CN104266592A - Sleeper-based displacement measuring method for mobile device inside tunnel - Google Patents
Sleeper-based displacement measuring method for mobile device inside tunnel Download PDFInfo
- Publication number
- CN104266592A CN104266592A CN201410513066.0A CN201410513066A CN104266592A CN 104266592 A CN104266592 A CN 104266592A CN 201410513066 A CN201410513066 A CN 201410513066A CN 104266592 A CN104266592 A CN 104266592A
- Authority
- CN
- China
- Prior art keywords
- displacement
- tunnel
- sleeper
- image
- mobile device
- 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
- 238000006073 displacement reaction Methods 0.000 title claims abstract description 108
- 238000000034 method Methods 0.000 title claims abstract description 40
- 241001669679 Eleotris Species 0.000 title claims description 56
- 238000005259 measurement Methods 0.000 claims abstract description 41
- 238000009826 distribution Methods 0.000 claims abstract description 7
- 230000006835 compression Effects 0.000 claims description 24
- 238000007906 compression Methods 0.000 claims description 24
- 238000001514 detection method Methods 0.000 claims description 24
- 230000008569 process Effects 0.000 claims description 12
- 230000007547 defect Effects 0.000 claims description 10
- 238000012937 correction Methods 0.000 claims description 7
- 238000001914 filtration Methods 0.000 claims description 6
- 238000003709 image segmentation Methods 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 230000010339 dilation Effects 0.000 claims description 5
- 230000003252 repetitive effect Effects 0.000 claims description 5
- 230000000877 morphologic effect Effects 0.000 claims description 4
- 230000007797 corrosion Effects 0.000 claims description 3
- 238000005260 corrosion Methods 0.000 claims description 3
- 238000003708 edge detection Methods 0.000 claims description 3
- 238000011946 reduction process Methods 0.000 claims description 3
- 238000000691 measurement method Methods 0.000 claims 2
- 238000007405 data analysis Methods 0.000 abstract description 2
- 238000003860 storage Methods 0.000 description 15
- 201000010099 disease Diseases 0.000 description 6
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 6
- 238000009434 installation Methods 0.000 description 6
- 230000033001 locomotion Effects 0.000 description 6
- 230000011218 segmentation Effects 0.000 description 6
- 238000012544 monitoring process Methods 0.000 description 5
- 238000005286 illumination Methods 0.000 description 4
- 230000010354 integration Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000005520 cutting process Methods 0.000 description 2
- 230000003628 erosive effect Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000005021 gait Effects 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 230000008595 infiltration Effects 0.000 description 2
- 238000001764 infiltration Methods 0.000 description 2
- 230000005764 inhibitory process Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000013138 pruning Methods 0.000 description 2
- 238000003786 synthesis reaction Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000011514 reflex Effects 0.000 description 1
- 238000004062 sedimentation Methods 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
Landscapes
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention provides a displacement measuring method for a mobile device inside a track transportation tunnel. More accurate displacement data are provided for the mobile device to obtain a tunnel image. According to the displacement measuring method, the tunnel is divided into a plurality of given sections by means of track sleepers, distance measurement reset signals of displacement sensors are output to pipes in an infrared mode, and the problem that errors caused by long-distance displacement measurement are accumulated continuously is effectively solved; on the basis of repeated measurements and a large amount of data analysis, a probability distribution model of measurement errors, caused when a wheeltrack skids and wheels move in an S-shaped path, of the displacement sensors is obtained, the expectation of the measurement errors of the displacement sensors between the adjacent sleepers is estimated, the measurement errors of the displacement sensors are corrected, and the purpose of reducing the measurement errors is achieved.
Description
Technical field
The present invention relates to a kind of displacement measuring method based on mobile device in the tunnel of sleeper, it may be used for the displacement of advancing of measuring the mobile devices such as tunnel car.
Background technology
In recent years, have relevant unit and scholar to propose tunnel surface information automation acquisition system, this system mainly utilizes optical image technology, installs camera apparatus on a mobile platform, takes high-definition image and record image-capturing positions to subway tunnel.If utilize Digital Image Processing and computer vision technique, intellectualized detection analysis is carried out to subway tunnel image, then can realize subway tunnel intelligent information monitoring and control, significant to the security maintenance of metro operation.On subway tunnel, high-speed mobile shooting tunnel surface image has very large difficulty.Mobile platform, in the process of advancing at utmost speed, produces tractive force by wheel-rail friction, inevitably there is wheel-slip and advance with zigzag due to wheel track gap in this process.This brings very large challenge to the displacement measurement of mobile mobile platform.Due to the shooting of tunnel imaging technique many employings high speed linear array camera, measurement displacement is utilized to carry out image co-registration, so the precision of mobile platform displacement measurement directly affects the accuracy of tunnel imaging.In addition, the sedimentation of tunnel rail affects vehicle operating safety on the one hand, too increases the difficulty of mobile platform displacement measurement on the other hand.
Summary of the invention
The present invention proposes a kind of displacement measuring method based on mobile device in the tunnel of sleeper, utilize image detection algorithm identification track sleeper, obtain the range finding distance of current sleeper displacement transducer, calculate adjacent sleeper pitch and do error analysis with the corresponding sleeper pitch of actual measurement, estimating to skid due to wheel track by repetitive measurement and mass data analysis and wheel zigzag is advanced the displacement transducer range error caused.
Technical scheme of the present invention is achieved in that
A displacement measuring method for mobile device in the tunnel of sleeper, is characterized in that comprising the following steps:
The tie distance of manual measurement test tracks, adds up sleeper number as actual sleeper spacing;
Signal transmitting and receiving pair is installed respectively in the two ends and mobile device in given interval, tunnel, and linear array CCD camera and displacement transducer are fixed on mobile device;
In mobile device moving process, displacement transducer provides displacement signal, and linear array CCD camera obtains the image of tunnel rail and sleeper according to this displacement signal, and records displacement;
Along tunnel bearing of trend, image block compression is stored, noise reduction process;
Utilize Gradient edge detection algorithm and image-region gray feature to Image Segmentation Using, and make morphological dilations corrosion and binary conversion treatment, obtain contour feature;
Utilize graph line detection algorithm, extracted the profile information of sleeper by the contour feature of rail image, and obtained the measuring distance of current sleeper by displacement transducer;
Obtain the measuring distance between adjacent rail sleeper by each sleeper distance of displacement sensor, be called measurement spacing;
Calculate the error between sleeper measurement spacing and actual pitch;
Repetitive measurement, displacement sensor error distribution between statistics adjacent rail sleeper, evaluated error is expected;
In actual track displacement measurement process, utilize image detection algorithm, detect sleeper, utilize above-mentioned error to expect the sleeper distance that correction displacement transducer provides.
In the displacement measuring method based on mobile device in the tunnel of sleeper of the present invention, tunnel is divided between several given areas, between each interval, infrared transceiver device is set.
In the displacement measuring method based on mobile device in the tunnel of sleeper of the present invention, after between given area, the shift value of displacement transducer resets.
In the displacement measuring method based on mobile device in the tunnel of sleeper of the present invention, the noise-reduction method of image comprises image enhaucament and filtering algorithm.
In the displacement measuring method based on mobile device in the tunnel of sleeper of the present invention, described signal transmitting and receiving forms by infrared launcher and receiving trap.
Implement the displacement measuring method based on mobile device in the tunnel of sleeper of the present invention, there is following beneficial effect: tunnel is divided between some given areas by the present invention, utilize infrared tube output displacement sensor instrument distance reset signal, efficiently solve the error that long-distance displacement measurement brings and constantly accumulate problem; Image algorithm is utilized to detect sleeper position, calculate tie distance and compare with actual measurement distance, the error of calculation, analyzed by mass data, draw by the probability Distribution Model of the sensor displacement measuring error that wheel track skids and zigzag causes, evaluated error is expected, thus revises the measuring distance between adjacent rail sleepers, reduces the measuring error between each adjacent rail sleeper.
Accompanying drawing explanation
Fig. 1 is tunnel of the present invention segmentation schematic diagram;
Fig. 2 is mobile device in tunnel under one embodiment of the present invention;
The process flow diagram of the displacement detection method of mobile device in Fig. 3 tunnel of the present invention;
Fig. 4 is the system architecture schematic diagram of Fig. 3;
Fig. 5 represents under the prerequisite that there is skidding and the advance of S route, the displacement curve that car body actual motion displacement curve and sensor record.
Fig. 6 is two width figure to be spliced;
Fig. 7 is the splicing charts for finned heat of two width figure splicing regions;
Fig. 8 is the splicing result of Fig. 6.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described.
With reference to Fig. 4, displacement transducer can be a measuring wheel, and generally, wheel is considered as uniform motion.Due to the existence of skidding and S advances, car body actual motion speed changes, and car body displacement is generally less than the displacement that displacement transducer records, as Fig. 5.In actual applications, mobile device needs shooting tunnel image and demarcates camera site, therefore the gait of march that recorded by displacement transducer and displacement correction is needed to be actual speed and the displacement of car body, the difference of degree of namely advancing according to the skidding of each position and S, suitably cuts down the displacement data that displacement transducer records.
As Fig. 1, the displacement measuring method based on mobile device in the tunnel of sleeper of the present invention, comprises the following steps:
S01: be divided in experiment test tunnel between several given areas, such as length of tunnel 800m, be divided into eight intervals that 100m is long, there is again multiple sleeper in each interval, as shown in Figure 2.Signal transmitting and receiving pair is installed respectively in the two ends and mobile device in given interval, tunnel, and linear array CCD camera and displacement transducer are fixed on mobile device, with reference to Fig. 4.
S02: in mobile device moving process, displacement transducer provides displacement signal, linear array CCD camera according to this displacement signal obtain rail in tunnel and image.The image of linear array CCD camera is merged, obtains image I.And by fixing unit length, { I is stored to image block compression along tunnel bearing of trend
1, I
2... I
n, n represents image block number.
S03: along tunnel bearing of trend, image block compression is stored, noise reduction process.Image segmentation algorithm selects gradient operator
wherein, G
x, G
ybe respectively the image detected through transverse direction and longitudinal edge, B
x, B
yfor 3*3 parameter matrix.Coarse contour is characterized as
the half-tone information in image outline region is as gray feature
utilize histogram equalization and gaussian filtering method, to I
i, i ∈ 1,2 ..., n} carries out strengthening and filtering, reduces illumination effect and picture noise.The discrete Gaussian filter function of two dimension zero-mean
wherein σ is noise variance.The noise of this wave filter to the Normal Distribution in image has good inhibition.Image filtering result is designated as
S04: utilize Gradient edge detection algorithm and image-region gray feature to Image Segmentation Using, and make morphological dilations corrosion and binary conversion treatment, obtain contour feature.To coarse contour feature
carry out dilation erosion process, pruning refinement is carried out to coarse contour feature, obtain better segmentation effect, be designated as
right
carry out the binary conversion treatment of adaptive threshold, obtain the contour feature of image
S05: utilize graph line detection algorithm, by the profile information of tunnel rail profile feature extraction sleeper, and is obtained the position of current sleeper by displacement transducer.Show that tunnel sleeper actual pitch is respectively s by the length (this length is measured in advance, the spacing that adjacent infrared receiving/transmission is right) between given area
12, s
23..., s
n (n-1), n is the quantity of the track sleeper identified in given area, and displacement transducer can measure the moving displacement s ' of mobile device between tunnel sleeper
12, s '
23..., s '
n (n-1), Δ
i (i-1)=s '
i (i-1)-s
i (i-1), i ∈ 2,3 ... n} obtains sleeper pitch measuring error.
S06: repetitive measurement, the measuring error of comparative analysis sleeper pitch, obtains the probability Distribution Model of displacement sensor error, and then draws the expectation of measuring error
S07: to actual measurement track, utilizes above-mentioned image-recognizing method, detects track sleeper position, obtains measuring distance S by displacement transducer
i, corrected range is
wherein, i represents the counting numbering of current sleeper, and sleeper counting is from 1.
In the present invention, tunnel inner wall is arc, and area is large, and single camera cannot obtain all images in whole tunnel, needs multiple CCD camera to coordinate and realizes.The image that CCD camera obtains can intersect mutually, therefore after single camera obtains image, needs to splice image.Therefore after obtaining the image of the fixed measure of single camera, utilize infrared laser to mark the image obtained different CCD camera and carry out splicing fusion, specific as followsly to state.
Greyscale transformation, hard-threshold segmentation are carried out to image, obtain two-value contour feature figure, Hough algorithm is adopted to detect the straight line of contour feature figure, straight line should be less than 5 DEG C with the angle at least one limit of image, two straight lines being less than setting threshold xi of adjusting the distance regard as repetition straight line, only retain wherein one.Mark lengths is greater than the straight line of setting value, and rectilinear coordinates are pos (pos is horizontal ordinate or the ordinate of straight line), and image mosaic interval is [pos – index, pos+index], and index is width parameter between splice region.Threshold xi is less than 10 pixels.
Generate splicing figure parameters y
x=0.5*exp (-0.5*x
2/ σ
2), wherein, x ∈ [-index, index], σ=20; Stitching image
Wherein j is the coordinate on image, I
jdenotation coordination is the pixel line of j, I
1, I
2represent two adjacent width images, I
1jrepresent I
1upper coordinate is the pixel line of j, pos
maxrepresent I
1, I
2one that middle pos coordinate is larger, pos
minrepresent I
1, I
2one that middle pos coordinate is less, W=max (pos1, pos2)+W
2the width that-min (pos1, pos2) is composograph.I
1, I
2the laser rays position pos that two width images are corresponding is different, is divided into pos1, pos2; For larger pos value (pos
max) figure we retain its left half figure, figure (pos of less pos value
min) we retain right half part, the size of composograph is: max (pos1, pos2)+W
2-min (pos1, pos2), the image coordinate system after synthesis is consistent with left half-image.Outside integration region, stitching image is respectively at I
1, I
2unanimously, in integration region (interval [pos – index, pos+index]), stitching image is by I
1, I
2be multiplied by its splicing parameter respectively to form.The value of splicing parameter changes along with the change of coordinate j, and Fig. 7 shows the splicing charts for finned heat in image mosaic region, Fig. 6 and Fig. 8 shows the image before and after splicing.
Displacement detection method of the present invention can adopt the system shown in Fig. 3 to realize, and it is made up of main control unit, linear array CCD camera, image compression unit, image storage unit, isochronous controller, displacement transducer etc.The input end of this isochronous controller is connected to described main control unit, output terminal is connected to described image compression unit, the output terminal of this image compression unit is connected to described linear array CCD camera, the output terminal of this image compression unit is also connected to described image storage unit, and the output terminal of this image storage unit is connected to described main control unit.Displacement transducer provides raw bits shifting signal, view data is captured successively by main control unit, isochronous controller, image compression unit control line array CCD camera, the displacement signal that this view data is communicated with this moment is supplied to image compression unit by linear array CDD camera, and image, displacement data are sent to image storage unit by image compression unit.View data is supplied to main control unit by image storage unit, and main control unit, to Image Segmentation Using, detection, obtains sleeper position.In addition, for ensureing brightness of image, can also arrange illumination compensation system, it can comprise area source.Movable storage device can so that the data reading that image storage unit etc. stored.
The displacement detection method of mobile device of the present invention can revise the error of wheel range finding, realizes mobile device, such as, measures the displacement of car.It may be used in tunnel system, the identification of such as tunnel defect and the high speed detection of tunnel defect.Tunnel system all requires to provide a displacement transducer to the measurement of tunnel inner wall, and this displacement transducer can obtain the speed of body movement.Because single wheel intelligence provides the speed of wheel, need to adopt displacement monitoring method of the present invention to be revised.
The high-speed detection system of tunnel defect at least comprises moveable detection platform, main control unit, linear laser, area array CCD camera, displacement transducer, isochronous controller, high speed memory modules, memory module, correction module and inertial navigator.Detection platform is the bogey of native system, and it can be tunnel dolly, and in the process that this tunnel dolly is walked in orbit, the image information of the acquisition tunnel medial surface that system is real-time, obtains full-view image.Main control unit is central processing unit part of the present invention, and each unit data of major control transmit line correlation calculating etc. of going forward side by side.Image compression unit is stored to high speed memory modules by after compression of images through main control unit.Linear laser is arranged in detection platform, this linear laser launches linear laser, and linear laser is beaten inside tunnel, reflexes to area array CCD camera, visible ray cutoff filter visible light on camera, the reflected light that camera only obtains infrared laser line forms highlighted lines.Area array CCD camera is arranged in described detection platform.It is ccd image sensor, CCD has the electric capacity of many marshallings, can respond to light, via the control of external circuit, each small capacitances can by its with electric charge be given to its adjacent electric capacity, final and image is transformed into digital signal.Displacement transducer slides in orbit, detects the position of dolly, to measure omnidistance tunnel according to tachometric survey.Isochronous controller controls described control image compression unit according to the shift value of displacement transducer, and image compression unit chain of command array CCD camera is made a video recording, and this isochronous controller can be accommodated in main control unit.After system installation, with memory module record installation parameter, comprise the length of installation baseline, linear laser and the angle of this installation baseline, the focal length of area array CCD camera.This parameter is stored in described memory module, identical with isochronous controller, and this memory module is accommodated in described main control unit.Correction module is all memory device, and when first time measures, this correction module provides an empirical value, after this revises gradually, final storage one optimum corrected parameter offset.This corrected parameter can be revised and measure dolly, tunnel track etc. to the impact of measurement result.Horizontal direction and the working direction of inertial navigator measurement detection platform shake, and provide corrected parameter according to this horizontal vibrating
the high-speed detection system of this tunnel of the present invention defect adopts linear laser as light source, and many group detection platform real time emission high intensity laser beam signals, can obtain highdensity testing result.The present invention simultaneously adopts unique space modeling method, can carry out high-precision three-dimensional modeling exactly to tunnel inner wall.
The high speed detection method of tunnel defect, adopts following detecting step:
S01: be fixed in identical platform by linear laser 2 and area array CCD camera 3, camera lens and the linear laser of area array CCD camera are in same plane.Store the length s of the installation baseline that linear laser 2 limits with the focus of area array CCD camera 3, linear laser 2 and the angle β of this installation baseline and the focal distance f of area array CCD camera 3.Focus is F.FL＝s，P
1LP＝β，FF
1＝f。O is the mid point of FL, so FO is also known quantity.
S02: linear laser Emission Lasers line, area array CCD camera 3 obtains laser reflection image, identify laser beam, with image center (this image center can be the projection of focus on image-forming component) for initial point sets up coordinate system, the direction of X-axis is perpendicular to laser rays.Foundation
calculate laser rays mid point P
1, the N number of mid point P of continuous print
1composition beats the center line 01 of the laser rays on tunnel, and this center line 01 corresponds to the center line 03 on image-forming component 32.
S03: calculate mid point P
1with the distance of linear laser 2
wherein offset is corrected parameter, and PixelSize is Pixel Dimensions, x
1for the picture P of mid point
1' and true origin between number of pixels.In the accompanying drawings, d
1=P
1l.P
1l, OL and β, according to sine, can calculate P
1o.Known with the moving direction of inspection vehicle for X, the direction of laser centerline is Y, x
1corresponding P
1' horizontal ordinate.
S04: according to three cosine laws, calculates each point P outside mid point
2with the distance of linear laser 2.On image-forming component 32, P
1' and P
2' between distance obtained by number of pixels and Pixel Dimensions between point.Triangle P
1' P
2' F triangle P
1p
2f is similar, P
1' F and P
1f and known, can calculate P
1p
2.P again
1o is known, can calculate P
2o, P
1oP
2.By P
1o and P
1l and β, can calculate P
1oP.P
1p2 perpendicular to face P1FL, according to three cosine laws, cosP
2oP=cosP
1oP
2* cosP
1oP, known P
2oP, OL and P
2o, can calculate P
2l.Known mid point P
1, the distance of other each points and laser instrument beyond mid point on laser rays can be calculated, in other words, calculating the position of each mid point on center line 01, the position of each point on all laser beams can be calculated.
S05: when measuring first, offset substitutes into empirical value, and repetitive measurement, determines the impact of this offset on measurement result, and determine offset value according to measurement result.The introducing of Offset parameter can reduce special mounting position and operating mode to the impact of measurement result.After drawing the correlativity of offset and measurement result, according to measurement result bias direction, substitute into different offset, to reduce the error of measurement result.
S06: measure the vibrations error detecting dolly level in advance process
and each measured distance to current tunnel cross section
revise
the vibrations error of level is measured by inertial navigator, and data are supplied to main control unit 1.In the present invention, inertial navigator can measure the error in xyz tri-directions, and revises measurement result.
S07: utilize mobile vehicle to carry out panorama measurement to whole tunnel, obtains the three dimensions cloud data in tunnel, according to cloud data reconstruction tunnel model.Tunnel survey speed is fast, and the preferred high speed memory modules 9 that adopts stores view data, and compresses by image compression unit 31 pairs of image informations.
Above-mentioned steps details feasible computing method, for the purpose of the present invention, outbalance be adopt wire infrared laser to measure, computing formula is had no particular limits, according to the difference of accuracy requirement, also can adopt other feasible computing method.
Can obtain more satisfactory measurement result main cause is utilize camera calibration, range finding curve correcting parameter, the technology such as sub-pix segmented positioning algorithm ensure that accuracy and the stability of system, the tunnel defect inspection system utilizing the method to design can carry out modeling reconstruct to tunnels such as subways preparatively, and tunnel deformation monitoring becomes simple efficient.
The recognition system of tunnel defect, can be made up of main control unit, linear array CCD camera, image compression unit, image storage unit, isochronous controller, displacement transducer etc.The input end of this isochronous controller is connected to described main control unit, output terminal is connected to described image compression unit, the output terminal of this image compression unit is connected to described linear array CCD camera, the output terminal of this image compression unit is also connected to described image storage unit, and the output terminal of this image storage unit is connected to described main control unit.Displacement transducer provides raw bits shifting signal, view data is captured successively by main control unit, isochronous controller, image compression unit control line array CCD camera, the displacement signal that this view data is communicated with this moment is supplied to image compression unit by linear array CDD camera, and image, displacement data are sent to image storage unit by image compression unit.View data is supplied to main control unit by image storage unit, and main control unit splices image, analysis, obtains disease position.In addition, for ensureing brightness of image, can also arrange illumination compensation system, it can comprise area source.Movable storage device can so that the data reading that image storage unit etc. stored.
The recognition methods of tunnel defect, mainly comprises the steps:
S1: obtain image.Utilize multiple high speed linear array CCD camera, displacement transducer and illumination compensation auxiliary lighting system to obtain tunnel internal high-definition image, obtain image I, and by fixing unit length, { I is stored to image block compression along tunnel bearing of trend
1, I
2... I
n, n represents image block number.For a kind of shooting results of disease.
S2: by I
i, i ∈ 1,2 ..., n} is converted into gray level image, and carries out denoising, and denoising result is
profit is rung and picture noise.The discrete Gaussian filter function of two dimension zero-mean
wherein σ is noise variance.The noise of this wave filter to the Normal Distribution in image has good inhibition.Image filtering result is designated as
S3: right
carry out Iamge Segmentation, extract coarse contour feature
and gray feature
image segmentation algorithm selects gradient operator
wherein, G
x, G
ybe respectively the image detected through transverse direction and longitudinal edge, B
x, B
yfor 3*3 parameter matrix.Coarse contour is characterized as
the half-tone information in image outline region is as gray feature
S4: to coarse contour feature
carry out Morphological scale-space and to feature binaryzation, obtain
to coarse contour feature
carry out dilation erosion process, pruning refinement is carried out to coarse contour feature, obtain better segmentation effect, be designated as
right
carry out the binary conversion treatment of adaptive threshold, obtain the contour feature of image
S5: set up tunnel surface normal profile feature and Disease Characters database E
1, E
2, E
3, E
4.Utilize above step to carry out contour feature and gray feature extraction to testing tunnel, utilize a large amount of test result to set up normal tunnel surface profile property data base E
1, fracture profile property data base E
2, infiltration contour feature database E
3and contour feature E is peeled off in lining cutting
4.
S6: monitoring tunnel surface.Above-mentioned identical algorithms is utilized to extract tunnel surface contour feature and gray feature, by comparing with known type contour feature database, carry out contour feature classification, obtain current monitoring tunnel surface contour feature type, and improve classification accuracy by repeated detection.
S7: tunnel surface disease parameter calculates.Disease contour feature is utilized to estimate relevant disease parameter: infiltration area, area etc. is peeled off in fracture width and lining cutting.
In the present invention, tunnel inner wall is arc, and area is large, and single camera cannot obtain all images in whole tunnel, needs multiple CCD camera to coordinate and realizes.The image that CCD camera obtains can intersect mutually, therefore after single camera obtains image, needs to splice image.Therefore after single camera obtains image, preferably carry out following steps: greyscale transformation, hard-threshold segmentation are carried out to image, obtain two-value contour feature figure, Hough algorithm is adopted to detect the straight line of contour feature figure, straight line should be less than 5 DEG C with the angle at least one limit of image, two straight lines being less than setting threshold xi of adjusting the distance regard as repetition straight line, only retain wherein one.Mark lengths is greater than the straight line of setting value, and rectilinear coordinates are pos (pos is horizontal ordinate or the ordinate of straight line), and image mosaic interval is [pos – index, pos+index], and index is width parameter between splice region.Threshold xi is less than 10 pixels.Generate splicing figure parameters y
x=0.5*exp (-0.5*x
2/ σ
2), wherein, x ∈ [-index, index], σ=20; Stitching image
Wherein j is the coordinate on image, I
jdenotation coordination is the pixel line of j, I
1, I
2represent two adjacent width images, I
1jrepresent I
1upper coordinate is the pixel line of j, pos
maxrepresent I
1, I
2one that middle pos coordinate is larger, pos
minrepresent I
1, I
2one that middle pos coordinate is less, W=max (pos1, pos2)+W
2the width that-min (pos1, pos2) is composograph.I
1, I
2the laser rays position pos that two width images are corresponding is different, is divided into pos1, pos2; For larger pos value (pos
max) figure we retain its left half figure, figure (pos of less pos value
min) we retain right half part, the size of composograph is: max (pos1, pos2)+W
2-min (pos1, pos2), the image coordinate system after synthesis is consistent with left half-image.Outside integration region, stitching image is respectively at I
1, I
2unanimously, in integration region (interval [pos – index, pos+index]), stitching image is by I
1, I
2be multiplied by its splicing parameter respectively to form.The value of splicing parameter changes along with the change of coordinate j.
Displacement transducer can be a measuring wheel, and generally, wheel is considered as uniform motion.Due to the existence of skidding and S advances, car body actual motion speed changes, and car body displacement is generally less than the displacement that displacement transducer records.In actual applications, mobile device needs shooting tunnel image and demarcates camera site, therefore the gait of march that recorded by displacement transducer and displacement correction is needed to be actual speed and the displacement of car body, the difference of degree of namely advancing according to the skidding of each position and S, suitably cuts down the displacement data that displacement transducer records.
The unaccomplished matter of the application, can see described in 201410275604.7 and 201410275647.5.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (7)
1., based on a displacement measuring method for mobile device in the tunnel of sleeper, it is characterized in that comprising the following steps:
The tie distance of manual measurement test tracks, adds up sleeper number as actual sleeper spacing;
Signal transmitting and receiving pair is installed respectively in the two ends and mobile device in given interval, tunnel, and linear array CCD camera and displacement transducer are fixed on mobile device;
In mobile device moving process, displacement transducer provides displacement signal, and linear array CCD camera obtains the image of tunnel rail and sleeper according to this displacement signal, and records displacement;
Along tunnel bearing of trend, image block compression is stored, noise reduction process;
Utilize Gradient edge detection algorithm and image-region gray feature to Image Segmentation Using, and make morphological dilations corrosion and binary conversion treatment, obtain contour feature;
Utilize graph line detection algorithm, extracted the profile information of sleeper by the contour feature of rail image, and obtained the measuring distance of current sleeper by displacement transducer;
Obtain the measuring distance between adjacent rail sleeper by each sleeper distance of displacement sensor, be called measurement spacing;
Calculate the error between sleeper measurement spacing and actual pitch;
Repetitive measurement, displacement sensor error distribution between statistics adjacent rail sleeper, evaluated error is expected;
In actual track displacement measurement process, utilize image detection algorithm, detect sleeper, utilize above-mentioned error to expect the sleeper distance that correction displacement transducer provides.
2. the displacement measuring method based on mobile device in the tunnel of sleeper according to claim 1, is characterized in that, is divided in tunnel between several given areas, arranges infrared transceiver device between each interval.
3. the displacement measuring method based on mobile device in the tunnel of sleeper according to claim 1, is characterized in that, after between given area, the shift value of displacement transducer resets.
4. the displacement measuring method based on mobile device in the tunnel of sleeper according to claim 1, is characterized in that, the noise-reduction method of image comprises image enhaucament and filtering algorithm.
5. the displacement measuring method based on mobile device in the tunnel of sleeper according to claim 1, it is characterized in that, described signal transmitting and receiving forms by infrared launcher and receiving trap.
6. a recognition methods for tunnel defect, is characterized in that, adopts the displacement of the displacement measurement method measurement update platform of any one described in claim 1 to 5.
7. a high speed detection method for tunnel defect, is characterized in that, adopts the displacement of the displacement measurement method correct detection platform described in claim 1 to 5 any one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410513066.0A CN104266592A (en) | 2014-09-29 | 2014-09-29 | Sleeper-based displacement measuring method for mobile device inside tunnel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410513066.0A CN104266592A (en) | 2014-09-29 | 2014-09-29 | Sleeper-based displacement measuring method for mobile device inside tunnel |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104266592A true CN104266592A (en) | 2015-01-07 |
Family
ID=52158134
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410513066.0A Pending CN104266592A (en) | 2014-09-29 | 2014-09-29 | Sleeper-based displacement measuring method for mobile device inside tunnel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104266592A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106327475A (en) * | 2016-08-26 | 2017-01-11 | 曾美枝 | Image processing method and device |
| CN108204985A (en) * | 2018-01-16 | 2018-06-26 | 上海同岩土木工程科技股份有限公司 | A kind of shield tunnel mileage localization method |
| CN108225202A (en) * | 2016-12-10 | 2018-06-29 | 南京理工大学 | Steel rail displacement on-line measurement system and measuring method based on line array CCD |
| CN108596968A (en) * | 2018-04-02 | 2018-09-28 | 成都精工华耀科技有限公司 | A kind of taking count of crosstie method based on track 3D depth images |
| CN112325935A (en) * | 2020-10-30 | 2021-02-05 | 湖北省水利水电规划勘测设计院 | Safety performance monitoring system of deeply buried tunnel |
| CN114612364A (en) * | 2020-11-24 | 2022-06-10 | 中山大学附属第七医院(深圳) | A screening method for asymptomatic people with novel coronavirus pneumonia |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1457191A (en) * | 2002-05-10 | 2003-11-19 | 原相科技股份有限公司 | Motion Correction and Compensation Methods |
| JP2009115587A (en) * | 2007-11-06 | 2009-05-28 | Mitsubishi Heavy Ind Ltd | In-vehicle apparatus for detecting running route |
| US20100185390A1 (en) * | 2007-07-04 | 2010-07-22 | Yasuhiro Monde | Navigation system |
| CN102211590A (en) * | 2011-04-07 | 2011-10-12 | 同济大学 | Detecting vehicle for tunnel |
| CN102706288A (en) * | 2012-06-19 | 2012-10-03 | 上海地铁盾构设备工程有限公司 | Posture compensation method and device based on image measurement and tunnel settlement and deformation monitoring system |
| CN104048969A (en) * | 2014-06-19 | 2014-09-17 | 樊晓东 | Tunnel defect recognition method |
| CN104048970A (en) * | 2014-06-19 | 2014-09-17 | 樊晓东 | High-speed detection system and method of tunnel defects |
-
2014
- 2014-09-29 CN CN201410513066.0A patent/CN104266592A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1457191A (en) * | 2002-05-10 | 2003-11-19 | 原相科技股份有限公司 | Motion Correction and Compensation Methods |
| US20100185390A1 (en) * | 2007-07-04 | 2010-07-22 | Yasuhiro Monde | Navigation system |
| JP2009115587A (en) * | 2007-11-06 | 2009-05-28 | Mitsubishi Heavy Ind Ltd | In-vehicle apparatus for detecting running route |
| CN102211590A (en) * | 2011-04-07 | 2011-10-12 | 同济大学 | Detecting vehicle for tunnel |
| CN102706288A (en) * | 2012-06-19 | 2012-10-03 | 上海地铁盾构设备工程有限公司 | Posture compensation method and device based on image measurement and tunnel settlement and deformation monitoring system |
| CN104048969A (en) * | 2014-06-19 | 2014-09-17 | 樊晓东 | Tunnel defect recognition method |
| CN104048970A (en) * | 2014-06-19 | 2014-09-17 | 樊晓东 | High-speed detection system and method of tunnel defects |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106327475A (en) * | 2016-08-26 | 2017-01-11 | 曾美枝 | Image processing method and device |
| CN108225202A (en) * | 2016-12-10 | 2018-06-29 | 南京理工大学 | Steel rail displacement on-line measurement system and measuring method based on line array CCD |
| CN108225202B (en) * | 2016-12-10 | 2021-05-04 | 南京理工大学 | On-line measurement system and measurement method of rail displacement based on linear CCD |
| CN108204985A (en) * | 2018-01-16 | 2018-06-26 | 上海同岩土木工程科技股份有限公司 | A kind of shield tunnel mileage localization method |
| CN108596968A (en) * | 2018-04-02 | 2018-09-28 | 成都精工华耀科技有限公司 | A kind of taking count of crosstie method based on track 3D depth images |
| CN108596968B (en) * | 2018-04-02 | 2022-04-19 | 成都精工华耀科技有限公司 | Sleeper counting method based on track 3D depth image |
| CN112325935A (en) * | 2020-10-30 | 2021-02-05 | 湖北省水利水电规划勘测设计院 | Safety performance monitoring system of deeply buried tunnel |
| CN114612364A (en) * | 2020-11-24 | 2022-06-10 | 中山大学附属第七医院(深圳) | A screening method for asymptomatic people with novel coronavirus pneumonia |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104266591A (en) | Displacement detection method for moving device in tunnel | |
| CN104266592A (en) | Sleeper-based displacement measuring method for mobile device inside tunnel | |
| US20040095585A1 (en) | Wheel profile inspection apparatus and method | |
| Acikgoz et al. | Sensing dynamic displacements in masonry rail bridges using 2D digital image correlation | |
| CN101021417B (en) | Long straight rail geometric parameter photogrammetry | |
| CN108802043B (en) | Tunnel detection device, tunnel detection system and tunnel defect information extraction method | |
| RU2018103751A (en) | SYSTEM AND METHOD FOR MONITORING RAILWAYS | |
| JP5494286B2 (en) | Overhead position measuring device | |
| CN106871805A (en) | vehicle-mounted rail gauge measuring system and measuring method | |
| CN104048970A (en) | High-speed detection system and method of tunnel defects | |
| CN104048969A (en) | Tunnel defect recognition method | |
| KR20120022943A (en) | Device for measuring displacement of pantograph and method for detecting hard spot of trolley wire | |
| EP2966400B1 (en) | Overhead line position measuring device and method | |
| JPWO2018179281A1 (en) | Object detection device and vehicle | |
| KR102315405B1 (en) | Apparatus and method for inspecting railway line | |
| CN104406523A (en) | Shield tunnel segment shift detection method | |
| CN115147566A (en) | Modeling method, system and device of tunnel lining three-dimensional structure based on multi-line laser | |
| Adham et al. | Railway tracks detection of railways based on computer vision technique and GNSS data | |
| EP4082867B1 (en) | Automatic camera inspection system | |
| CN205991784U (en) | A kind of railway tunnel gauge dynamic detection system based on industrial computer control | |
| CN105674893B (en) | Absolute grating scale based on cmos image sensor and its measurement method | |
| Qiu et al. | Non-disruptive rail track geometry measurement system using an unmanned aerial vehicle and a light detection and ranging sensor | |
| CN120068001B (en) | Traffic track deformation detection method and system based on multimodal three-dimensional point cloud fusion | |
| JP2015068641A (en) | Stereo image processing device | |
| Alidoost et al. | Development of a machine vision system for damage and object detection in tunnels using convolutional neural networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150107 |
|
| WD01 | Invention patent application deemed withdrawn after publication |