CN106996748A - A kind of wheel footpath measuring method based on binocular vision - Google Patents
A kind of wheel footpath measuring method based on binocular vision Download PDFInfo
- Publication number
- CN106996748A CN106996748A CN201710167507.XA CN201710167507A CN106996748A CN 106996748 A CN106996748 A CN 106996748A CN 201710167507 A CN201710167507 A CN 201710167507A CN 106996748 A CN106996748 A CN 106996748A
- Authority
- CN
- China
- Prior art keywords
- dimensional
- wheel
- method described
- camera
- image
- 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
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/08—Measuring arrangements characterised by the use of optical techniques for measuring diameters
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B5/00—Measuring arrangements characterised by the use of mechanical techniques
- G01B5/0025—Measuring of vehicle parts
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a kind of wheel footpath measuring method based on binocular vision.This method realizes three-dimensional correction to taking turns using Bouguet algorithms to binocular image plane progress re-projection first, and the simplified NCC algorithms of application realize the Stereo matching of characteristic point.Obtained wheel will be extracted by three-dimensional reconstruction and coordinate transform to be mapped in two dimensional surface Edge Feature Points and finally fit wheel footpath parameter, experiment is detected to flange radius by taking turns, the error for demonstrating this method is less than 0.1mm, with preferable accuracy of detection, it is larger and the problem of be difficult to three-dimensional fitting that the invention solves current measuring method characteristic point Stereo matching error, meets the requirement that actual field is quickly measured.
Description
Art
The invention belongs to field of visual inspection, it is related to binocular identification and matching technique, the particularly energy of line-structured light guiding
Re-projection is enough carried out to wheel footpath binocular image plane by Bouguet algorithms and realizes three-dimensional correction, and the simplified NCC algorithms of application
Feature Points Matching is realized, obtained parameter error is finally fitted smaller.
Background technology
Wheel is taken turns and the accuracy of modular construction parameter is measured to Bogie Designs to being the important running part of rail vehicle
Manufacture and characteristic performance analysis suffer from significance, in actual moving process, due to there may be track irregularity, track
Profile and its material matching is incorrect and too big to etc. many reasons of tractive force, causes abrasion quickening of the wheel to each several part, have impact on
The normal operation of rail vehicle.
China is Mechanical measurement to taking turns the Main Means measured structural parameters, and this measuring method efficiency is low and measures
It is more inaccurate.With the development of machine vision technique, the image processing method measurement based on ccd video camera has been achieved for necessarily
Progress, this kind of method has the advantages that noncontact, detection speed are fast, but current measuring method is probably due to wheel is smooth to surface
Degree is high, and camera optical axis can not possibly be perpendicular to wheel to surface during reflective strong and measurement so that Edge Feature Points are extracted and vertical
Body matching is difficult, and the wheel footpath parameter that final calculating is obtained has larger error.
To solve above problem, by designing a kind of binocular vision non-contact detecting technology based on line-structured light, pass through
The three-dimensional correction of image and the NCC algorithms simplified realize the Stereo matching of characteristic point, are re-introduced into coordinate transform thought by characteristic point
Transform to the fitting and calculating that wheel footpath parameter is realized in two dimensional surface.
The content of the invention
The present invention can effectively realize the feature point extraction and Stereo matching of image, and the algorithm is shown not by real example
Only measure precisely and disclosure satisfy that the requirement of actual field quick detection.
The technical scheme that the present invention solves above-mentioned technical problem is to propose that a kind of binocular vision based on line-structured light is non-to connect
Touch detection technique.Its detailed process is as follows:
Step 1: using the good scaling board of pre-production, carrying out intrinsic parameter demarcation to left and right camera, obtaining left and right camera
Intrinsic parameter and stereo calibration parameter.
Step 2: the pre-fixed shooting equipment of regulation, makes what left and right cameras was irradiated to line-structured light respectively
Wheel shoots 5 width images, and will shoot obtained Digital Image Transmission to image processor to taking pictures in real example.
Step 3: the wheel footpath image gathered to industrial camera carries out image preprocessing work, its specific steps includes:
Step 3-1:Medium filtering is carried out to image, sequential scanning is carried out along image by the filter window of medium filtering,
So that the pixel larger with neighboring pixel gray scale difference eliminates most of noise close to neighboring pixel value with this.
Step 3-2:Three-dimensional correction is carried out to two images, passes through the plane of delineation of the Bouguet algorithms to two video cameras
Carry out re-projection so that they accurately fall in approximately the same plane, and image row be perfectly aligned over it is preceding to parallel
In structure.
Step 3-3:Stereo matching is carried out to characteristic point, using simplified NCC algorithms, according to binocular wheel to correction chart as
Feature, method of the wheel to Edge Feature Points is searched for using along line direction, completes Stereo matching of the wheel to external diameter characteristic point.
Step 4: obtain taking turns after the feature point coordinates for matching left and right camera by Stereo matching, it is three-dimensional with reference to video camera
Demarcation, according to Bouguet algorithms, pixel can be mapped in three-dimensional by re-projection, marked further according to binocular camera solid
Fixed result, obtains re-projection matrix Q.
Step 4-1:The parallax d that characteristic point is associated can be obtained by Stereo matching.
Step 4-2:The parallax d associated according to characteristic point, passes through coordinate transformProjecting characteristic points are arrived
Among three dimensional coordinate space relative to left video camera, error during due to installing, left and right camera can not can guarantee that and be substantially parallel,
It is also impossible to guarantee and is exactly perpendicularly to wheel to plane, therefore takes turns to place plane relative to camera coordinate system { Oc-XcYcZc}
(hereinafter referred to as { C }) is space plane, therefore, setting up new Fitting Coordinate System { OL-XLYLZL(hereinafter referred to as { L }).
Step 4-3:According to the obtained three-dimensional coordinate (X/W, Y/W, Z/W) under left camera coordinates, by sitting
Mark transformLPi=(CRL)T·Pi+P1, (i=1 ... n) is mapped to fitted coordinate system, realizes two dimensional surface.
Step 5: carrying out plane fitting to the three-dimensional data of external diameter feature point extraction according to obtained wheel, and obtain
Wheel is to the fit Plane π (a, b, c, d) relative to camera coordinate system.
Step 5-1:According to obtained fit Plane parameter, fitted coordinate system is further obtained relative to left video camera
The transformation matrix of coordinate systemCTL。
Step 5-2:Parameter fitting is carried out in the two dimensional surface where wheel footpath characteristic point, and with P1For origin OL, P1Point to
PnDirection be XL, using the normal direction of π planes as ZL, YL=ZL×XL, thenWhereinBy P1, P2…PnPass through formulaLPi=(CRL)T·Pi+P1, (i=1 ... n) can be transformed into { L } by { C }, by
InLPiIn π, therefore it can carry out justifying fitting in the plane and obtain final wheel pair radius R.
The invention provides a kind of binocular vision wheel footpath measuring method based on line-structured light, this method can be effectively real
The feature point extraction and Stereo matching of existing image, show that the invention not only accurate positioning but also disclosure satisfy that reality by real example
The requirement of field quick detection.
Brief description of the drawings
Fig. 1 is measuring process FB(flow block) of the wheel to parameter.
Fig. 2 is that NCC matches schematic diagram.
Fig. 3 is three-dimensional reconstruction of the wheel to external diameter Edge Feature Points.
Fig. 4 calculates schematic diagram for wheel pair radius.
Fig. 5 is detection means schematic diagram.
Embodiment
It is the binocular vision detection method flow block diagram based on cable architecture as shown in Figure 1, it is below according to accompanying drawing and specifically real
Implementation of the example to the present invention is described further.
Step one:Using the good scaling board of pre-production, intrinsic parameter demarcation is carried out to left and right camera, left and right camera is obtained
Intrinsic parameter and stereo calibration parameter matrix, the relative position of the Intrinsic Matrix of left and right cameras and left and right cameras in real example
Appearance matrix is respectively:
Step 2:The pre-fixed shooting equipment of regulation, as shown in figure 5, making left and right cameras respectively to knot
The wheel of structure light irradiation is to taking pictures, from S1Place photographs SnPlace.Take turns to some row points P on outline1, P2…PnIn left and right phase
Corresponding imaging point a is formed on machine respectively1, a2…anAnd b1, b2…bn, as illustrated in figure 2 of the appended drawings.5 width are shot in real example
Image, and obtained Digital Image Transmission will be shot to image processor.
Step 3:The wheel footpath image gathered to industrial camera carries out image preprocessing work, and its specific steps includes:
Step 3-1:Medium filtering is carried out to image, sequential scanning is carried out along image by the filter window of medium filtering,
In real example use 5 × 5 filter window so that larger pixel is close to neighboring pixel value with neighboring pixel gray scale difference, with this
Eliminate most of noise.
Step 3-2:Three-dimensional correction is carried out to two images, passes through the plane of delineation of the Bouguet algorithms to two video cameras
Carry out re-projection so that they accurately fall in approximately the same plane, and image row be perfectly aligned over it is preceding to parallel
In structure.
Step 3-3:Stereo matching is carried out to characteristic point, using simplified NCC algorithms, according to binocular wheel to correction chart as
Feature, method of the wheel to Edge Feature Points is searched for using along line direction, completes Stereo matching of the wheel to external diameter characteristic point, match party
Method is as shown in figure 3, for image I of the binocular wheel to correction1And I2, it can use normalizing with the gray scale correlation of a line epigraph point
The correlation function measure of change:
In the matching process, certain threshold value is set first, when calculating some correlations more than threshold value, it is believed that be a time
Match point is selected, then turns around and asks for image I2The upper match point is in I1In candidate matches point.
Step 4: obtain taking turns after the feature point coordinates for matching left and right camera by Stereo matching, it is three-dimensional with reference to video camera
Demarcation, according to Bouguet algorithms, pixel can be mapped in three-dimensional by re-projection, marked further according to binocular camera solid
Fixed result, obtains re-projection matrix
Wherein (cx, cy) be left image principal point, c 'xIt is the x coordinate of right figure principal point, f is left image focal length, TxTaken the photograph for two
Camera projection centre translation vector x-component, above-mentioned numerical value can be obtained by the stereo calibration of binocular camera.
Step 4-1:The parallax that characteristic point is associated can be obtained by Stereo matching, if v1The match point that row is obtained is m1
(u1, v1) and m2(u2, v1), then the parallax between match point is d1=u2-u1。
Step 4-2:The parallax d associated according to characteristic point, passes through coordinate transformProjecting characteristic points are arrived
Among three dimensional coordinate space relative to left video camera, due to the error of installation, left and right camera can not can guarantee that and be substantially parallel,
It is also impossible to guarantee and is exactly perpendicularly to wheel to plane, therefore takes turns flat for space relative to camera coordinate system { C } to place plane
Face, therefore, setting up new fitted coordinate system { L }.
Step 4-3:According to the obtained three-dimensional coordinate (X/W, Y/W, Z/W) under left camera coordinates, by sitting
Mark transformLPi=(CRL)T·Pi+P1, (i=1 ... n) is mapped to fitted coordinate system, realizes two dimensional surface.
Step 5:Plane fitting is carried out to the three-dimensional data of external diameter feature point extraction according to obtained wheel, and obtained
Wheel is to the fit Plane π (0.2680x-0.3176y-z-347.9605=0) relative to camera coordinate system.
Step 5-1:According to obtained fit Plane parameter, fitted coordinate system is further obtained relative to left video camera
The transformation matrix of coordinate system, the transformation matrix data in actual measurement are as follows:
Step 5-2:Parameter fitting is carried out in the two dimensional surface where wheel footpath characteristic point, as shown in figure 4, with P1For origin
OL, P1Point to PnDirection be XL, using the normal direction of π planes as ZL, YL=ZL×XL, thenCRL=By P1, P2…PnPass through formulaLPi=(CRL)T·Pi+P1, (i=1 ... n) can be by { C }
{ L } is transformed into, due toLPiIn π, therefore it can carry out justifying fitting in the plane and obtain final wheel pair radius R.
The above-mentioned binocular vision based on line-structured light is intended for e measurement technology of the steering framing wheel to parameter, can use quick
Efficient NCC algorithms realize the Stereo matching of characteristic point, and binocular image plane is thrown again to taking turns using Bouguet algorithms
Shadow realizes three-dimensional correction.Two dimensional surface is mapped to Edge Feature Points by obtained wheel is extracted by three-dimensional reconstruction and coordinate transform
It is interior and finally fit wheel footpath parameter, flange radius is detected tested by taking turns, the error for demonstrating this method is less than 0.1mm,
With preferable accuracy of detection, it is larger and be difficult to 3 D Quasi that the invention solves current measuring method characteristic point Stereo matching error
The problem of conjunction, for promoting the development tool of Automation Industry to be of great significance.
Claims (11)
1. a kind of wheel footpath measuring method based on binocular vision, its step is as follows:
Step 1: being demarcated to structured light vision sensor, the intrinsic parameter and stereo calibration parameter of video camera are obtained;
Step 2: carrying out IMAQ to wheel footpath using the shooting equipment fixed;
Step 3: the wheel footpath image gathered to industrial camera carries out image preprocessing work,
Step 3-1:Medium filtering is carried out to image;
Step 3-2:Three-dimensional correction is carried out to two images;
Step 3-3:Stereo matching is carried out to characteristic point;
Step 4: the characteristic point image coordinate for asking for obtaining using step 3 carries out three-dimensional reconstruction and two dimensional surface;
Step 5: being fitted and calculating to parameter to taking turns.
2. according to the method described in claim 1, it is characterised in that in step one, by Matlab self-calibrations algorithm and
Stereo calibration algorithm, obtains the intrinsic parameter and stereo calibration parameter of left and right cameras.
3. according to the method described in claim 1, it is characterised in that in step 2, by the homemade equipment that shoots to wheel footpath
It is continuously shot, obtains the image information of wheel footpath.
4. according to the method described in claim 1, it is characterised in that the wheel gathered in step 3 is big to binocular original image
Small is 768 × 576 pixels, and good image processing method can reduce the complexity of feature point extraction, improves and extracts result
Accuracy, its committed step is as follows:Step 3-1:Using 5 × 5 filter window, the window is along image sequential scanning so that with
The larger pixel of neighboring pixel gray scale difference is close to neighboring pixel value, so as to largely eliminate noise;Step 3-2:Pass through
Bouguet algorithms carry out re-projection to the plane of delineation of two video cameras so that they accurately fall at grade, complete
The effect of three-dimensional correction;Step 3-3:On the basis of step 3-2, along line direction search wheel to Edge Feature Points, its feature exists
In utilizing NCC algorithms, characteristic point Stereo matching is completed in the case of high efficiency.
5. according to the method described in claim 1, it is characterised in that in step 4, obtain taking turns to left and right by Stereo matching
After the feature point coordinates of camera matching, three-dimensional coordinate of the characteristic point relative to left camera is obtained using three-dimensional reconstruction, with reference to shooting
Machine stereo calibration, according to Bouguet algorithms, pixel can be mapped in three-dimensional by re-projection, further according to binocular camera
The result of stereo calibration, obtains re-projection matrix.
6. spy according to the method described in claim 1, it is characterised in that in step 4, can also be obtained by Stereo matching
Levy a parallax for association.
7. the parallax according to the method described in claim 1, it is characterised in that in step 4, associated according to characteristic point, passes through
Coordinate transform by projecting characteristic points to the three dimensional coordinate space relative to left video camera among.
8. according to the method described in claim 1, it is characterised in that in step 4, according to having obtained in left video camera
Three-dimensional coordinate under coordinate, is mapped to fitted coordinate system by coordinate transform formula, realizes two dimensional surface.
9. it is according to the method described in claim 1, it is characterised in that in step 5, special to external diameter according to obtained wheel
Levy a three-dimensional data extracted and carry out plane fitting, and obtain taking turns the fit Plane to relative to camera coordinate system.
10. according to the method described in claim 1, it is characterised in that in step 5, joined according to obtained fit Plane
Number, further obtains transformation matrix of the fitted coordinate system relative to left camera coordinate system.
11. according to the method described in claim 1, it is characterised in that in step 5, put down in the two dimension where wheel footpath characteristic point
Parameter fitting is carried out in face, match value is obtained.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710167507.XA CN106996748A (en) | 2017-03-16 | 2017-03-16 | A kind of wheel footpath measuring method based on binocular vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710167507.XA CN106996748A (en) | 2017-03-16 | 2017-03-16 | A kind of wheel footpath measuring method based on binocular vision |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106996748A true CN106996748A (en) | 2017-08-01 |
Family
ID=59431743
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710167507.XA Pending CN106996748A (en) | 2017-03-16 | 2017-03-16 | A kind of wheel footpath measuring method based on binocular vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106996748A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107677223A (en) * | 2017-10-25 | 2018-02-09 | 烟台大学 | The wheel shooting measurement apparatus and measuring method of a kind of non-contact four-wheel position finder |
CN107967685A (en) * | 2017-12-11 | 2018-04-27 | 中交第二公路勘察设计研究院有限公司 | A kind of bridge pier and tower crack harmless quantitative detection method based on unmanned aerial vehicle remote sensing |
CN108267104A (en) * | 2018-01-22 | 2018-07-10 | 浙江大学 | A kind of axial workpiece radius size measuring method based on binocular vision |
CN108917595A (en) * | 2018-06-19 | 2018-11-30 | 杭州蓝蜓科技有限公司 | Glass on-line measuring device based on machine vision |
CN109470170A (en) * | 2018-12-25 | 2019-03-15 | 山东大学 | Stereoscopic vision space circle pose high-precision measuring method and system based on optimal projection plane |
CN109598687A (en) * | 2018-12-04 | 2019-04-09 | 深慧视(深圳)科技有限公司 | Binocular Stereo Vision System and method for correcting image |
CN109751973A (en) * | 2017-11-01 | 2019-05-14 | 欧姆龙株式会社 | Three-dimensional measuring apparatus, method for three-dimensional measurement and storage medium |
CN110567345A (en) * | 2019-09-04 | 2019-12-13 | 北京信息科技大学 | Non-contact type pipe wall thickness measuring method and system based on machine vision |
CN112284696A (en) * | 2019-07-11 | 2021-01-29 | 中国科学院半导体研究所 | Gas cylinder volume detection method and system based on 3D vision technology |
CN117315033A (en) * | 2023-11-29 | 2023-12-29 | 上海仙工智能科技有限公司 | Neural network-based identification positioning method and system and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2006118901A (en) * | 2004-10-20 | 2006-05-11 | Hitachi Industries Co Ltd | Wheel profile measuring device |
CN100999220A (en) * | 2006-12-18 | 2007-07-18 | 杭州电子科技大学 | On-line detection method and device for vehicle wheel set diameter |
JP2008180619A (en) * | 2007-01-25 | 2008-08-07 | Act Denshi Kk | Wheel measuring method and wheel measuring apparatus therefor |
CN103693073A (en) * | 2014-01-06 | 2014-04-02 | 北京交通大学 | Contactless vehicle wheel diameter dynamical measuring device and method |
CN105528785A (en) * | 2015-12-03 | 2016-04-27 | 河北工业大学 | Binocular visual image stereo matching method |
CN105894499A (en) * | 2016-03-25 | 2016-08-24 | 华南理工大学 | Binocular-vision-based rapid detection method for three-dimensional information of space object |
-
2017
- 2017-03-16 CN CN201710167507.XA patent/CN106996748A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2006118901A (en) * | 2004-10-20 | 2006-05-11 | Hitachi Industries Co Ltd | Wheel profile measuring device |
CN100999220A (en) * | 2006-12-18 | 2007-07-18 | 杭州电子科技大学 | On-line detection method and device for vehicle wheel set diameter |
JP2008180619A (en) * | 2007-01-25 | 2008-08-07 | Act Denshi Kk | Wheel measuring method and wheel measuring apparatus therefor |
CN103693073A (en) * | 2014-01-06 | 2014-04-02 | 北京交通大学 | Contactless vehicle wheel diameter dynamical measuring device and method |
CN105528785A (en) * | 2015-12-03 | 2016-04-27 | 河北工业大学 | Binocular visual image stereo matching method |
CN105894499A (en) * | 2016-03-25 | 2016-08-24 | 华南理工大学 | Binocular-vision-based rapid detection method for three-dimensional information of space object |
Non-Patent Citations (3)
Title |
---|
何东健等: "《数字图像处理(第三版)》", 28 February 2015, 西安电子科技大学出版社 * |
宋西平等: "基于机器视觉的葡萄采摘点三维空间定位系统的研究", 《江苏农业科学》 * |
张亮: "基于双目视觉的列车轮对检测系统研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107677223A (en) * | 2017-10-25 | 2018-02-09 | 烟台大学 | The wheel shooting measurement apparatus and measuring method of a kind of non-contact four-wheel position finder |
CN109751973B (en) * | 2017-11-01 | 2020-12-11 | 欧姆龙株式会社 | Three-dimensional measuring device, three-dimensional measuring method, and storage medium |
CN109751973A (en) * | 2017-11-01 | 2019-05-14 | 欧姆龙株式会社 | Three-dimensional measuring apparatus, method for three-dimensional measurement and storage medium |
CN107967685A (en) * | 2017-12-11 | 2018-04-27 | 中交第二公路勘察设计研究院有限公司 | A kind of bridge pier and tower crack harmless quantitative detection method based on unmanned aerial vehicle remote sensing |
CN108267104A (en) * | 2018-01-22 | 2018-07-10 | 浙江大学 | A kind of axial workpiece radius size measuring method based on binocular vision |
CN108917595A (en) * | 2018-06-19 | 2018-11-30 | 杭州蓝蜓科技有限公司 | Glass on-line measuring device based on machine vision |
CN109598687A (en) * | 2018-12-04 | 2019-04-09 | 深慧视(深圳)科技有限公司 | Binocular Stereo Vision System and method for correcting image |
CN109470170A (en) * | 2018-12-25 | 2019-03-15 | 山东大学 | Stereoscopic vision space circle pose high-precision measuring method and system based on optimal projection plane |
CN109470170B (en) * | 2018-12-25 | 2020-01-07 | 山东大学 | Stereoscopic vision space circular attitude high-precision measurement method and system based on optimal projection plane |
CN112284696A (en) * | 2019-07-11 | 2021-01-29 | 中国科学院半导体研究所 | Gas cylinder volume detection method and system based on 3D vision technology |
CN110567345A (en) * | 2019-09-04 | 2019-12-13 | 北京信息科技大学 | Non-contact type pipe wall thickness measuring method and system based on machine vision |
CN117315033A (en) * | 2023-11-29 | 2023-12-29 | 上海仙工智能科技有限公司 | Neural network-based identification positioning method and system and storage medium |
CN117315033B (en) * | 2023-11-29 | 2024-03-19 | 上海仙工智能科技有限公司 | Neural network-based identification positioning method and system and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106996748A (en) | A kind of wheel footpath measuring method based on binocular vision | |
CN110033447B (en) | High-speed rail heavy rail surface defect detection method based on point cloud method | |
CN104484648B (en) | Robot variable visual angle obstacle detection method based on outline identification | |
CN108510530B (en) | Three-dimensional point cloud matching method and system | |
CN104574393B (en) | A kind of three-dimensional pavement crack pattern picture generates system and method | |
CN105674880B (en) | Contact net geometric parameter measurement method and system based on binocular principle | |
CN103559711B (en) | Based on the method for estimating of three dimensional vision system characteristics of image and three-dimensional information | |
CN104930985B (en) | Binocular vision 3 D topography measurement method based on space-time restriction | |
CN106340009B (en) | A kind of electric power line detecting method and system based on parallel binocular | |
CN109615654B (en) | Method for measuring corrosion depth and area of inner surface of drainage pipeline based on binocular vision | |
CN107767456A (en) | A kind of object dimensional method for reconstructing based on RGB D cameras | |
CN106679567A (en) | Contact net and strut geometric parameter detecting measuring system based on binocular stereoscopic vision | |
CN106643555B (en) | Connector recognition methods based on structured light three-dimensional measurement system | |
CN110223355B (en) | Feature mark point matching method based on dual epipolar constraint | |
CN111091076B (en) | Tunnel limit data measuring method based on stereoscopic vision | |
CN111996883B (en) | Method for detecting width of road surface | |
CN115482195B (en) | Train part deformation detection method based on three-dimensional point cloud | |
CN111768417B (en) | Railway wagon overrun detection method based on monocular vision 3D reconstruction technology | |
CN109974618A (en) | The overall calibration method of multisensor vision measurement system | |
CN106500625A (en) | A kind of telecentricity stereo vision measuring apparatus and its method for being applied to the measurement of object dimensional pattern micron accuracies | |
CN115330684A (en) | Underwater structure apparent defect detection method based on binocular vision and line structured light | |
Kim et al. | The effective acquisition and processing of 3D photogrammetric data from digital photogrammetry for construction progress measurement | |
CN110349209A (en) | Vibrating spear localization method based on binocular vision | |
CN104614372B (en) | Detection method of solar silicon wafer | |
CN108180825A (en) | A kind of identification of cuboid object dimensional and localization method based on line-structured light |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170801 |