CN107967804A - A kind of more rotors carry the vehicle cab recognition and vehicle speed measurement device and method of laser radar - Google Patents
A kind of more rotors carry the vehicle cab recognition and vehicle speed measurement device and method of laser radar Download PDFInfo
- Publication number
- CN107967804A CN107967804A CN201711260040.XA CN201711260040A CN107967804A CN 107967804 A CN107967804 A CN 107967804A CN 201711260040 A CN201711260040 A CN 201711260040A CN 107967804 A CN107967804 A CN 107967804A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- data
- laser radar
- point
- cab recognition
- 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
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/015—Detecting movement of traffic to be counted or controlled with provision for distinguishing between two or more types of vehicles, e.g. between motor-cars and cycles
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/04—Detecting movement of traffic to be counted or controlled using optical or ultrasonic detectors
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed
Abstract
The present invention relates to the vehicle cab recognition and vehicle speed measurement device and method that a kind of more rotors carry laser radar, belong to intelligent transportation system technical field.This method is acquired using UAV flight's laser radar road pavement information, laser radar acquisition hardware system is divided into airborne acquisition subsystem and ground control subsystem two parts, and airborne acquisition subsystem is made of battery, DC DC modules, laser radar, Raspberry Pi and five part of wireless data transfer module.Ground control subsystem is made of wireless data transfer module and computer two parts.Data interaction is carried out by medium of radio signal by wireless data transfer module between airborne acquisition subsystem and ground control subsystem.Since more rotors can promptly fly to the detection site specified, can meet the needs of for regions such as Emergent detections.
Description
Technical field
The present invention relates to the vehicle cab recognition and vehicle speed measurement device and method that a kind of more rotors carry laser radar, belong to intelligence
Traffic system technical field.
Background technology
With the fast development of China's economy, vehicle number quickly increases per capita, and the increase of vehicle is bringing people's traffic just
While sharp, huge pressure is also brought to intelligent transportation system.Intelligent transportation system be by it is advanced science and technology in full
It is effectively combined according to communication transfer technology, Electronic transducer technology, information technology and computer technology etc. and applies to whole ground
Traffic control system and establish it is a kind of in a wide range of, it is comprehensive play a role, in real time, accurately and efficiently comprehensive traffic
Transportation management system.And vehicle cab recognition and vehicle speed measurement are two most important parameter indexes of intelligent transportation system, for intelligence
Traffic system carries out vehicle flowrate, charge station administration etc. and provides important basic data, has important research significance.
Current model recognizing method mainly has ground induction coil detection, ultrasound examination, detection based on image procossing etc..
Vehicle speed measurement method has based on video image, buried induction coil, infrared detection etc..These methods are all to install sensor
There is the fixed position of certain distance on ground or from the ground, what the data by handling sensor carried out, there is technology application
Extensively, the advantages that high certainty of measurement, technique study more go deep into.But the vehicle that these methods are not suitable for emergent place is known
Other and vehicle speed measurement.Therefore, set forth herein using multi-rotor unmanned aerial vehicle to carry laser radar as acquisition platform, research is based on laser
The method that radar data completes vehicle cab recognition and vehicle speed measurement at the same time.Due to more rotors with can promptly flying to the detection specified
Point, therefore can meet the needs of for regions such as Emergent detections.
The content of the invention
The problem of the purpose of the present invention is not being suitable for emergent place for current vehicle cab recognition and vehicle speed measurement method, carry
A kind of vehicle cab recognition and vehicle speed measurement method that laser radar is carried based on more rotors is gone out.
The purpose of the present invention is what is be achieved through the following technical solutions:
A kind of more rotors carry the vehicle cab recognition and vehicle speed measurement device of laser radar, and laser radar is installed on unmanned plane
On, for airborne acquisition vehicle data, realize the identification to vehicle and the measurement of speed.
This method is acquired using UAV flight's laser radar road pavement information, laser radar acquisition hardware system point
For airborne acquisition subsystem and ground control subsystem two parts, airborne acquisition subsystem is by battery, DC-DC module, laser thunder
Reach, Raspberry Pi and five part of wireless data transfer module composition.Ground control subsystem is by wireless data transfer module and electricity
Brain two parts form.By wireless data transfer module with aerogram between airborne acquisition subsystem and ground control subsystem
Number for medium carry out data interaction.
A kind of more rotors carry the vehicle cab recognition and vehicle speed measurement method of laser radar, comprise the following steps that:
Step 1: data prediction:According to radar signal feature and the characteristic of more rotor platforms, first to radar data
Carry out the levelling of coordinate system conversion, the extraction of target information of road surface data, information of road surface data and radar site.Laser radar is adopted
The significant figure strong point collected is transformed into Descartes's rectangular coordinate system by polar coordinate system, obtains preprocessed features, so as to follow-up car
Type identifies and the use of vehicle speed measurement algorithm.
Step 2: vehicle cab recognition:Preprocessed features according to step 1 to lidar measurement data, to existing vehicle into
Row is classified again.The characteristic parameter of the height, width and compartment number of vehicle as vehicle cab recognition is chosen, according to the car being calculated
Type characteristic parameter numerical value, contrasts vehicle classification standard, exports vehicle cab recognition result.
Step 3: vehicle speed measurement:For the data of this method vehicle speed measurement using overall width, overall height as input, vehicle commander is output, because
The estimated value of vehicle commander allows there are error, therefore selects polynomial fitting method to establish model to given data.Intended according to multinomial
Conjunction method, establish by input, vehicle commander of overall width and overall height be output mathematical model, calculating vehicle commander and then according to car
The frame numbers of data calculates the time, and then obtains speed.
The step 2, vehicle cab recognition comprise the following steps that:
Step 1, choose the characteristic parameter of overall height, overall width and compartment number as identification vehicle;
Step 2, according to laser radar data feature classify vehicle again, establish using overall height, overall width and compartment number as
The vehicle classification standard of feature.
Step 3, extract the data for belonging to vehicle from the data of laser radar;
Vehicle is numbered in step 4, the sequencing for driving into according to vehicle laser radar scanning scope;
Step 5, three compartment number, overall height and overall width characteristic parameters for calculating vehicle in laser radar data;
Step 6, the overall height obtained according to step 5, the characteristic parameter of overall width and compartment number, i.e. vehicle cab recognition, with step 2 car
Type criteria for classification contrasts, and exports vehicle cab recognition result.
Beneficial effect
The present invention proposes to use multi-rotor unmanned aerial vehicle to carry laser radar as acquisition platform, studies and be based on laser radar data
The method for completing vehicle cab recognition and vehicle speed measurement at the same time.Since more rotors can promptly fly to the detection site specified,
Can meet the needs of for regions such as Emergent detections.
Brief description of the drawings
Fig. 1 is acquisition system composition frame chart;
Fig. 2 is vehicle cab recognition and vehicle speed measurement method flow chart;
Fig. 3 is hardware system for being electrically connected;
Fig. 4 is hardware communications connection figure;
Fig. 5 is laser radar initial data reconstruct image;
Fig. 6 is laser radar road acquisition state figure;
Fig. 7 chooses flow chart for vehicle characteristic parameter;
Fig. 8 is laser radar automobile transverse direction schematic diagram data;
Fig. 9 is laser radar automobile longitudinal schematic diagram data;
Figure 10 is two wagon overall height statistical charts;
Figure 11 is vehicle extraction algorithm flow chart;
Figure 12 extracts flow chart for target road data;
Figure 13 is the flow chart for extracting doubtful vehicle data;
Figure 14 is extraction number of vehicles flow chart;
Figure 15 is radar data file part vehicle schematic diagram;
Figure 16 is longitudinal direction of car schematic diagram, wherein (a) is three-box car longitudinal direction schematic diagram, (b) is two-box automobile longitudinal direction schematic diagram;
Vehicle data characteristic when Figure 17 is opposite radar different azimuth, wherein (a) is radar left side vehicle data, (b)
For vehicle data immediately below radar;
Figure 18 asks for overall width schematic diagram for radar left side vehicle.
Embodiment
The invention will be further described with embodiment below in conjunction with the accompanying drawings.
Embodiment 1
Using Matlab as software development environment, corresponding vehicle cab recognition and vehicle speed measurement processing software has been worked out.Institute
The case study on implementation of description is the part of the present invention.Based on the embodiments of the present invention, those of ordinary skill in the art are not having
All other embodiments obtained on the premise of creative work are made, belong to the scope of protection of the invention.
A kind of more rotors carry the vehicle cab recognition and vehicle speed measurement device of laser radar, and acquisition system frame diagram is as shown in Figure 1.
A kind of more rotors carry the vehicle cab recognition and vehicle speed measurement method of laser radar, as shown in Fig. 2, specific implementation step is such as
Under:
Step 1:Acquisition platform is built
This method acquisition platform carries laser radar for more rotors.
When hardware system is built firstly the need of the power demands for supplying electrical connection, ensureing each module for considering each several part.This
Method hardware system is as shown in Figure 3 for being electrically connected.Laser radar operating voltage 19.2-28.8V, and more rotors are polymerize using 6S lithiums
Thing battery, supply voltage 21.6-25.2V, it is possible to powered using flying platform power battery directly to laser radar.Tree
It is 5V/2A that the certain kind of berries, which sends power demands, can not directly be powered using power battery, therefore it is Raspberry Pi to increase 5V/3A DC-DC modules
Power supply, while also power for wireless data transfer module, the effect of Power Management Unit is similar to DC-DC, and lithium battery is exported
22.2V voltage conversions export to flight control units (input voltage 5V) and GPS (input voltage 5V).Electron speed regulator exports
It with signal intensity is not a fixed value that voltage to DC brushless motor, which is,.
Meet each hardware system for be electrically connected after, it is necessary to according to each module communication protocol etc. complete data input and
Output, the communication connection of this method hardware system are as shown in Figure 4.In the air in acquisition subsystem, Raspberry Pi and laser thunder
Communicated between reaching using ICP/IP protocol, communication speed is up to 100Mbit/s.According to the physical layer standard of ICP/IP protocol
And laser radar communications protocol, 4 data lines must be connected between Raspberry Pi and laser radar altogether, using Raspberry Pi as host definition number
According to line entitled TX+, TX-, RX+, RX-, therebetween using six type shielding twisted pair line connection of high-quality.Raspberry Pi and without line number
Carried out data transmission according between transport module using asynchronous serial communication mode, 2 data lines need to be connected altogether, based on Raspberry Pi
Machine defines data cable entitled TX and RX.
The LMS511 two-dimensional laser radars that selection pulse frequency is 27Kz, more rotor platform mission requirementses, flying height are big
It is more than 5min in 10m, flying speed 0m/s, hovering flight, flight time, payload is more than 4kg.
Step 2:Data prediction
In the data file bag of laser radar, each frame data are all comprising the equal data point of quantity, each data point bag
Containing distance value and an angle value is corresponding with, the initial data of laser radar is represented in the form of polar.For the ease of handling,
Polar coordinates are transformed into rectangular coordinate system first, such as Fig. 5.Rectangular coordinate system:Using the position of laser radar as origin, road is horizontal
To being ordinate for abscissa, vertical direction.Formula (1) and (2) are Formula of Coordinate System Transformation.
xi=ri*cosθi (1)
yi=ri*sinθi (2)
Wherein:riFor laser radar and the distance value of scanned object, θiFor with riCorresponding angle value, xiRepresent horizontal
Direction distance, yiVertical direction distance.
In the information gathered due to laser radar in addition to target information of road surface, also comprising road surface landscape, roadside fence
With roadside building etc., therefore, it is necessary to target information of road surface data are extracted.
For frame data, laser radar scanning be a certain cross section in road surface data, target road surface data are similar
Straight line.Clustering method of the present invention use based on distance threshold, state diagram when being gathered according to laser radar, such as Fig. 6, and
The laser radar data feature on road surface extracts the data point for belonging to target information of road surface.
After target information of road surface data are extracted, road surface information data point mathematical modulo is gone out using least square fitting
Type, formula (3).
Y=k*x+b (3)
Wherein:Y is the y of formula (2)iValue, x are the x of formula (1)iValue, k and b are model of fit coefficient.
Step 3:Vehicle cab recognition
Vehicle cab recognition is according to the data of existing vehicle vehicle commander, overall width and overall height and the analysis result of laser radar initial data
Vehicle is reclassified, vehicle characteristic parameter of the selection for vehicle cab recognition, the algorithm of Design vehicle characteristic parameter, according to
Pavement test verifies vehicle targets.Embodiment is as follows:
1. vehicle characteristic parameter is chosen
Vehicle characteristic parameter is chosen firstly the need of understanding which classification existing vehicle has, and understands the principle of these classification,
Then existing vehicle characteristic parameter species, and the definition of these characteristic parameters, then the feature of analyte sensors data are analyzed, really
Which fixed sensing data can use in data handling with vehicle characteristic parameter, finally determine the classification of vehicle and make
Vehicle characteristics parameter.The flow chart that this method vehicle characteristic parameter is chosen is as shown in Figure 7.Fig. 8 and Fig. 9 is respectively laser thunder
The schematic diagram data horizontal and vertical up to automobile, it can be seen that the height of vehicle can be calculated according to the Characteristics of LIDAR Data
Degree, width and compartment number.Consider and choose compartment number, overall height, overall width as characteristic parameter, because these three parameters are sharp
Optical radar can scan and more can significantly distinguish different vehicle type
2. vehicle classification
This method mainly for common car (including:Minicar, compact car, compact sedan, in-between car, medium-and-large-sized car,
Large car), SUV and MPV be identified, and classified again to these vehicles according to laser radar data feature.Table 1 is statistics
The main characteristic parameters table of the obtained 800 various vehicles obtained at random.
1 vehicle dimension average value statistics table of table
According to statistical analysis, to two box and wing-rooms on either side of a one-story house SUV, it is also necessary to set a threshold value to distinguish overall height.Work as car
When high threshold is arranged to 1629mm, the car of vehicle ratio of the wing-rooms on either side of a one-story house SUV overall height more than 1629mm and two box less than 1629mm
Respective classification accounting it is maximum, such as Figure 10, is respectively 86% and 95%.Therefore, 1629mm is arranged to overall height threshold value, car
Two-box automobile of the height less than 1629mm is considered two box, and the two-box automobile more than 1629mm is considered wing-rooms on either side of a one-story house SUV.
3. extract vehicle data
The premise for carrying out vehicle cab recognition is that the data for belonging to vehicle are extracted from the data of laser radar, this method
The flow chart of vehicle extraction algorithm is as shown in figure 11.
This method gathers the information such as road vehicles, the information of collection straight down using laser radar in the height of about 10m
Except target road surface data in data, there is the influence in roadside vegetation, guideboard, bypass and other non-targeted tracks, it is therefore desirable to
Target road data are extracted first, and according to step 2, the method for data prediction extracts target road data, target track
Circuit-switched data extraction flow chart is as shown in figure 12.
A frame radar data is gathered, after extracting target road data, starts to adjust road data position, i.e., according to road
Road benchmark model adjusts inclined road data.Shown in the formula such as formula (4) that adjustment road data uses
In above formula, hyIt is true vertical distance of the laser radar to road surface, k is being averaged for four frame model coefficients of formula (3)
Value, xiAnd yiIt is formula (1) and formula (2) as a result, iRFor total points of road data.
By road data position after levelling is tilted to, start the extraction of doubtful vehicle data, and by these data points
Store in variable data, only storage belongs to the value in the data point of doubtful vehicle horizontally and vertically direction in data.Extract doubtful
The flow chart of vehicle data is as shown in figure 13.
, it is necessary to identify that these data points include several cars after the doubtful vehicle number strong point of one frame data is extracted, that is, need
Extract the number of vehicles in doubtful vehicle data.The flow chart of this method extraction number of vehicles is as shown in figure 14,
4. car number
Identify after the vehicle data that frame data contains, it is necessary to be carried out to all vehicle datas of whole data file
Numbering, such as Figure 15.The vehicle driving mode travelled on road has two kinds:One kind is sequence of cars by below laser radar, i.e.,
Every time by a car, another kind is more cars in parallel through below laser radar, and more cars can be with the comparison side of two cars
On the basis of formula, compare two-by-two, distinguish the tandem of more cars, vehicle is numbered successively.
5. calculate vehicle characteristic parameter numerical value
This method uses the characteristic parameter of overall height, overall width and compartment number as identification vehicle, and here is these characteristic parameters
Computational methods.
(1) overall height
Overall height is that vehicle is in the horizontal plane of vehicle under the idle condition of operating status, can be supported to arrive and vehicle highest protrusion
Distance (mm) between horizontal plane where position.Such as Fig. 9, the maximum height of Y-axis is exactly the height of vehicle, takes all single frames height
Height of the maximum as vehicle.
(2) compartment number
Such as Figure 16, the top length of two-box automobile and three-box car is than (the ratio between roof frame number and vehicle commander's frame number) and tail length ratio (tailstock frame
The ratio between number and vehicle commander's frame number) there is significant difference, this method distinguishes compartment number according to the two ratios.By vehicle classification standard
Statistical result, is arranged to 0.51 by the ratio between roof and the frame number of vehicle commander threshold value, the tailstock and the ratio between vehicle commander's frame number threshold value is arranged to
Optimum can be obtained when 0.15, i.e., when a car roof and the frame number of vehicle commander are than less than 0.51 and the tailstock and the frame number of vehicle commander
Than thinking that the car is two-box automobile less than 0.15, be unsatisfactory for this condition is considered three-box car.
(3) overall width
From the definition of overall width, overall width is the distance that lateral direction of car is most wide in addition to rearview mirror.Come for frame data
Say, when vehicle is located at the different azimuths such as laser radar left side, right side and underface, the data characteristics of collection is different, such as
Figure 17.The vehicle of underface can directly calculate the horizontal distance of the data point of the leftmost side and the rightmost side as overall width.Positioned at thunder
Up to the data point of the left and right sides, there is shortage of data in not homonymy respectively.
As shown in figure 18, by taking radar left side vehicle as an example, leftmost side point is found first, then using interpolation method by vehicle
Data point establishes fitting a straight line, draws a horizontal line by leftmost side point, seeks the intersection point of horizontal line and the straight line;Then leftmost side point
Midtread coordinate with intersection point is exactly the midpoint of vehicle;The horizontal distance of middle point coordinates and the frame rightmost side point is exactly that vehicle is wide
The half of angle value.Above-mentioned overall width numerical value is all taken to all frame data of vehicle, maximum of the statistics per frame overall width numerical value is used as should
The overall width of car.
6. export vehicle cab recognition result
According to the vehicle characteristic parameter numerical value being calculated, vehicle classification is contrasted, exports vehicle cab recognition result.
Step 4:Vehicle speed measurement
This method establish by input, vehicle commander of overall width and overall height be output mathematical model, calculate vehicle commander and then
Time is calculated according to the frame number of vehicle data, and then obtains speed.Embodiment is as follows:
1. establish vehicle commander's mathematical model
The data of this method are using overall width, overall height as input, and vehicle commander is output, because the estimated value of vehicle commander allows have certain mistake
Difference, therefore select polynomial fitting method to establish model to given data, for simplified model, using a fitting of a polynomial, one
Shown in the formula of order polynomial such as formula (5)
Z=po+p1*x+p2y (5)
In above formula, z is the sample value of vehicle commander, and x is overall height sample value, and y is overall width sample value, p0、p1And p2It is model ginseng
Number.
2. calculate vehicle commander
According to vehicle commander's mathematical model of foundation, input overall height, overall width, calculate output vehicle commander.
3. export vehicle speed measurement result
Shown in the calculation formula of speed such as formula (6):
V is speed (unit in above formula:Km/h), L is Vehicle length (unit:M), f is that laser radar scanning frequency is (single
Position:Hz), n is the data frame number for belonging to the car in laser radar data.
Flight test verification result is as shown in table 2:
2 flight test result of table
As shown in Table 2, the mean error of overall height is 3.3mm, which can ignore for vehicle always high 1460mm
Disregard, and compartment number discrimination is 100%.
The overall width mean error of flight test is because more rotors can be because aerial wind speed increases suddenly in hovering flight
The small-scale rotation of generation, is caused when gathering road vehicle data, the scanning plane of radar is no longer parallel with vehicle cross section, meeting
The small-scale inclination of generation, sideling enters below laser radar, therefore, obtained result can be more inclined than actual value equivalent to vehicle
Greatly.Speed error contrasts the error of speedometer in 0.4km/h to 3km/h, which is to receive for speed calculating
's.
Method of the present invention is based on multi-rotor unmanned aerial vehicle and carries laser radar platform collection information of road surface, can be effective
Solve to may not apply to emergent section vehicle cab recognition and car using fixed platform carrying sensor in vehicle cab recognition and vehicle speed measurement
The problem of speed measurement;And the method for giving and vehicle cab recognition and vehicle speed measurement being completed based on laser radar data is studied, can be at the same time
Output vehicle cab recognition and vehicle speed measurement are as a result, realize the result for providing two parameter indexes of ITS at the same time using a kind of sensor.
Based on method provided by the invention, flight test platform is built, and this method is demonstrated by flight test
Validity, wherein:In vehicle classification, vehicle targets are 100% for non-black vehicle identification rate, black vehicle identification
Rate is 57.1%, and non-black vehicle identification rate is very high, and black vehicle is low relative to non-black vehicle identification rate;Vehicle speed measurement knot
Fruit meets the requirement of GB15082-2008.
Claims (5)
1. a kind of more rotors carry the vehicle cab recognition and vehicle speed measurement method of laser radar, it is characterised in that:Comprise the following steps that:
Step 1: data prediction:According to radar signal feature and the characteristic of more rotor platforms, radar data is carried out first
Coordinate system conversion, the extraction of target information of road surface data, the levelling of information of road surface data and radar site;Laser radar is collected
Significant figure strong point be transformed into by polar coordinate system in Descartes's rectangular coordinate system, obtain preprocessed features, so as to follow-up vehicle know
Other and vehicle speed measurement algorithm use;
Step 2: vehicle cab recognition:Preprocessed features according to step 1 to lidar measurement data, carry out again existing vehicle
Classification;The characteristic parameter of the height, width and compartment number of vehicle as vehicle cab recognition is chosen, it is special according to the vehicle being calculated
Parameter values are levied, contrast vehicle classification standard, export vehicle cab recognition result;
Step 3: vehicle speed measurement:According to polynomial fitting method, establish using overall width and overall height as number that input, vehicle commander are output
Model is learned, vehicle commander is being calculated and then the time is calculated according to the frame number of vehicle data, and then obtaining speed.
2. a kind of more rotors as claimed in claim 1 carry the vehicle cab recognition and vehicle speed measurement method of laser radar, its feature exists
In:Vehicle cab recognition described in step 2 comprises the following steps that:
Step 1, choose the characteristic parameter of overall height, overall width and compartment number as identification vehicle;
Step 2, according to laser radar data feature classify vehicle again, establishes characterized by overall height, overall width and compartment number
Vehicle classification standard;
Step 3, extract the data for belonging to vehicle from the data of laser radar;
Vehicle is numbered in step 4, the sequencing for driving into according to vehicle laser radar scanning scope;
Step 5, three compartment number, overall height and overall width characteristic parameters for calculating vehicle in laser radar data;
Step 6, the overall height obtained according to step 5, the characteristic parameter of overall width and compartment number, i.e. vehicle cab recognition, with step 2 vehicle point
Class standard contrasts, and exports vehicle cab recognition result.
3. a kind of more rotors as claimed in claim 1 carry the vehicle cab recognition and vehicle speed measurement method of laser radar, its feature exists
In:Foundation described in step 3 is as the mathematical model that input, vehicle commander are output using overall width and overall height:
Z=po+p1*x+p2y (5)
In above formula, z is the sample value of vehicle commander, and x is overall height sample value, and y is overall width sample value, p0、p1And p2It is model parameter.
4. a kind of more rotors as claimed in claim 2 carry the vehicle cab recognition and vehicle speed measurement method of laser radar, its feature exists
In:The method that vehicle classification standard is established described in step 2 is:According to Principle of Statistics, vehicle is divided into two box, wing-rooms on either side of a one-story house
4 SUV, three box car and cart types;Wherein two box and wing-rooms on either side of a one-story house SUV is similar, according to statistical result, to two class overall height
One threshold value is set to distinguish, 1629mm is arranged to overall height threshold value, two-box automobile of the overall height less than 1629mm is considered wing-rooms on either side of a one-story house
Car, the two-box automobile more than 1629mm are considered wing-rooms on either side of a one-story house SUV.
5. a kind of more rotors as claimed in claim 2 carry the vehicle cab recognition and vehicle speed measurement method of laser radar, its feature exists
In:The method of calculating characteristic parameter is described in step 5:
(1) overall height:Take height of the maximum of all single frames height as vehicle;
(2) compartment number:It is long than distinguishing compartment number with tail length ratio according to top;By vehicle classification canonical statistics as a result, by roof with
The ratio between frame number of vehicle commander threshold value is arranged to 0.51, can be obtained when the ratio between the tailstock and vehicle commander's frame number threshold value are arranged to 0.15 optimal
As a result, i.e. when the frame number of a car roof and vehicle commander are than less than 0.51 and the tailstock and the frame number of vehicle commander ratio think this less than 0.15
Car is two-box automobile, and be unsatisfactory for this condition is considered three-box car;
(3) overall width:Any one frame in the data collected from radar calculates characteristic parameter;
If vehicle to be measured is located on the left of radar, leftmost side point is found first, is then established the data point of vehicle using interpolation method
Fitting a straight line, draws a horizontal line by leftmost side point, seeks the intersection point of horizontal line and the straight line;Then in leftmost side point and intersection point
Point horizontal coordinate is exactly the midpoint of vehicle;The horizontal distance of middle point coordinates and the frame rightmost side point is exactly the one of vehicle width value
Half;Above-mentioned overall width numerical value is all taken to all frame data of vehicle, is counted per car of the maximum of frame overall width numerical value as this car
It is wide;
If vehicle to be measured is located on the right side of radar, rightmost side point is found first, is then established the data point of vehicle using interpolation method
Fitting a straight line, draws a horizontal line by rightmost side point, seeks the intersection point of horizontal line and the straight line;Then in rightmost side point and intersection point
Point horizontal coordinate is exactly the midpoint of vehicle;The horizontal distance of middle point coordinates and the frame leftmost side point is exactly the one of vehicle width value
Half;Above-mentioned overall width numerical value is all taken to all frame data of vehicle, is counted per car of the maximum of frame overall width numerical value as this car
It is wide;
If vehicle to be measured is located at immediately below radar, the horizontal direction of leftmost side data point and rightmost side data point can be directly calculated
Range difference is as overall width numerical value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711260040.XA CN107967804A (en) | 2017-12-04 | 2017-12-04 | A kind of more rotors carry the vehicle cab recognition and vehicle speed measurement device and method of laser radar |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711260040.XA CN107967804A (en) | 2017-12-04 | 2017-12-04 | A kind of more rotors carry the vehicle cab recognition and vehicle speed measurement device and method of laser radar |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107967804A true CN107967804A (en) | 2018-04-27 |
Family
ID=61999300
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711260040.XA Pending CN107967804A (en) | 2017-12-04 | 2017-12-04 | A kind of more rotors carry the vehicle cab recognition and vehicle speed measurement device and method of laser radar |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107967804A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109598947A (en) * | 2018-12-26 | 2019-04-09 | 武汉万集信息技术有限公司 | A kind of vehicle identification method and system |
CN109887307A (en) * | 2019-02-28 | 2019-06-14 | 南京瑞贻电子科技有限公司 | A kind of car radar speed measuring device |
CN112230222A (en) * | 2020-12-11 | 2021-01-15 | 北京雷信科技有限公司 | Microwave radar device for recognizing vehicle type |
CN113514837A (en) * | 2021-06-02 | 2021-10-19 | 东南大学 | Pavement track detection device and method |
CN113869196A (en) * | 2021-09-27 | 2021-12-31 | 中远海运科技股份有限公司 | Vehicle type classification method and device based on laser point cloud data multi-feature analysis |
Citations (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1605033A (en) * | 2001-12-14 | 2005-04-06 | 卡波施交通公司 | Method and device for the geometric measurement and speed determination of vehicles |
CN101661669A (en) * | 2009-09-11 | 2010-03-03 | 北京物资学院 | Road vehicle type identification method and system thereof |
CN102436738A (en) * | 2011-09-26 | 2012-05-02 | 同济大学 | Traffic monitoring device based on unmanned aerial vehicle (UAV) |
CN103050010A (en) * | 2012-12-31 | 2013-04-17 | 北京万集科技股份有限公司 | Integrated laser scanning traffic survey device and integrated laser scanning traffic survey method |
CN103730011A (en) * | 2012-10-12 | 2014-04-16 | 博隆建亚科技(北京)有限公司 | Vehicle type recognition detection method and system based on laser |
CN104064030A (en) * | 2014-07-01 | 2014-09-24 | 武汉万集信息技术有限公司 | Vehicle type identification method and vehicle type identification system |
CN203941616U (en) * | 2014-07-01 | 2014-11-12 | 武汉万集信息技术有限公司 | A kind of model recognition system |
CN104183133A (en) * | 2014-08-11 | 2014-12-03 | 广州普勒仕交通科技有限公司 | Method for acquiring and transmitting road traffic flow dynamic information |
CN104766481A (en) * | 2015-04-29 | 2015-07-08 | 深圳市保千里电子有限公司 | Method and system for unmanned plane to conduct vehicle tracking |
KR20150114236A (en) * | 2014-04-01 | 2015-10-12 | 메타빌드주식회사 | Vehicles classification system using radar and the method of the same, toll collection system |
CN105046968A (en) * | 2015-06-19 | 2015-11-11 | 武汉万集信息技术有限公司 | Motor bus identification and snapshot method, apparatus and system |
CN105427614A (en) * | 2015-08-28 | 2016-03-23 | 北京动视元科技有限公司 | Model classification system and method |
CN105761494A (en) * | 2016-05-12 | 2016-07-13 | 招商局重庆交通科研设计院有限公司 | Abnormal traffic information collecting method based on unmanned aerial vehicle |
CN205621284U (en) * | 2016-05-10 | 2016-10-05 | 武汉万集信息技术有限公司 | Traffic flow investigates system based on unmanned aerial vehicle |
CN106097743A (en) * | 2016-08-13 | 2016-11-09 | 张选琪 | Vehicle identification UAS |
CN205788760U (en) * | 2016-05-10 | 2016-12-07 | 武汉万集信息技术有限公司 | A kind of vehicle detecting system based on unmanned plane |
CN107358796A (en) * | 2016-05-10 | 2017-11-17 | 武汉万集信息技术有限公司 | A kind of vehicle checking method based on unmanned plane |
-
2017
- 2017-12-04 CN CN201711260040.XA patent/CN107967804A/en active Pending
Patent Citations (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1605033A (en) * | 2001-12-14 | 2005-04-06 | 卡波施交通公司 | Method and device for the geometric measurement and speed determination of vehicles |
CN101661669A (en) * | 2009-09-11 | 2010-03-03 | 北京物资学院 | Road vehicle type identification method and system thereof |
CN102436738A (en) * | 2011-09-26 | 2012-05-02 | 同济大学 | Traffic monitoring device based on unmanned aerial vehicle (UAV) |
CN103730011A (en) * | 2012-10-12 | 2014-04-16 | 博隆建亚科技(北京)有限公司 | Vehicle type recognition detection method and system based on laser |
CN103050010A (en) * | 2012-12-31 | 2013-04-17 | 北京万集科技股份有限公司 | Integrated laser scanning traffic survey device and integrated laser scanning traffic survey method |
KR20150114236A (en) * | 2014-04-01 | 2015-10-12 | 메타빌드주식회사 | Vehicles classification system using radar and the method of the same, toll collection system |
CN104064030A (en) * | 2014-07-01 | 2014-09-24 | 武汉万集信息技术有限公司 | Vehicle type identification method and vehicle type identification system |
CN203941616U (en) * | 2014-07-01 | 2014-11-12 | 武汉万集信息技术有限公司 | A kind of model recognition system |
CN104183133A (en) * | 2014-08-11 | 2014-12-03 | 广州普勒仕交通科技有限公司 | Method for acquiring and transmitting road traffic flow dynamic information |
CN104766481A (en) * | 2015-04-29 | 2015-07-08 | 深圳市保千里电子有限公司 | Method and system for unmanned plane to conduct vehicle tracking |
CN105046968A (en) * | 2015-06-19 | 2015-11-11 | 武汉万集信息技术有限公司 | Motor bus identification and snapshot method, apparatus and system |
CN105427614A (en) * | 2015-08-28 | 2016-03-23 | 北京动视元科技有限公司 | Model classification system and method |
CN205621284U (en) * | 2016-05-10 | 2016-10-05 | 武汉万集信息技术有限公司 | Traffic flow investigates system based on unmanned aerial vehicle |
CN205788760U (en) * | 2016-05-10 | 2016-12-07 | 武汉万集信息技术有限公司 | A kind of vehicle detecting system based on unmanned plane |
CN107358796A (en) * | 2016-05-10 | 2017-11-17 | 武汉万集信息技术有限公司 | A kind of vehicle checking method based on unmanned plane |
CN105761494A (en) * | 2016-05-12 | 2016-07-13 | 招商局重庆交通科研设计院有限公司 | Abnormal traffic information collecting method based on unmanned aerial vehicle |
CN106097743A (en) * | 2016-08-13 | 2016-11-09 | 张选琪 | Vehicle identification UAS |
Non-Patent Citations (2)
Title |
---|
卢玥,刘学军,晋蓓: "基于视频的车辆几何参数提取与车型分类", 《计算机工程》 * |
王寒凝: "基于激光测距的车型识别分类系统", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109598947A (en) * | 2018-12-26 | 2019-04-09 | 武汉万集信息技术有限公司 | A kind of vehicle identification method and system |
CN109887307A (en) * | 2019-02-28 | 2019-06-14 | 南京瑞贻电子科技有限公司 | A kind of car radar speed measuring device |
CN109887307B (en) * | 2019-02-28 | 2020-07-03 | 南京瑞贻电子科技有限公司 | Automobile radar speed measuring device |
CN112230222A (en) * | 2020-12-11 | 2021-01-15 | 北京雷信科技有限公司 | Microwave radar device for recognizing vehicle type |
CN112230222B (en) * | 2020-12-11 | 2021-03-16 | 北京雷信科技有限公司 | Microwave radar device for recognizing vehicle type |
CN113514837A (en) * | 2021-06-02 | 2021-10-19 | 东南大学 | Pavement track detection device and method |
CN113869196A (en) * | 2021-09-27 | 2021-12-31 | 中远海运科技股份有限公司 | Vehicle type classification method and device based on laser point cloud data multi-feature analysis |
CN113869196B (en) * | 2021-09-27 | 2022-04-19 | 中远海运科技股份有限公司 | Vehicle type classification method and device based on laser point cloud data multi-feature analysis |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107967804A (en) | A kind of more rotors carry the vehicle cab recognition and vehicle speed measurement device and method of laser radar | |
Han et al. | Research on road environmental sense method of intelligent vehicle based on tracking check | |
CN110780305B (en) | Track cone detection and target point tracking method based on multi-line laser radar | |
CN203063940U (en) | Pilotless automobile device | |
CN108256413B (en) | Passable area detection method and device, storage medium and electronic equipment | |
CN103065151B (en) | Vehicle identification method based on depth information | |
CN109278752B (en) | Energy optimization control method for plug-in hybrid electric vehicle based on cooperative sensing | |
CN102944224B (en) | Work method for automatic environmental perception systemfor remotely piloted vehicle | |
CN102722989B (en) | Expressway microclimate traffic early warning method based on fuzzy neural network | |
CN106428558B (en) | A kind of track synthesis method for inspecting based on the dual-purpose unmanned plane of sky-rail | |
CN102737510A (en) | Real-time traffic condition acquisition method based on mobile intelligent terminal | |
CN106741890A (en) | A kind of high-speed railway safety detecting system based on the dual-purpose unmanned plane of empty rail | |
CN110378416A (en) | A kind of coefficient of road adhesion estimation method of view-based access control model | |
CN109283930A (en) | A kind of spherical intelligent inspection robot | |
CN104182957A (en) | Traffic video detection information method and device | |
CN102831768A (en) | Hybrid power bus driving condition forecasting method based on internet of vehicles | |
CN108045375A (en) | A kind of unmanned automatic driving automobile autonomous operation method for controlling driving speed | |
CN109910881A (en) | A kind of detection processing method and system of turning road vision dead zone information | |
CN113792566A (en) | Laser point cloud processing method and related equipment | |
CN101419065A (en) | Geometrical parameter dynamic measurement device for vehicle | |
CN113885045A (en) | Method and device for detecting lane line | |
CN106494611A (en) | A kind of dual-purpose patrol unmanned machine of empty rail | |
CN114332722A (en) | Real-time estimation method for adhesion coefficient of mixed ice and snow road surface based on video data | |
CN106803348A (en) | A kind of overpass vehicle travel time Forecasting Approach for Short-term based on bayonet socket | |
CN202817605U (en) | UAV routing-inspection line corridor device based on millimeter-wave radar |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180427 |
|
WD01 | Invention patent application deemed withdrawn after publication |