopic vision can travel area detection method |
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 |