CN105741278A - On-line monitoring method of inhaul cable distribution stress on the basis of computer vision - Google Patents
On-line monitoring method of inhaul cable distribution stress on the basis of computer vision Download PDFInfo
- Publication number
- CN105741278A CN105741278A CN201610054665.XA CN201610054665A CN105741278A CN 105741278 A CN105741278 A CN 105741278A CN 201610054665 A CN201610054665 A CN 201610054665A CN 105741278 A CN105741278 A CN 105741278A
- Authority
- CN
- China
- Prior art keywords
- line
- drag
- stress
- district
- dimensional
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention discloses an on-line monitoring method of inhaul cable distribution stress on the basis of computer vision. The on-line monitoring method comprises the following specific implementation steps: A: erecting two high-speed cameras, and debugging camera shooting parameters; B: carrying out the three-dimensional construction of a testing zone, system calibration and the initial setting of a system; C: carrying out three-dimensional target characteristic point extraction and three-dimensional space model deformation calculation; and D: carrying out automatic on-line monitoring and real-time storage of inhaul cable testing zone stress.
Description
Technical field
The present invention relates to use vision sensor and the stress state of various load action downhauls is carried out on-line monitoring by digital image processing method.
Background technology
Steel cable is the important load-carrying member of Loads of Long-span Bridges (such as cable-stayed bridge, suspension bridge and arch bridge etc.) structure, and its stress directly influences the suitability and the safety of whole bridge structure.Stayed structure is in its stretching construction and bridge operation process, owing to material property, Construction Condition and environmental condition factor etc. affect, stress loss or stress can be occurred excessive, thus causing that drag-line stress distribution is beyond predetermined safe range, the safety of whole bridge structure is on the hazard.Easily being caused security incident by the abnormal bridge structure damage caused of drag-line stress state, this is unallowed at bridge construction and operation stage.The Suo Li of steel cable is monitored the vital task having become as in monitoring structural health conditions field.But, up to the present the Suo Li of steel cable accurately monitors or an extremely challenging task.
Stress state for drag-line is monitored, and traditional measuring method mainly includes contact method for measuring stress and non-contact stress measuring method two kinds, carries out relevant commentary below for its classification.
1, contact drag-line method for measuring stress:
(1) based on the Cable force monitoring methods of strain-type pressure rings.Strain-type pressure rings is to utilize structure and the mechanics of materials relative theory deformation that pressure rings produces to be changed to generate the sensor of pulling force suffered by its pressure experienced and then assessment drag-line.Needing in the middle of practical application to be arranged on pressure rings between ground tackle and the anchor plate of guy anchor fixed end, drag-line can drive ground tackle in tension process so that pressure rings is clamped between ground tackle and anchor plate.Can be shown that the pressure suffered by pressure rings is exactly the drag-line stress in end by the balance of power and transitive relation.This drag-line stress monitoring method is just for the stress state of stay cable end, and for super long stayed cable, due to drag-line under gravity because its stress distribution of cable sag effect is uneven, in this way the integrated stress state of drag-line can not be carried out comprehensive assessment.Owing to pressure rings is transmitted generally by the form of voltage signal and stores, its transmission range is restricted.And strain gauge transducer is easily subject to noise jamming, cause that cable force monitoring produces bigger error.Further, since being continually changing of Cable power can cause that the pressure rings under pressured state is susceptible to destroy and lost efficacy, the life-span of pressure rings itself can be greatly affected, and pressure rings is replaced fairly cumbersome.
(2) based on the Cable force monitoring methods of frequency of vibration.This method is a kind of method being most widely used in the middle of the detection of current bridge cable.By installing acceleration transducer on drag-line, to drag-line vibrational state measurement under artificial excitation or environmental excitation, adopt frequency spectrum analysis method can obtain the frequency of vibration of drag-line.The Suo Li of drag-line just can be identified by the relation utilizing Cable power and its frequency of vibration.But this relation is very sensitive for the boundary condition of drag-line and himself rigidity, it is easily caused cable force measurement misalignment to the simplification of drag-line boundary condition and himself rigidity is inaccurate.Further, since spectrum analysis needs the Acceleration time course signal taking a period of time to carry out frequency-domain analysis, so Cable power can not be carried out real time on-line monitoring.
(3) based on the Cable force monitoring methods of fiber Bragg grating strain sensor.This method needs to be pasted onto by fiber-optic grating sensor in the middle of drag-line Steel Wire Surface or steel wire internal gutter.Can deform after drag-line under tension, due to fiber grating and steel wire close contact, it is possible to the strain regime of drag-line is estimated by the strain produced by fiber grating.The stress state of drag-line can be calculated by stress-strain relation in the mechanics of materials.This drag-line stress monitoring method requires higher for technological level, and optical fiber needs carry out special protection and need close contact between optical fiber and steel wire.The tight degree direct influence that bonds between optical fiber and steel wire is to the strain transfer between drag-line and optical fiber, thus affecting the stress monitoring of drag-line.
2, contactless drag-line method for measuring stress:
(1) based on the drag-line stress monitoring method of magnetic flux transducer.This method is to utilize the counter magnetostriction effect of ferromagnetic material, carrys out the stress state of anti-push-pull cable by monitoring drag-line magnetic field parameter change under stress effect.Magnetic flux transducer is in the process carrying out cable force monitoring, and sensor is easily subject to the interference of electromagnetic environment, causes measuring the bigger noise of appearance and error.Additionally sensor is subject to variations in temperature impact, the current technique for temperature compensation error that well solution does not bring due to variations in temperature.
(2) based on the drag-line stress monitoring method of supersonic guide-wave.This method is to launch supersonic guide-wave in drag-line one end, and guided wave is collected by the other end, can the stress of drag-line be identified by guided wave mutation analysis.But current this monitoring method is in conceptual phase, its feature being subject to electromagnetic interference and the complex nonlinear problem between guided wave and stress, material geometry etc. directly affects the popularization and application of the method.
Sum up above-mentioned several drag-line stress monitoring method, traditional drag-line stress monitoring method or anchoring end cable force situation can only be monitored and drag-line can not realize distributed measurement (strain-type pressure rings), or on-line monitoring (frequency method) can not be realized by boundary condition, or processing technique is required very high (fiber-optic grating sensor), or to temperature effects very sensitive (magnetic flux transducer), or because the non-linear conversion relation extremely complex (supersonic guide-wave) of physical quantity, or because major part monitoring method is susceptible to electromagnetic interferences, these all can bring very big error to measuring.Additionally conventional monitoring methods needs bridge to carry out the operations such as wiring, measures process relative complex, and sensor is not easily changed.In order to ensure the suitability and the safety of bridge, simultaneously take account of the quick and convenient of drag-line stress monitoring, find a kind of drag-line stress monitoring method accurate, reliable, convenient Longspan Bridge most important.
In recent years, by means of the continuous progress of computer hardware technology, digital image processing techniques and signal processing technology, computer vision technique achieves swift and violent development.Computer vision technique also because it is advanced, accuracy and convenience etc. receive the favor of civil engineering technicians gradually, be constantly applied in the middle of bridge, the dynamic displacement monitoring of building structure and outward appearance detection.Computer vision methods is used to be monitored drag-line stress being a good selection.
Summary of the invention
The present invention to overcome the deficiency of tradition drag-line stress monitoring method, it is proposed to a kind of contactless drag-line distributed stress on-line monitoring method based on computer vision technique.This measuring method adopts the eyes of two camera simulation people, and drag-line picture on surface is carried out three-dimensional perception, by contrasting drag-line visual pattern change under stress and initial unstress state, it is thus achieved that the stress state of drag-line.Measurement apparatus includes high-speed camera head and computer.
The problem that the invention solves the problems that the following aspects:
One is solve tradition drag-line stress monitoring method because the loaded down with trivial details problem of operation and the measurement noise brought due to line transmission and error problem are measured in upper bridge wiring etc..Here camera head is arranged in the position away from bridge floor, realizes the remote contactless monitoring to drag-line stress state by adjusting camera lens.
Two is solve tradition based on the problem that can not realize continuous on-line monitoring in the drag-line stress monitoring of frequency method.
Three is solve to measure in tradition drag-line stress monitoring method to be subject to electromagnetic interference, easily affected by temperature effects and measurement error problem that complex nonlinear problem in measurement process causes.
Four is simplify sensor processing and mounting process, improves the ease for use of drag-line stress monitoring method, makes cable tension sensor have easy replaceability.
A kind of drag-line distributed stress on-line monitoring method based on computer vision of the present invention, is embodied as step as follows:
1., based on a drag-line distributed stress on-line monitoring method for computer vision, it is embodied as step as follows:
A. set up two high-speed camera heads and carry out camera parameter debugging；
A1. two high-speed camera headstocks are located at away under the bridge of drag-line, the pattern test zone that alignment lens drag-line pre-sets；
A2. repeatedly adjust the space angle of cut of two high-speed camera heads, regulate the lens focus of photographic head, aperture size and amplification etc. so that the tested region of drag-line simultaneously appears in the visual field of two photographic head；
A3. adjust two photographic head time of exposure and yield value, obtain the picture rich in detail in the tested region of drag-line and make it be positioned at the correct position of image, finally give drag-line and survey the optimized image in district；
B. survey district's three-dimensional structure, system calibrating and system initialization to set；
B1. the locus in district is surveyed towards angle, space length and drag-line according to the space of two high-speed camera heads, it is determined that the space geometry relation of three；
B2. shoot by the image of geodesic structure with two high-speed camera heads respectively, drag-line is surveyed district and carries out three-dimensional structure, set up image coordinate and by the space coordinates mapping relations of geodesic structure；
B3. biocular systems is demarcated, it is determined that practical structures space displacement pixel displacement on two images, find out calibration coefficient matrix；
C. objective feature point extraction and three-dimensional space model deformation calculation；
C1. transfer drag-line and survey initialization pattern 3-dimensional image model model corresponding to real space (threedimensional model under unstress state) in district；
C2. extracting multiple characteristic points in initialization pattern 3-dimensional image model, these characteristic points can fully characterize drag-line and survey the three-dimensional geometry form in district, and using these characteristic points as the relevant benchmark of three-dimensional digital image；
C3., reference templates carries out digital picture relevant matches in the threedimensional model of the picture construction photographed by two high-speed camera heads, and the characteristic point that search is extracted is in existing position in a model, and realizes multi-characteristic points tracking；
C4. multi-characteristic points tracing process is carried out machine learning and training, optimizes tracking task, and add the change interference to following the trail of such as adaptive algorithm reduction light, until tracing characteristic points meets requirement；
C5. multi-characteristic points has been followed the trail of, and 3-dimensional image model is transformed into actual three-dimensional space model, is contrasted with initializing three-dimensional space model by actual three-dimensional space model now, obtains drag-line and surveys the actual three-dimensional deformation in district；
D. drag-line surveys the monitoring of zone position stress automatic on-line and real-time storage；
D1. according to strain and stress relation in the mechanics of materials, the drag-line obtained in step C is utilized to survey the three-dimensional actual deformation relationship in district, can obtaining surveying the stress field in district, stress field can survey the stress distribution in district as drag-line, takes the average of stress field as surveying district's drag-line stress value；
D2. require to formulate data sampling frequency and storage strategy according to the stress monitoring of tested drag-line；
D3. two photographic head are constantly taken pictures, described in step C, the model of the picture construction that each frame photographs is carried out multi-characteristic points tracking, tracking task completes be transformed into actual threedimensional model afterwards and calculate deformation, and then obtains drag-line survey district's stress field and average stress；
D4. checking whether D3 completes the D2 acquisition strategies proposed and gather store tasks, if completed, drag-line stress monitoring task completes.
The imagery exploitation gigabit Ethernet collected is transmitted by the high-speed camera head being previously mentioned in above-mentioned steps, is saved in hard disc of computer and immediately processes.
Shoot carrying out needing in the middle of image three-dimensional structure from different angles drag-line being surveyed district with two photographic head, two images obtained are carried out threedimensional model structure, it is achieved three-dimensional stereoscopic visual function.After the space angle of cut of photographic head and space length are determined, the threedimensional model that its picture construction photographed obtains and real space model are one to one.The transition matrix of image and real space can be set up by the object of physical dimension known in the middle of real space at the Pixel Dimensions of threedimensional model, and be used as system calibrating.
Threedimensional model tracing characteristic points process needs the threedimensional model of the picture construction photographed is tracked training and study, optimize tracing process.The drag-line survey district mentioned in flow chart refers to and makes the pattern of certain texture in advance as surveying district in survey zone position under drag-line unstress state, and with camera head, drag-line survey district is scanned, set up drag-line and survey the initialization pattern 3-dimensional image model in district, and by measuring its actual space geometry size, 3-dimensional image model and the foundation of real space model are contacted, obtain the reference value under drag-line unstress state.Follow-up drag-line is in installation, stretch-draw and bridge operation process, it is possible to and then obtain working as fore stay survey district stress state relative to the deformation of unstress state spatial model (reference value) by contrasting its current spatial model.The flow chart that the present invention illustrates only is lifted a drag-line and has surveyed district, actually by adjusting photographic head parameter, multiple survey districts of many drag-lines have been included within the scope of camera view, thus the stress distribution formula realizing many drag-lines is measured.
Except the high-speed camera head being previously mentioned and computer etc. in the present invention, additionally provide a set of contactless drag-line distributed stress on-line monitoring system software platform being stored in the middle of computer based on computer vision technique.
This advantage is:
1, solve tradition drag-line stress monitoring method to need bridge, drag-line is installed sensor and carries out the loaded down with trivial details operations such as wiring, it is achieved the remote non-contact measurement of drag-line stress；
2, solve the problem of rope force signal limited transmission distance in tradition drag-line stress monitoring method, avoid the signal noise owing to long range propagation brings simultaneously, improve certainty of measurement；
3, this method is by adding adaptive algorithm, it is possible to significantly alleviate the tracking error owing to environmental factors (light change and fog edge block) etc. causes；
4, high-speed camera head has higher sample frequency, achieve the long-term real time on-line monitoring of drag-line stress state, and the power of drag-line can be changed and be further analyzed, the real-time online warning function of drag-line can be realized in conjunction with the dynamic stress change information of drag-line and faulted condition is automatically monitored and identified；
5, solve tradition drag-line stress monitoring and certain point of drag-line can only be carried out the drawback of stress monitoring, it is achieved that full-bridge drag-line distributed synchronization is measured, and can obtain the information of force field in certain region of drag-line；
6, solve tradition drag-line stress monitoring method and be subject to electromagnetic interference, the weakness that variations in temperature is sensitive, it is to avoid the nonlinear problem occurred in measurement process, it is ensured that the stability of measurement and reliability；
7, due to adopt high speed gigabit Ethernet by camera collection to image information be transferred to computer, real-time Transmission is fast, it is achieved online data processes immediately, it is possible to real-time to monitor video and drag-line stress state is combined and checks, it is possible to achieve event perfect information playback；
8, the present invention supports the collaborative work of more multi-cam, by arranging that multiple photographic head can realize the stress state on-line monitoring of all drag-lines within the scope of full-bridge at diverse location；
9, simplify sensor processing technology thereof, improve the ease for use of drag-line stress monitoring method, solve the tradition not easily changeable problem of drag-line stress monitoring sensor.
10, compared to other measuring methods, measure that same pointing precision is high, cost is low, simple to operate, be capable of portable drag-line stress monitoring.
Accompanying drawing explanation
Fig. 1 assembly of the invention schematic diagram.
The measurement procedure figure of Fig. 2 present invention.
Marginal data: the code name in Fig. 1 represents respectively:
1 Longspan Bridge,
2 drag-lines,
3 drag-lines survey district,
4 first high-speed camera heads,
5 second high-speed camera heads,
6 computers,
7 first target images that high-speed camera head photographs,
8 second target images that high-speed camera head photographs,
Remarks: choose on Longspan Bridge a certain drag-line Shang Mouduance district in the present invention as monitoring object, first high-speed camera head and second high-speed camera head shoot drag-line survey district simultaneously and obtain two images, the 3-dimensional image model of shooting area is formed by three-dimensional constructing technology, survey the texture in district as the feature identified and follow the trail of, in order to be highlighted, drag-line surface be marked with band gray scale herein.Only having lifted a drag-line survey district herein is example, can essentially be entirely included within sweep of the eye by multiple survey districts of many drag-lines of full-bridge by adjustment photographic head parameter, it is achieved many ropes distributed monitoring.
Detailed description of the invention
The present invention is expanded on further below in conjunction with the case shown in Fig. 1 and the workflow shown in Fig. 2.
Referring to Fig. 1 and Fig. 2, in the present invention, lifted case is utilize a kind of contactless drag-line distributed stress on-line monitoring system based on computer vision technique that the drag-line on one Longspan Bridge (here for arch bridge) is carried out stress state monitoring, and concrete enforcement step is as follows:
A. set up two high-speed camera heads 4,5 and carry out camera parameter debugging；
A1. two high-speed camera heads 4,5 are erected under the bridge away from drag-line 2, the pattern test zone 3 that alignment lens drag-line pre-sets；
A2. repeatedly adjust the space angle of cut of two high-speed camera heads 4,5, regulate the lens focus of photographic head, aperture size and amplification etc. so that the tested region 3 of drag-line simultaneously appears in the visual field of two photographic head；
A3. adjust two high-speed camera head time of exposure and yield value, obtain the picture rich in detail in the tested region 3 of drag-line and be located at the correct position of image, finally give drag-line and survey the optimized image in district 3；
B. survey district's three-dimensional structure, system calibrating and system initialization to set；
B1. the locus in district 3 is surveyed towards angle, space length and drag-line according to the space of two high-speed camera heads 4,5, it is determined that the space geometry relation of three；
B2. shoot by the image of geodesic structure with two high-speed camera heads 4,5 respectively, drag-line is surveyed district 3 and carries out three-dimensional structure, set up image coordinate and by the space coordinates mapping relations of geodesic structure；
B3. biocular systems is demarcated, it is determined that practical structures space displacement pixel displacement on two images 7,8, find out calibration coefficient matrix；
C. objective feature point extraction and three-dimensional space model deformation calculation；
C1. transfer drag-line and survey initialization pattern 3-dimensional image model model corresponding to real space (threedimensional model under unstress state) in district；
C2. extracting multiple characteristic points in initialization pattern 3-dimensional image model, these characteristic points can fully characterize drag-line and survey the three-dimensional geometry form in district, and using these characteristic points as the relevant benchmark of three-dimensional digital image；
C3., reference templates carries out digital picture relevant matches in the threedimensional model of the picture construction photographed by two high-speed camera heads, and the characteristic point that search is extracted is in existing position in a model, and realizes multi-characteristic points tracking；
C4. multi-characteristic points tracing process is carried out machine learning and training, optimizes tracking task, and add the change interference to following the trail of such as adaptive algorithm reduction light, until tracing characteristic points meets requirement；
C5. multi-characteristic points has been followed the trail of, and 3-dimensional image model is transformed into actual three-dimensional space model, is contrasted with initializing three-dimensional space model by actual three-dimensional space model now, obtains drag-line and surveys the actual three-dimensional deformation in district；
D. drag-line surveys the stress automatic on-line monitoring of position, district 3 and real-time storage；
D1. according to strain and stress relation in the mechanics of materials, the drag-line obtained in step C is utilized to survey the three-dimensional actual deformation relationship in district 3, can obtaining surveying the stress field in district, stress field can survey the stress distribution in district as drag-line, takes the average of stress field as surveying district's drag-line stress value；
D2. require to formulate data sampling frequency and storage strategy according to the stress monitoring of tested drag-line；
D3. two photographic head 4,5 are constantly taken pictures, described in step C, the model of the picture construction that each frame photographs is carried out multi-characteristic points tracking, tracking task completes be transformed into actual threedimensional model afterwards and calculate deformation, and then obtains drag-line survey district 3 stress field and average stress；
D4. checking whether D3 completes the D2 acquisition strategies proposed and gather store tasks, if completed, drag-line stress monitoring task completes.
The imagery exploitation gigabit Ethernet collected is transmitted by the high-speed camera head being previously mentioned in above-mentioned steps, is saved in the middle of hard disc of computer and immediately processes.
In carrying out image three-dimensional structure, need from different angles drag-line to be surveyed district with two photographic head shoot, two images obtained are carried out threedimensional model structure, it is achieved three-dimensional stereoscopic visual function.After the space angle of cut of photographic head and space length are determined, the threedimensional model that its picture construction photographed obtains and real space model are one to one.The transition matrix of image and real space can be set up by the object of physical dimension known in the middle of real space at the Pixel Dimensions of threedimensional model, and be used as system calibrating.
Threedimensional model tracing characteristic points process needs the threedimensional model of the picture construction photographed is tracked training and study, optimize tracing process.The drag-line survey district mentioned in flow chart refers to and makes the pattern of certain texture in advance as surveying district in survey zone position under drag-line unstress state, and with camera head, drag-line survey district is scanned, set up drag-line and survey the initialization pattern 3-dimensional image model in district, and by measuring its actual space geometry size, 3-dimensional image model and the foundation of real space model are contacted, obtain the reference value under drag-line unstress state.Follow-up drag-line is in installation, stretch-draw and bridge operation process, it is possible to and then obtain working as fore stay survey district stress state relative to the deformation of unstress state spatial model (reference value) by contrasting its current spatial model.The flow chart that the present invention illustrates only is lifted a drag-line and has surveyed district, actually by adjusting photographic head parameter, multiple survey districts of many drag-lines have been included within the scope of camera view, thus the stress distribution formula realizing many drag-lines is measured.
Except the high-speed camera head being previously mentioned and computer etc. in the present invention, additionally provide a set of contactless drag-line distributed stress on-line monitoring system software platform being stored in the middle of computer based on computer vision technique.
Content described in this specification case study on implementation is only enumerating of the way of realization to inventive concept; protection scope of the present invention is not construed as being only limitted to the concrete form that case study on implementation is stated, protection scope of the present invention also and in those skilled in the art according to present inventive concept it is conceivable that equivalent technologies means.
Claims (1)
1., based on a drag-line distributed stress on-line monitoring method for computer vision, it is embodied as step as follows:
A. set up two high-speed camera heads and carry out camera parameter debugging；
A1. two high-speed camera headstocks are located at away under the bridge of drag-line, the pattern test zone that alignment lens drag-line pre-sets；
A2. repeatedly adjust the space angle of cut of two high-speed camera heads, regulate the lens focus of photographic head, aperture size and amplification etc. so that the tested region of drag-line simultaneously appears in the visual field of two photographic head；
A3. adjust two photographic head time of exposure and yield value, obtain the picture rich in detail in the tested region of drag-line and make it be positioned at the correct position of image, finally give drag-line and survey the optimized image in district；
B. survey district's three-dimensional structure, system calibrating and system initialization to set；
B1. the locus in district is surveyed towards angle, space length and drag-line according to the space of two high-speed camera heads, it is determined that the space geometry relation of three；
B2. shoot by the image of geodesic structure with two high-speed camera heads respectively, drag-line is surveyed district and carries out three-dimensional structure, set up image coordinate and by the space coordinates mapping relations of geodesic structure；
B3. biocular systems is demarcated, it is determined that practical structures space displacement pixel displacement on two images, find out calibration coefficient matrix；
C. objective feature point extraction and three-dimensional space model deformation calculation；
C1. transfer drag-line and survey initialization pattern 3-dimensional image model model corresponding to real space (threedimensional model under unstress state) in district；
C2. extracting multiple characteristic points in initialization pattern 3-dimensional image model, these characteristic points can fully characterize drag-line and survey the three-dimensional geometry form in district, and using these characteristic points as the relevant benchmark of three-dimensional digital image；
C3., reference templates carries out digital picture relevant matches in the threedimensional model of the picture construction photographed by two high-speed camera heads, and the characteristic point that search is extracted is in existing position in a model, and realizes multi-characteristic points tracking；
C4. multi-characteristic points tracing process is carried out machine learning and training, optimizes tracking task, and add the change interference to following the trail of such as adaptive algorithm reduction light, until tracing characteristic points meets requirement；
C5. multi-characteristic points has been followed the trail of, and 3-dimensional image model is transformed into actual three-dimensional space model, is contrasted with initializing three-dimensional space model by actual three-dimensional space model now, obtains drag-line and surveys the actual three-dimensional deformation in district；
D. drag-line surveys the monitoring of zone position stress automatic on-line and real-time storage；
D1. according to strain and stress relation in the mechanics of materials, the drag-line obtained in step C is utilized to survey the three-dimensional actual deformation relationship in district, can obtaining surveying the stress field in district, stress field can survey the stress distribution in district as drag-line, takes the average of stress field as surveying district's drag-line stress value；
D2. require to formulate data sampling frequency and storage strategy according to the stress monitoring of tested drag-line；
D3. two photographic head are constantly taken pictures, described in step C, the model of the picture construction that each frame photographs is carried out multi-characteristic points tracking, tracking task completes be transformed into actual threedimensional model afterwards and calculate deformation, and then obtains drag-line survey district's stress field and average stress；
D4. checking whether D3 completes the D2 acquisition strategies proposed and gather store tasks, if completed, drag-line stress monitoring task completes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610054665.XA CN105741278B (en) | 2016-01-26 | 2016-01-26 | A kind of drag-line distributed stress on-line monitoring method based on computer vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610054665.XA CN105741278B (en) | 2016-01-26 | 2016-01-26 | A kind of drag-line distributed stress on-line monitoring method based on computer vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105741278A true CN105741278A (en) | 2016-07-06 |
| CN105741278B CN105741278B (en) | 2018-09-14 |
Family
ID=56247751
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610054665.XA Active CN105741278B (en) | 2016-01-26 | 2016-01-26 | A kind of drag-line distributed stress on-line monitoring method based on computer vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105741278B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106530837A (en) * | 2016-11-14 | 2017-03-22 | 浙江大学 | Active ship collision preventing method for bridge based on machine vision |
| CN107146223A (en) * | 2017-04-27 | 2017-09-08 | 武汉大学 | A kind of analysis system and method for power transmission tower and power transmission line displacement coupling |
| CN109341903A (en) * | 2018-11-08 | 2019-02-15 | 东南大学 | A kind of Cable force measuring method based on limb recognition in computer vision |
| CN109871795A (en) * | 2019-01-31 | 2019-06-11 | 湖北交投智能检测股份有限公司 | A kind of highway condition state method, system and the product of automation |
| CN110220909A (en) * | 2019-04-28 | 2019-09-10 | 浙江大学 | A kind of Shield-bored tunnels Defect inspection method based on deep learning |
| CN111649857A (en) * | 2020-04-23 | 2020-09-11 | 河海大学 | Inhaul cable modal measurement method for target matching analysis |
| CN111691679A (en) * | 2020-06-24 | 2020-09-22 | 北京工业大学 | Digital twinning-based intelligent tensioning method for prestressed steel structure |
| CN112378507A (en) * | 2020-11-05 | 2021-02-19 | 浙江大学 | Computer vision structure vibration monitoring method based on motion compensation |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001153740A (en) * | 1999-11-26 | 2001-06-08 | Tokyo Seiko Co Ltd | Tension measurement method for wire rope |
| CN101055216A (en) * | 2007-05-28 | 2007-10-17 | 重庆交通大学 | Cable-stayed bridge cable force automatic monitoring method and device |
| CN101762346A (en) * | 2009-12-31 | 2010-06-30 | 北京市建筑工程研究院 | Method for measuring rope force of multi-span steel stay rope by using multi-frequency method |
| CN102542124A (en) * | 2012-02-29 | 2012-07-04 | 深圳大学建筑设计研究院 | Pull cable prestressing simulating analysis method for steel building structure |
| CN104111133A (en) * | 2014-06-10 | 2014-10-22 | 浙江省建筑科学设计研究院有限公司 | Cable force monitoring system and monitoring method of wide-span pre-stress stay cable steel structure |
-
2016
- 2016-01-26 CN CN201610054665.XA patent/CN105741278B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001153740A (en) * | 1999-11-26 | 2001-06-08 | Tokyo Seiko Co Ltd | Tension measurement method for wire rope |
| CN101055216A (en) * | 2007-05-28 | 2007-10-17 | 重庆交通大学 | Cable-stayed bridge cable force automatic monitoring method and device |
| CN101762346A (en) * | 2009-12-31 | 2010-06-30 | 北京市建筑工程研究院 | Method for measuring rope force of multi-span steel stay rope by using multi-frequency method |
| CN102542124A (en) * | 2012-02-29 | 2012-07-04 | 深圳大学建筑设计研究院 | Pull cable prestressing simulating analysis method for steel building structure |
| CN104111133A (en) * | 2014-06-10 | 2014-10-22 | 浙江省建筑科学设计研究院有限公司 | Cable force monitoring system and monitoring method of wide-span pre-stress stay cable steel structure |
Non-Patent Citations (1)
| Title |
|---|
| 严少波,裴丙志: "斜拉桥索塔拉索锚固区空间应力分析模型", 《国外公路》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106530837A (en) * | 2016-11-14 | 2017-03-22 | 浙江大学 | Active ship collision preventing method for bridge based on machine vision |
| CN106530837B (en) * | 2016-11-14 | 2019-02-05 | 浙江大学 | Actively anti-ship hits method for a kind of bridge based on machine vision |
| CN107146223A (en) * | 2017-04-27 | 2017-09-08 | 武汉大学 | A kind of analysis system and method for power transmission tower and power transmission line displacement coupling |
| CN107146223B (en) * | 2017-04-27 | 2019-06-11 | 武汉大学 | The analysis system and method for a kind of power transmission tower and power transmission line displacement coupling |
| CN109341903A (en) * | 2018-11-08 | 2019-02-15 | 东南大学 | A kind of Cable force measuring method based on limb recognition in computer vision |
| CN109871795A (en) * | 2019-01-31 | 2019-06-11 | 湖北交投智能检测股份有限公司 | A kind of highway condition state method, system and the product of automation |
| CN110220909A (en) * | 2019-04-28 | 2019-09-10 | 浙江大学 | A kind of Shield-bored tunnels Defect inspection method based on deep learning |
| CN111649857A (en) * | 2020-04-23 | 2020-09-11 | 河海大学 | Inhaul cable modal measurement method for target matching analysis |
| CN111691679A (en) * | 2020-06-24 | 2020-09-22 | 北京工业大学 | Digital twinning-based intelligent tensioning method for prestressed steel structure |
| CN112378507A (en) * | 2020-11-05 | 2021-02-19 | 浙江大学 | Computer vision structure vibration monitoring method based on motion compensation |
| CN112378507B (en) * | 2020-11-05 | 2021-10-01 | 浙江大学 | Computer vision structure vibration monitoring method based on motion compensation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105741278B (en) | 2018-09-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105741278A (en) | On-line monitoring method of inhaul cable distribution stress on the basis of computer vision | |
| US10627219B2 (en) | Apparatus and methods for monitoring movement of physical structures by laser deflection | |
| CN105865735B (en) | A kind of bridge vibration test and dynamic characteristics recognition methods based on video monitoring | |
| Poozesh et al. | Large-area photogrammetry based testing of wind turbine blades | |
| CN105547635B (en) | A kind of contactless structural dynamic response measurement method for wind tunnel test | |
| CN103398860B (en) | A kind of ocean platform safety monitoring method of deformation based sensor | |
| KR101309443B1 (en) | Apparatus and method for monitoring a structure using motion capture method | |
| CN102798456B (en) | Method, device and system for measuring working amplitude of engineering mechanical arm support system | |
| CN111174961B (en) | Cable force optical measurement method based on modal analysis and measurement system thereof | |
| Dilek et al. | Condition monitoring of wind turbine blades and tower via an automated laser scanning system | |
| CN105403143A (en) | Measurement method and system thereof for acquiring displacement changes of earthquake simulator | |
| CN107860538B (en) | Detachable system widely applicable to multipoint dynamic deflection calibration and application thereof | |
| CN103017672A (en) | Non-contact nondestructive testing method for bridge structure | |
| CN111608731B (en) | Shield tunnel safety state monitoring and early warning device and monitoring and early warning method thereof | |
| Alamdari et al. | Non-contact structural health monitoring of a cable-stayed bridge: Case study | |
| CN109696209A (en) | The identification of underground piping and monitoring system and application method | |
| CN111504261A (en) | Tunnel three-dimensional automatic laser ranging device with parameter identification function | |
| Opoka et al. | Damage detection and localization method based on a frequency spectrum change in a scaled tripod model with strain rosettes | |
| CN104501732B (en) | A kind of hydro-structure distortion distributed optical fiber sensing device and method | |
| CN204286374U (en) | A kind of hydro-structure distortion distributed optical fiber sensing device | |
| Pan et al. | Enhancement of external wall decoration material for the building in safety inspection method | |
| Brown et al. | Evaluation of a novel video-and laser-based displacement sensor prototype for civil infrastructure applications | |
| CN104501772A (en) | Device and method for monitoring differential settlement in portfolio for soil-stone combined region with hydraulic structure | |
| CN104501773A (en) | Device and method for monitoring vertical deformation of hydraulic construction | |
| Sabato et al. | Experimental modal analysis of a utility-scale wind turbine blade using a multi-camera approach |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |