CN103669182B - Pavement crack recognition method based on camera and line laser - Google Patents
Pavement crack recognition method based on camera and line laser Download PDFInfo
- Publication number
- CN103669182B CN103669182B CN201310573077.3A CN201310573077A CN103669182B CN 103669182 B CN103669182 B CN 103669182B CN 201310573077 A CN201310573077 A CN 201310573077A CN 103669182 B CN103669182 B CN 103669182B
- Authority
- CN
- China
- Prior art keywords
- pavement
- camera
- image data
- laser
- area array
- 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.)
- Active
Links
Landscapes
- Length Measuring Devices By Optical Means (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention belongs to the technical field of pavement crack recognition, and discloses a pavement crack recognition method based on a camera and a line laser. The pavement crack recognition method based on the camera and a line laser comprises the following steps of laser emitting, generation of a trigger signal, collection of pavement image date, extraction of strip-shaped euphotic belts, linear fitting, generation of an euphotic belt curved array, crack point recognition, and computing of length, width and depth of cracks. The method can detect the pavement cracks rapidly and efficiently, and has the characteristics of being high in recognition speed, high in automation and high in detection accuracy.
Description
Technical field
The invention belongs to pavement crack technology of identification field, know particularly to based on the pavement crack of camera and laser line generator
Other device and its recognition methodss.
Background technology
Being gradually improved of fast development with China's expressway construction and highway network construction, the maintenance work of highway
Also increasingly it is taken seriously.After highway builds up, by wheel repeatedly roll and the various natural cause such as ice, rain, snow shadow
Ring, road surface occurs crack, the disrepair phenomenon such as concavo-convex.Pavement crack is one of principal mode of pavement disease, can affect to drive a vehicle
Safety and highway service life, pavement crack be assessment road quality an important indicator.
The detection in road pavement crack mainly includes manual detection, camera calibration both approaches at present.Manual detection road surface
Need during crack testing staff to live road pavement crack depth, length and width situations such as measure and record, finally
Evaluated according to recorded data fracture.But because the method data of manual detection is by artificial treatment, so measurement
There is the impact of subjective factorss in result, and inefficiency is it is impossible to adapt to the fast-developing requirement of highway.
During camera calibration pavement crack, shot using camera road pavement, to obtain the view data on road surface.Then lead to
Cross and view data is filtered, after binaryzation etc. processes, detects the crack on road surface.But camera calibration method is easily subject to
The interference of the rubbish on road surface, is easily mistakenly identified as crack by sticking the foul on road surface, thus leading to recognition accuracy low.
Content of the invention
It is an object of the invention to proposing the pavement crack identifying device based on camera and laser line generator and its recognition methodss.
Should be few with the pavement crack identifying device investment cost of laser line generator based on camera, can carry out in road pavement crack rapidly and efficiently
Detection；This be had based on the pavement crack recognition methodss of camera and laser line generator, and recognition speed is fast, automatization is high, Detection accuracy
High feature.
For realizing above-mentioned technical purpose, the present invention adopts the following technical scheme that and is achieved.
Technical scheme one:
Pavement crack identifying device based on camera and laser line generator is it is characterised in that include: impulse scaler, industry control
Machine, vehicle, it is fixedly mounted on the laser line generator of headstock center of vehicle, extends anteriorly out at the headstock top center of vehicle
Flat board；One end of described flat board is fixed at the headstock top center of vehicle, and the lower section of the other end of described flat board is fixed with
Area array cameras；Towards ground, the laser head of described laser line generator and the angle of horizontal plane are 40 for the laser head of described laser line generator
Spend between 80 degree, the camera lens vertically ground of described area array cameras；Coaxially it is socketed with increasing in the trailing wheel rotating shaft of described vehicle
Amount formula rotary encoder；
Described incremental rotary encoder electrically connects the input of impulse scaler, the outfan electricity of described impulse scaler
It is connected with pci card；Described pci card is plugged in the pci draw-in groove of industrial computer, described area array cameras pass through camera link line with
Pci card is connected.Pci card is the pci card for gathering image.
The feature of the technical program and further improvement is that:
Described laser line generator is HONGGUANG laser line generator, and the laser head of described laser line generator and the angle of horizontal plane are 60 degree.
The frame speed of described area array cameras is 100fps.
Technical scheme two:
Based on the pavement crack recognition methodss of camera and laser line generator, based on the above-mentioned road surface based on camera and laser line generator
Crack identification device, comprises the following steps:
Area array cameras frame speed setting: area array cameras frame speed is set to default frame speed；
Laser emission: drive vehicle at the uniform velocity travel on road surface, laser line generator towards road surface-emission laser, now, on road
Strip euphotic belt is formed on face；
Trigger produces: during vehicle travels, incremental rotary encoder sends pulse signal to pulsimeter
Number device, impulse scaler pulse signals are counted；Impulse scaler module is just sent out to pci card every n pulse number
Send a trigger, n is the natural number more than 0；
Pavement image data acquisition: after pci card receives trigger, send control signal, chain of command battle array to area array cameras
Camera road pavement carries out captured in real-time；Area array cameras are sent shooting the pavement image data being formed to industrial computer by pci card
In；
Strip euphotic belt extracts: after vehicle stops travelling, in industrial computer, all pavement image data of storage is divided
Do not carry out Coordinate Conversion, so that all pavement image data is changed to three-dimensional world coordinate system from camera coordinates system, to Coordinate Conversion
All pavement image data afterwards carry out picture smooth treatment respectively；To all pavement images carrying out after picture smooth treatment
Data carries out binary conversion treatment respectively；In the every width pavement image data after binary conversion treatment, extract in strip euphotic belt
The pavement image data of heart line；
Fitting a straight line: the pavement image data to all strip euphotic belt centrages extracting, it is respectively adopted a young waiter in a wineshop or an inn
Multiplication will carry out fitting a straight line, obtain a plurality of straight line；Every straight line is carried out translation transformation and rotation transformation, makes every straight line position
In xoy plane, and putting in order between these straight lines is made to keep one with the acquisition order of corresponding pavement image data
Cause；Calculate translational movement and the rotation amount of every straight line respectively；
Euphotic belt curve battle array generates: the translational movement according to every straight line and rotation amount, will not carry out all of fitting a straight line
The pavement image data of strip euphotic belt centrage, carries out corresponding translation and rotates, obtain a plurality of curve；Row according to curve
Row order chooses curve successively, often chooses m bar curve, just described m bar curve is designated as a curve battle array, and m is oneself more than 500
So count；
Crack point identification: respectively surface fitting is carried out to each curve battle array, obtains fitting surface；In calculated curve battle array each
Beeline between the pavement image data point not being fitted and corresponding fitting surface, if described beeline is more than set
Definite value, then be identified as crack point by corresponding image data point；
Calculate fracture length, width and depth: identify that crack point carries out clustering processing to all, obtain several minimum
Envelope cuboid, the number of minimum envelop cuboid is the bar number of pavement crack, and the length of each minimum envelop cuboid is
For the length of corresponding pavement crack, the width of the as corresponding pavement crack of width of each minimum envelop cuboid, each is minimum
The height value of envelope cuboid is the depth value of corresponding pavement crack.
The feature of the technical program and further improvement is that:
During the speed setting of area array cameras frame, area array cameras are improved by the sampled pixel reducing area array cameras
Frame speed.
Before vehicle travels, the gear ratio of adjustment incremental rotary encoder and trailing wheel rotating shaft, so that vehicle is often advanced
5mm, incremental rotary encoder just produces 10 pulse signals.
After the speed setting of area array cameras frame, carry out area array cameras demarcation, then carry out Laser emission；Described area array cameras
Demarcation comprises the following steps: makes scaling board first, scaling board arranges uncalibrated image, area array cameras shoot diverse location respectively
Scaling board, gather image, the image collecting is demarcated, obtains the intrinsic parameter of camera and outer parameter；
During strip euphotic belt extracts, principle according to lens optical imaging and the intrinsic parameter of camera and outer
All pavement image data of storage are carried out Coordinate Conversion, make all pavement image data from camera coordinates system by parameter respectively
Change to three-dimensional world coordinate system.
During Laser emission, fast according to the frame of the sampled distance setting pavement image and area array cameras first, if
Determine the initial velocity of vehicle traveling, then drive vehicle and at the uniform velocity travel by described initial velocity on road surface.
The invention has the benefit that the pavement crack identifying device based on camera and laser line generator of the present invention, mainly
Part adopts high-speed area array camera, laser line generator, incremental rotary encoder, pci card and industrial computer, have investment cost few,
The feature that design is simple, reliability is high；The pavement crack identifying device method based on camera and laser line generator of the present invention, is based on
Above-mentioned pavement crack identifying device, the identification in its road pavement crack and calculating are mainly automatically performed by industrial computer, have intelligence
Change feature high, that certainty of measurement is high, and the recognition result in road pavement crack is directly perceived, reliable.
Brief description
Fig. 1 is the structural representation based on camera and the pavement crack identifying device of laser line generator of the present invention；
Fig. 2 is the circuit connection diagram based on camera and the pavement crack identifying device of laser line generator of the present invention；
Fig. 3 is the flow chart based on camera and the pavement crack recognition methodss of laser line generator of the present invention；
Fig. 4 is area array cameras and the laser line generator work process schematic diagram of the present invention；
Fig. 5 is that the curve battle array of the present invention arranges schematic diagram.
Specific embodiment
The invention will be further described below in conjunction with the accompanying drawings:
With reference to Fig. 1, it is the structural representation based on camera and the pavement crack identifying device of laser line generator of the present invention.Should
Vehicle 7 is included based on the pavement crack identifying device of camera and laser line generator, vehicle 7 is used for fixing area array cameras (high-speed area array
Camera), laser line generator etc..At vehicle 7 headstock top center, the flat board 8 extending forward is installed, one end of flat board 8 is fixed on
At the headstock top center of vehicle 7, the lower section of the other end of flat board 8 has been bolted area array cameras 1.The length of flat board 8
(length in vehicle forward direction) can be set as needed, for example, the length of flat board 8 20cm to 80cm it
Between.The camera lens of area array cameras 1 vertically road surface.Laser line generator 2 (for sending the laser instrument of line-like laser beam flux) uses bolt
It is fixed on the headstock center of vehicle 7, using tilting mounting means, the laser head of laser line generator 2 is towards ground, laser head and water
The angle of plane is between 40 degree to 80 degree it is preferable that laser head is 60 degree with the angle of horizontal plane.In the embodiment of the present invention,
Vehicle 7 is mounting platform, and laser line generator 2 is used for road surface-emission laser, thus forming strip euphotic belt, face battle array phase on road surface
Machine 1 is used for the pavement image data containing strip euphotic belt is acquired.The height of area array cameras 1 is more than laser line generator 2
Highly.
Incremental rotary encoder 3 passes through the coaxial socket of adapter and is fixed in vehicle rear wheel rotating shaft.Increment type rotation is compiled
Code device 3 is used for detecting the speed of vehicle, sends pulse signal.When installing incremental rotary encoder 3, adjustment increment type rotation
The gear ratio of encoder and vehicle rear wheel rotating shaft 5mm so that vehicle often advances, incremental rotary encoder just produces 10 pulses
Signal.In the embodiment of the present invention, incremental rotary encoder adopts the e6a2-cw3c incremental rotary encoder of omron company.
Vehicle 7 is internally provided with combined cabinet 6, deposits industrial computer 5 and impulse scaler 4 in combined cabinet 6.Industry control
It is inserted with pci card (for gathering image) in pci draw-in groove on machine 5.With reference to Fig. 2, it is the present invention based on camera and laser line generator
Pavement crack identifying device circuit connection diagram.Incremental rotary encoder 3 electrically connects the input of impulse scaler 4
End, the outfan of impulse scaler is electrically connected with pci card；Pci card is plugged in the pci draw-in groove of industrial computer 5, and area array cameras 1 lead to
Cross camera link line to be connected with pci card.In the embodiment of the present invention, pci card adopts the grablink of euresys company
Expert 2camera link image pick-up card.
In the embodiment of the present invention, incremental rotary encoder 3 adopts the e6a2-cw3c increment type rotation of Omron Corp to compile
Code device, can carry out two-phase and test the speed, have the resolution of 500p/r.Laser line generator 2 adopts Changchun NPD projects photoelectric technology limited
The mrl-iii-635 laser instrument of company, its wavelength is 635 ± 5nm, using 220v Alternating Current Power Supply, can send red laser.Face battle array
Camera 1 adopts the falcon 1.4m100 high speed cmos camera of teledyne dalsa company.Incremental rotary encoder 3 is by car
7 current travel speeds are sent to impulse scaler 4 after being converted into periodically pulsing signal, impulse scaler 4 record should
Pulse number, and just send a trigger to pci card every 10 pulse numbers.Pci clamping receives trigger
Afterwards, send control signal to area array cameras 1, control area array cameras collection pavement image data, and by pavement image data storage
To in high performance industrial computer 5.High-performance industrial computer 5, according to pavement image data, using image processing method, identifies and calculates
Obtain pavement crack parameter.
With reference to Fig. 3, it is the flow chart based on camera and the pavement crack recognition methodss of linear laser of the present invention.This base
Pavement crack recognition methodss in camera and linear laser comprise the following steps:
1) area array cameras frame speed (i.e. sample frequency) setting
Due to area array cameras, in sampled pixel highest, (area array cameras 1 adopt the falcon of teledyne dalsa company
1.4m100 high speed cmos camera, has the highest sampled pixel of 1400x1024) frame speed for 100fps it is impossible to meet collection essence
The requirement of degree, the therefore method by reducing the sampled pixel of area array cameras improve the frame speed of camera.In the embodiment of the present invention,
The sampled pixel of area array cameras is set to 140*102 so that the frame speed of camera improves to 1000fps.
2) area array cameras are demarcated
Make scaling board first, uncalibrated image is arranged on scaling board, high-speed area array camera shoots diverse location respectively
Scaling board, gather image, these images are imported in image processing software and are demarcated, obtain area array cameras intrinsic parameter,
Outer parameter and area array cameras inside and outside parameter corresponding relation；Its detailed process is as follows:
Set up Two-Dimensional rectangular coordinate system on a blank sheet of paper, and draw chequered with black and white on blank sheet of paper, the length of side is for 10cm just
Square, forms and demarcates paper.When area array cameras are demarcated, some zoness of different demarcating paper and being placed on road surface are allowed face battle array phase
Machine shoots to it, then imports to these images in Mathematical treatment software matlab2007 and is demarcated.At mathematics
In reason software matlab2007, after importing above-mentioned image, manually select out initial point, y-axis and x-axis, each image processed,
The corresponding relation of this area array cameras intrinsic parameter, outer parameter and area array cameras inside and outside parameter is just obtained after the completion of process.
3) set the initial velocity that vehicle travels
In view of the accuracy of road pavement crack identification, set the sampled distance d of pavement image as 5mm, due to face battle array phase
Sample frequency f of machine is 1000fps, therefore set travel speed v of vehicle as:
V=d*f=5 × 10-3* 1000=5m/s
4) Laser emission
With reference to Fig. 4, it is area array cameras and the laser line generator work process schematic diagram of the present invention.After vehicle starts running, will
Laser line generator supplies 220v alternating current, starts laser line generator, and laser line generator sends linear laser, irradiates road surface with 60 degree of angle of incidence,
Strip euphotic belt is formed on road surface.
5) trigger produces
Because vehicle is by driver's pilot steering, actual travel speed can become in the left and right of setting speed (5m/s)
Dynamic.Therefore incremental rotary encoder starts the Real-time Collection to detection car travel speed, and impulse scaler recording increment formula is revolved
Turn pulse number produced by encoder, and every 10 pulse numbers, just send a trigger to pci card.
6) pavement image data acquisition
Pci card, upon receipt of trigger, just sends control signal to area array cameras.In conjunction with Fig. 4, area array cameras receive
To after control signal, just the pavement image data containing strip euphotic belt for captured in real-time one pair (extends vehicle by adjusting flat board
The length of front top is it is ensured that the pavement image packet euphotic belt containing strip that shoots of area array cameras), and by pavement image data
By in pci card real-time storage to high performance industrial computer.
7) strip euphotic belt extracts
When vehicle stop travel after, in industrial computer, according to lens optical imaging principle and camera intrinsic parameter and
All pavement image data of storage are carried out Coordinate Conversion, make all pavement image data from camera coordinates by outer parameter respectively
System's conversion is to three-dimensional world coordinate system；Then all pavement image data after Coordinate Conversion are carried out at image smoothing respectively
Reason；Binary conversion treatment is carried out respectively to all pavement image data carrying out after picture smooth treatment；After binary conversion treatment
Every width pavement image data in, extract the pavement image data of discrete strip euphotic belt centrage.
In carrying out the every width pavement image data after picture smooth treatment, each pixel of this pavement image data
Gray value changes between 0～255.Due to there is the difference in obvious gray scale between strip euphotic belt and road surface, pass through
Select a suitable threshold value, the gray value of each pixel of pavement image data is compared with this threshold value, if greater than
This point value is then changed into 1 by this threshold value, otherwise is 0.After view picture pavement image data operation is completed, just comprise only in image
Numeral 0 and 1, such binarization just completes, in pavement image data, the profile that formed by the picture point by 1 for the numerical value
It is the appearance profile of strip euphotic belt.In this appearance profile, extract the pavement image data of strip euphotic belt centrage
(extracting the lines of strip euphotic belt centrage).
8) fitting a straight line
Pavement image data to all discrete strip euphotic belt centrage extracting, is respectively adopted method of least square
Fitting a straight line will be carried out, obtain a plurality of straight line (obtaining a plurality of straight line equation).Every straight line is carried out translation transformation and rotation
Transformation is changed, and so that every straight line is located in xoy plane (in three-dimensional world coordinate system), and makes the arrangement between these straight lines suitable
Sequence is consistent (being from left to right arranged in order) with the acquisition order of corresponding pavement image data it is preferable that straight by every
During line carries out translation transformation and rotation transformation, every straight line and the intersection point of x-axis also to be made to be evenly arranged in x-axis, phase
Adjacent two intersection point distances are steady state value (for example, 5mm).After translation transformation and rotation transformation are carried out to every straight line, respectively
Calculate translational movement and the rotation amount of every straight line.
9) euphotic belt curve battle array generates
Translational movement according to every straight line and rotation amount, the every width pavement image data not carrying out fitting a straight line (is carried
The pavement image data of the strip euphotic belt centrage taking out), carry out corresponding translation and rotate, obtain a plurality of curve.According to
Putting in order of curve chooses curve successively, often chooses 1000 curves, just 1000 curves is designated as a curve battle array, and each is bent
Linear array is to carry out the basis of crack point identification.With reference to Fig. 5, it is the curve battle array arrangement schematic diagram of the present invention.In Figure 5, with many
The camber line of bar curve intersection represents vehicle driving trace.
10) crack point identification
According to the features of shape of road cross section, using cubic surface fitting algorithm, to each curve battle array march respectively
Face matching, obtains fitting surface.Then in calculated curve battle array, each is not fitted pavement image data point and corresponding matching
Beeline between curved surface, if beeline is more than setting value (general setting value is 5mm), by corresponding view data
Point is identified as crack point.
11) fracture length, width and depth are calculated
Identify that crack point carries out clustering processing to all, obtain several minimum envelop cuboids, minimum envelop is rectangular
The number of body is the bar number of pavement crack, and the length of each minimum envelop cuboid is the length of corresponding pavement crack, often
The width of individual minimum envelop cuboid is the width of corresponding pavement crack, and the height value of each minimum envelop cuboid is as right
Answer the depth value of pavement crack.
The ultimate principle of the present invention is: when vehicle is travelled on the road of required detection with the speed setting, line laser
Laser launched by device, thus forming strip euphotic belt on road surface；Area array cameras gather the data of this strip euphotic belt simultaneously, and with
The form real-time storage of view data is in high performance industrial computer.Industrial computer according to the pavement image data being stored, first
Using binarization method, the centreline data of all euphotic belts is extracted from view data, and the side using image mosaic
Euphotic belt centreline data is from left to right arranged in three-dimensional cartesian coordinate system by method successively, forms reflection pavement behavior
Curve battle array.Then using cubic surface fitting algorithm, curve battle array is fitted to after cubic surface, and in calculated curve battle array each (from
Scattered) the distance between euphotic belt centreline data point and cubic surface, will this data point when this distance is more than setting value
It is identified as crack point.After all slits point that finally will identify that is clustered, form several minimum envelop cuboids, according to every
The length and width of individual minimum envelop cuboid can be calculated the length and width of every crack, according to each minimum envelop cuboid
Height value can be calculated the depth of every crack.
Obviously, those skilled in the art can carry out the various changes and modification essence without deviating from the present invention to the present invention
God and scope.So, if these modifications of the present invention and modification belong to the scope of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to comprise these changes and modification.
Claims (5)
1. the pavement crack recognition methodss based on camera and laser line generator, are known using based on the pavement crack of camera and laser line generator
Other device, described is included based on the pavement crack identifying device of camera and laser line generator: impulse scaler (4), industrial computer (5),
Vehicle (7), it is fixedly mounted on the laser line generator (2) of headstock center of vehicle (7), at the headstock top center of vehicle (7)
The flat board (8) extending anteriorly out；One end of described flat board (8) is fixed at the headstock top center of vehicle (7), described flat board
(8) lower section of the other end is fixed with area array cameras (1)；The laser head of described laser line generator (2) towards ground, described line laser
, between 40 degree to 80 degree, the camera lens of described area array cameras (1) is vertically for the angle of the laser head of device (2) and horizontal plane
Face；Coaxially it is socketed with incremental rotary encoder (3) in the trailing wheel rotating shaft of described vehicle (7)；Described incremental rotary encoder
(3) input of electrical connection impulse scaler (4), the outfan of described impulse scaler is electrically connected with pci card；Described pci card
It is plugged in the pci draw-in groove of industrial computer (5)；Described area array cameras (1) are connected with pci card by camera link line；Its feature
It is that the described pavement crack recognition methodss based on camera and laser line generator comprise the following steps:
Area array cameras frame speed setting: area array cameras frame speed is set to default frame speed；
Laser emission: drive vehicle at the uniform velocity travel on road surface, laser line generator towards road surface-emission laser, now, on road surface
Form strip euphotic belt；
Trigger produces: during vehicle travels, incremental rotary encoder sends pulse signal to impulse scaler,
Impulse scaler pulse signals are counted；Impulse scaler module just sends once to pci card every n pulse number
Trigger, n is the natural number more than 0；
Pavement image data acquisition: after pci card receives trigger, send control signal to area array cameras, control area array cameras
Road pavement carries out captured in real-time；Area array cameras are sent shooting the pavement image data being formed to industrial computer by pci card；
Strip euphotic belt extracts: after vehicle stops travelling, in industrial computer, all pavement image data of storage is entered respectively
Row Coordinate Conversion, makes all pavement image data change to three-dimensional world coordinate system from camera coordinates system, after Coordinate Conversion
All pavement image data carry out picture smooth treatment respectively；To all pavement image data carrying out after picture smooth treatment
Carry out binary conversion treatment respectively；In the every width pavement image data after binary conversion treatment, extract strip euphotic belt centrage
Pavement image data；
Fitting a straight line: the pavement image data to all strip euphotic belt centrages extracting, it is respectively adopted method of least square
Carry out fitting a straight line, obtain a plurality of straight line；Every straight line is carried out translation transformation and rotation transformation, makes every straight line be located at xoy
In plane, and putting in order between these straight lines is made to be consistent with the acquisition order of corresponding pavement image data；Point
Do not calculate translational movement and the rotation amount of every straight line；
Euphotic belt curve battle array generates: the translational movement according to every straight line and rotation amount, will not carry out all strips of fitting a straight line
The pavement image data of euphotic belt centrage, carries out corresponding translation and rotates, obtain a plurality of curve；Arrangement according to curve is suitable
Sequence chooses curve successively, often chooses m bar curve, just described m bar curve is designated as a curve battle array, m is the nature more than 500
Number；
Crack point identification: respectively surface fitting is carried out to each curve battle array, obtains fitting surface；In calculated curve battle array, each does not enter
Beeline between the pavement image data point of row matching and corresponding fitting surface, if described beeline is more than set
Value, then be identified as crack point by corresponding image data point；
Calculate fracture length, width and depth: identify that crack point carries out clustering processing to all, obtain several minimum envelop
Cuboid, the number of minimum envelop cuboid is the bar number of pavement crack, and the length of each minimum envelop cuboid is as right
Answer the length of pavement crack, the width of each minimum envelop cuboid is the width of corresponding pavement crack, each minimum envelop
The height of cuboid is the depth of corresponding pavement crack.
2. the pavement crack recognition methodss based on camera and laser line generator as claimed in claim 1 are it is characterised in that in face battle array
During the speed setting of camera frame, improve the frame speed of area array cameras by the sampled pixel reducing area array cameras.
3. the pavement crack recognition methodss based on camera and laser line generator as claimed in claim 1 are it is characterised in that adjustment increases
Amount formula rotary encoder and the gear ratio of trailing wheel rotating shaft, make vehicle often advance 5mm, and incremental rotary encoder just produces 10 arteries and veins
Rush signal.
4. the pavement crack recognition methodss based on camera and laser line generator as claimed in claim 1 are it is characterised in that in face battle array
After the speed setting of camera frame, carry out area array cameras demarcation, then carry out Laser emission；Described area array cameras are demarcated and are included following walking
Rapid: to make scaling board first, scaling board arranges uncalibrated image, area array cameras shoot the scaling board of diverse location respectively, collection
Image, demarcates to the image collecting, and obtains the intrinsic parameter of camera and outer parameter；
During strip euphotic belt extracts, the principle according to lens optical imaging and the intrinsic parameter of camera and outer parameter,
Coordinate Conversion is carried out to all pavement image data of storage, so that all pavement image data is changed to three-dimensional from camera coordinates system
World coordinate system.
5. the pavement crack recognition methodss based on camera and laser line generator as claimed in claim 1 are it is characterised in that in laser
During transmitting, fast according to the frame of the sampled distance setting pavement image and area array cameras first, set the first of vehicle traveling
Beginning speed, then drives vehicle and at the uniform velocity travels by described initial velocity on road surface.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310573077.3A CN103669182B (en) | 2013-11-15 | 2013-11-15 | Pavement crack recognition method based on camera and line laser |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310573077.3A CN103669182B (en) | 2013-11-15 | 2013-11-15 | Pavement crack recognition method based on camera and line laser |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103669182A CN103669182A (en) | 2014-03-26 |
| CN103669182B true CN103669182B (en) | 2017-01-25 |
Family
ID=50307898
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310573077.3A Active CN103669182B (en) | 2013-11-15 | 2013-11-15 | Pavement crack recognition method based on camera and line laser |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103669182B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104005325B (en) * | 2014-06-17 | 2016-01-20 | 武汉武大卓越科技有限责任公司 | Based on pavement crack checkout gear and the method for the degree of depth and gray level image |
| CN105277561B (en) * | 2015-10-26 | 2017-11-28 | 镇江苏仪德科技有限公司 | Silicon chip and cell piece regularity detection means and detection method based on machine vision |
| CN107421963A (en) * | 2017-09-18 | 2017-12-01 | 重庆大学 | A kind of nuclear fuel green compact crack detection device and crack detecting method |
| CN108733053A (en) * | 2018-04-23 | 2018-11-02 | 上海圭目机器人有限公司 | A kind of Intelligent road detection method based on robot |
| CN108460767B (en) * | 2018-05-06 | 2023-05-09 | 吉林大学 | Road surface homography active solution device and method for vehicle-mounted road surface crack detection system |
| WO2020026917A2 (en) * | 2018-07-30 | 2020-02-06 | Ricoh Company, Ltd. | Measurement apparatus, movable apparatus, and vehicle |
| CN111207304B (en) * | 2018-11-22 | 2021-09-14 | 北京世纪东方通讯设备有限公司 | Railway tunnel leaky cable vision inspection device and product positioning detection method |
| CN111664830A (en) * | 2020-06-01 | 2020-09-15 | 中公诚科(吉林)工程检测有限公司 | Road longitudinal section elevation and flatness measuring method based on three-dimensional laser scanning |
| CN113420732B (en) * | 2021-08-23 | 2022-02-01 | 深圳市城市交通规划设计研究中心股份有限公司 | Pavement disease detection method and device and storage medium |
| CN113579683A (en) * | 2021-08-23 | 2021-11-02 | 安徽艾瑞思信息科技有限公司 | Device for high-precision picking of assembly line parts based on Falcon camera |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS61281915A (en) * | 1985-06-07 | 1986-12-12 | Kokusai Kogyo Kk | Vehicle device for measuring properties of road surface |
| CN201354024Y (en) * | 2009-03-03 | 2009-12-02 | 长安大学 | High-speed road condition inspection vehicle |
| CN101845788B (en) * | 2010-04-09 | 2013-04-17 | 同济大学 | Cement concrete road surface dislocation detection device and method based on structured light vision |
| CN202055155U (en) * | 2011-03-03 | 2011-11-30 | 上海普勒斯道路交通技术有限公司 | Laser inertia high-speed pavement smoothness detection instrument |
| CN102706880B (en) * | 2012-06-26 | 2014-04-02 | 哈尔滨工业大学 | Road information extraction device based on two-dimensional image and depth information and road crack information detection method based on same |
-
2013
- 2013-11-15 CN CN201310573077.3A patent/CN103669182B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN103669182A (en) | 2014-03-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103669182B (en) | Pavement crack recognition method based on camera and line laser | |
| CN206114546U (en) | Printing defect vision detection system | |
| CN103217111B (en) | A kind of non-contact contact line geometric parameter detection method | |
| CN103217122B (en) | A kind of online dimension measurement method of aluminium alloy wheel hub of vehicle finished product and device | |
| CN101226108B (en) | Method for testing droplet distribution consistency degree | |
| CN107121093A (en) | A kind of gear measurement device and measuring method based on active vision | |
| CN104608799A (en) | Information fusion technology based train wheel set tread damage online detection and recognition method | |
| WO2004046644A2 (en) | Wheel profile inspection apparatus and method | |
| CN1843822A (en) | Dynamic detection system of train wheel pair and detection method thereof | |
| CN102608126A (en) | On-line detection method and device for surface defects of high-temperature continuously cast bloom | |
| CN206563560U (en) | A kind of phone housing quality on-line detecting device based on line laser structured light | |
| CN109974582A (en) | A kind of the conductor diameters non-contact vision detection device and method of automotive wire bundle | |
| CN101832764A (en) | Three-dimensional laser scanning on-line detection profiler | |
| CN103900494A (en) | Homologous point rapid matching method used for binocular vision three-dimensional measurement | |
| CN106248680A (en) | A kind of engine commutator quality detecting system based on machine vision and detection method | |
| CN202548069U (en) | On-line detection device for surface defects of high temperature continuous casting billet | |
| CN108311545A (en) | A kind of y-type rolling mill tandem rolling centering and pass detecting system and method | |
| CN105160754A (en) | Coin surface quality detection apparatus and coin surface quality detection method based on height measurement | |
| CN104458764A (en) | Curved uneven surface defect identification method based on large-field-depth stripped image projection | |
| CN107806824A (en) | The detection method and device of contact net geometric parameter under a kind of lower-speed state | |
| CN107891012B (en) | Pearl size and circularity sorting device based on equivalent algorithm | |
| CN110488261B (en) | LiDAR-based shelf positioning hole detection device and method | |
| CN104528508B (en) | A kind of contactless escalator Multi-parameter Measurement Method based on photoelectricity testing part | |
| CN116989881A (en) | Contact line vibration measurement system and method based on line laser target identification | |
| CN104878679A (en) | Vehicle-mounted rut real-time detection system and detection method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Chen Shengke Inventor after: Han Yi Inventor after: Zhang Xiaochun Inventor after: Chen Zhongyang Inventor after: Wu Guopeng Inventor after: Chen Shengxian Inventor after: Mao Peng Inventor after: Ning Lan Inventor before: Chen Shengke Inventor before: Han Yi Inventor before: Mao Peng Inventor before: Ning Lan |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220411 Address after: 310000 room A325, No. 9, Xiyuan Road, Sandun Town, Xihu District, Hangzhou City, Zhejiang Province Patentee after: Hangzhou Hualu Zhongjian Testing Technology Co.,Ltd. Address before: 200135 room 709-715, Shanhai building, No. 1480, Pudong Avenue, Fengxian District, Shanghai Patentee before: JIAJUE INDUSTRIAL (CHINA) Ltd. |