CN110207605A - A kind of measuring device and method of the metal structure deformation based on machine vision - Google Patents
A kind of measuring device and method of the metal structure deformation based on machine vision Download PDFInfo
- Publication number
- CN110207605A CN110207605A CN201910511836.0A CN201910511836A CN110207605A CN 110207605 A CN110207605 A CN 110207605A CN 201910511836 A CN201910511836 A CN 201910511836A CN 110207605 A CN110207605 A CN 110207605A
- Authority
- CN
- China
- Prior art keywords
- target
- light source
- film
- film target
- module
- 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
- 239000002184 metal Substances 0.000 title claims abstract description 51
- 229910052751 metal Inorganic materials 0.000 title claims abstract description 51
- 238000000034 method Methods 0.000 title claims abstract description 23
- 230000011218 segmentation Effects 0.000 claims description 13
- 230000009466 transformation Effects 0.000 claims description 12
- 238000000691 measurement method Methods 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims description 7
- 238000006073 displacement reaction Methods 0.000 claims description 6
- 239000000463 material Substances 0.000 claims description 4
- 230000000007 visual effect Effects 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000010586 diagram Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 230000009897 systematic effect Effects 0.000 claims description 3
- 241000208340 Araliaceae Species 0.000 claims 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims 1
- 235000003140 Panax quinquefolius Nutrition 0.000 claims 1
- 235000008434 ginseng Nutrition 0.000 claims 1
- 238000005259 measurement Methods 0.000 abstract description 11
- 238000005516 engineering process Methods 0.000 abstract description 5
- 230000003068 static effect Effects 0.000 abstract description 4
- 230000008901 benefit Effects 0.000 abstract description 2
- 230000003247 decreasing effect Effects 0.000 abstract description 2
- 239000003205 fragrance Substances 0.000 description 6
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 4
- 230000008859 change Effects 0.000 description 4
- 229910052742 iron Inorganic materials 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 229910003460 diamond Inorganic materials 0.000 description 1
- 239000010432 diamond Substances 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 239000011888 foil Substances 0.000 description 1
- 230000005389 magnetism Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
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
- G01B11/16—Measuring arrangements characterised by the use of optical techniques for measuring the deformation in a solid, e.g. optical strain gauge
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The present invention relates to metal structure deformation measurement technical fields, refer in particular to the measuring device and method of a kind of metal structure deformation based on machine vision, the present invention is compared with prior art, have the function of following remarkable advantage: having measurement metal structure static state and dynamic deformation amount, the especially deformation measurement in metal structure dynamic level direction；Special target light source target module and reference light source target module are devised, compared to other machine vision metrology technologies, there is stronger environment resistant light interference performance；Using reference light source target module, the camera shake problem due to caused by environment is eliminated, precision is improved；The multiple positions of metal component to be measured or specific position can targetedly be measured by increasing and decreasing, moving target light source target module；Using high resolution CMOS camera and sub-pixel positioning algorithm, precision can reach sub-pixel；It will not influence tested metal structure.
Description
Technical field
The present invention relates to metal structure deformation measurement technical fields, refer in particular to a kind of metal structure based on machine vision
The measuring device and method of deformation.
Background technique
With the continuous construction of China's infrastructure, the application of metal structure and realization scene are more and more, such as portal formula
Crane, derrick crane, autocrane, large bridge metal structure prefabricated component etc., but metallic magnetism is by external force
Influence and deformation occurs, once deformation it is excessive or be more than elastic limit, will deform excessively, be broken, cause thing
Therefore wait, it is therefore necessary to detect band large-scale metal structure static state and dynamic plasticity deformation, as camber, lower amount of deflection, angularity, across
The important parameters such as horizontal direction deflection under degree and external force.
In existing metal structure deformation measurement method, common method has instrument measuring method, sensor measurement method, GPS
Measurement method etc., instrument measuring method utilize the general measures such as theodolite, total station, level, rangefinder angle and displacement
Instrument obtains three-dimensional coordinate, these methods are unable to measure dynamic deformation in the presence of that can only measure static deformation, and measuring speed is slow, nothing
The problems such as method measures multiple spot simultaneously, measurement accuracy is influenced by operator；Sensor measurement method is currently used LVDT
Meter, fiber grating, foil gauge etc. are moved, these methods need and testee is contacted or is embedded in inside testee, can be to survey
Amount object itself interferes；GPS measurement method is non-contact measurement, and measurement is more convenient, but not can be carried out dynamic and survey
Amount, measurement accuracy are not able to satisfy metal structure requirement, and at high cost.Therefore existing instrument can not carry out metal structure dynamical state
Deformation measurement, can not quickly continuous precise measurement metal structure deformation under static state.
Summary of the invention
The present invention for problem of the prior art provide one kind can high-precision, it is remote it is non-contact, can be real-time continuous dynamic
State, and it is easily installed the measuring device and method of the metal structure deformation based on machine vision used.
In order to solve the above-mentioned technical problem, the present invention adopts the following technical scheme:
A kind of measuring device of metal structure deformation based on machine vision provided by the invention, including host computer, camera
Support frame, CMOS camera, target light source target module, reference light source target module and power supply unit, the target light source target
Module includes the first strong magnets, the first LED area light source and target film target, and the reference light source target module includes second
Strong magnets, the second LED area light source and reference film target, the CMOS camera is fixed on the camera tripod, described
CMOS camera configuration has zoom lens, and the host computer regulates and controls the CMOS camera, the target film target via network connection
Mark is pasted on first LED area light source surface, and first strong magnets are pasted on first LED area light source back side, institute
It states and is pasted on second LED area light source surface with reference to film target, second strong magnets are pasted on the 2nd face LED
Light source back surface, said supply unit are used for first LED area light source, the second LED area light source, the CMOS camera and described
Host computer power supply, the target light source target module and the reference light source target module pass through first strong magnetic respectively
Iron, second strong magnets are pasted on metal component to be measured.
Preferably, the target film target and the film target that refers to are provided in round, the target film target
Mark and the concentric loop that unlike material is provided with reference to film target, annulus in the middle part of the target film target are set
The center point there are two fan-shaped and two fan-shaped central angles in the target film target is set, it is described in the middle part of film target
The setting of annulus there are two fan-shaped and two fan-shaped central angles in the center point with reference to film target.
Preferably, the light that first LED area light source and second LED area light source issue is 1000-10000K
Natural light is provided with high light transmission on the light-emitting surface of first LED area light source and on the light-emitting surface of second LED area light source
PWMA panel.
Preferably, first LED area light source and second LED area light source are disposed as circle, the first LED
Area source outer end and second LED area light source outer end are provided with frame, the frame of first LED area light source and described
The width of the frame of two LED area light sources is 0.5-2cm, the face of the first LED area light source area and second LED area light source
Product is all larger than the area and the area with reference to film target of the target film target, and the target film target fits in
It is described to be fitted in the middle part of second LED area light source with reference to film target in the middle part of first LED area light source.
Preferably, the quantity of the target light source target module is multiple.
A kind of measurement method of the metal structure deformation based on machine vision, steps are as follows:
Step 1: target film target module is affixed at the endpoint of metal component to be measured by the first strong magnets, it will
It is affixed on metal component to be measured with reference to film target module by the second strong magnets, adjusts camera tripod, make CMOS camera
Face target film target module and reference film target module, target film target and reference film target module are in CMOS
In the visual field of camera, and it is imaged as close to the middle position of image；
Step 2: opening the first LED area light source, the focal length and aperture of the zoom lens of CMOS camera are adjusted, is acquired to be measured
Target film target module and the image with reference to film target module on metal component, and image data is passed by GigE agreement
Enter to host computer；
Step 3: being split using Threshold Segmentation Algorithm and prior information to image on host computer, it is luxuriant and rich with fragrance to obtain target
Woods target module and reference film target module region seek target film target module respectively and refer to film target mould
Coordinate (Xbeg_1, Ybeg_1), (Xbeg_2, Ybeg_2) of the block region starting point in original image；
Step 4: designing customized threshold using target film target and with reference to the existing physical size information of film target
Value-based algorithm carries out customized Threshold segmentation processing to target film target and with reference to film target region, obtains the target film
Target and the pixel edge round with reference to film target outermost, and obtain elliptical horizontal pixel point on the outside of target film target
Number xPiXcels, vertical pixel number yPiXcels；
Step 5: the pixel edge justified to the outermost of target film target and reference film target that the 4th step obtains
Sub-pixel edge is obtained using sub-pix detection and correcting algorithm, then sub-pixel edge is fitted, it is luxuriant and rich with fragrance to obtain target
Woods target and reference film target are in centre coordinate (Xti_1, Yti_1), (Xti_2, the Yti_2) for dividing region, finally
(Xbeg_1, Ybeg_1), (Xbeg_2, the Ybeg_2) sought using third step is obtained target film target and refers to film target
Centre coordinate (Xi_1, Yi_1), (Xi_2, the Yi_2) being marked in original image；
Step 6: according to xPiXcels, yPiXcels that the 4th step obtains, combining target film target outermost edges circle
Physical dimensional parameters and camera parameter, determine horizontal transformation factor kx from image coordinate system to world coordinate system, vertical
Transformation factor ky；
Step 7: centre coordinate (Xi_1, the Yi_ of the target film target and reference film target obtained using the 5th step
1), coordinate transform factor kx, ky that (Xi_2, Yi_2) and the 6th step obtain calculates each frame target film target and with reference to luxuriant and rich with fragrance
World coordinates (Xwi_1, Ywi_1), (Xwi_2, the Ywi_2) at woods target image center, and seek target film target and reference
The undulate quantity (Δ Xwi_1, Δ Ywi_1) at film target image center, (Δ Xwi_2, Δ Ywi_2)；
Step 8: the center undulate quantity of target film target image is subtracted the fluctuation with reference to film target image center
Amount obtains displacement variable (Δ Xwi, Δ Ywi), then draws out the position with initial position world coordinates difference Δ Xwi, Δ Ywi
Move curvilinear motion figure.
Preferably, customized Threshold Segmentation Algorithm described in the 4th step, specific step is as follows, and target film target is arranged
The initial threshold of logo image is 0, carries out Threshold segmentation to target film target, then seeks the target film target after threshold value separates
Logo image dimension information is compared with the physical size of target film target, when inconsistent, carry out to threshold value plus 1 operates, directly
Until when the target film target image dimension information after separation is consistent with the physical size information of target film target, currently
Threshold value be final threshold value.
Preferably, the detection of sub-pix described in the 5th step and correcting algorithm obtain sub-pixel edge, fitting, center
Specific step is as follows for coordinate, according to the pixel edge for the target film target outermost circle that the 4th step obtains, using spatial moment
Method obtains sub-pixel edge, and the systematic error eliminating this method and generating, Zhi Houli are then corrected to pixel edge
Ellipse fitting carried out to sub-pix point with least square method, obtain target film target cut zone image coordinate (Xti,
Yti), coordinate of the target film target center in original image is finally sought according to (Xbeg, the Ybeg) that third step obtains
(Xi,Yi)。
Preferably, the physical size information justified described in the 6th step according to the outermost edges of target film target,
Determine transformation factor kx, ky from image coordinate system to world coordinate system, transformation relation are as follows:
Beneficial effects of the present invention:
The measuring device and method of a kind of metal structure deformation based on machine vision provided by the invention, the present invention and existing
There is technology to compare, there is following remarkable advantage: devising special target light source target module and reference light source target module, phase
Than other machine vision metrology technologies, there is stronger environment resistant light interference performance；Using reference light source target module, eliminate
The camera shake problem due to caused by environment, improves precision；It can be by increasing and decreasing, moving target light source target module to be measured
The multiple positions of metal component or specific position are targetedly measured；Utilize high resolution CMOS camera and sub-pix
Location algorithm, precision can reach sub-pixel；It will not influence tested metal structure.
Detailed description of the invention
Fig. 1 is the structural diagram of the present invention.
Fig. 2 is the target light source target modular structure schematic diagram of the invention.
Fig. 3 is the target film target or described with reference to film target front view of the invention.
Appended drawing reference is respectively as follows:
Host computer -- 1, power supply unit -- 2, camera tripod -- 3, CMOS cameras -- 4, zoom lens -- 5, target light source
Target module -- 6, metal component to be measured -- 7, reference light source target module -- 8, target film target -- 9, the first strong magnetic
Iron -- 10, the first LED area light source -- 11.
Specific embodiment
For the ease of the understanding of those skilled in the art, the present invention is made further below with reference to embodiment and attached drawing
Bright, the content that embodiment refers to not is limitation of the invention.The present invention is described in detail below in conjunction with attached drawing.
Embodiment one
A kind of measuring device of metal structure deformation based on machine vision provided by the invention, including host computer 1, camera
Support frame 3, CMOS camera 4, target light source target module 6, reference light source target module 8 and power supply unit 2, the target light source
Target module 6 includes the first strong magnets 10, the first LED area light source 11 and target film target 9, the reference light source target mould
Block 8 includes the second strong magnets, the second LED area light source and refers to film target, and the CMOS camera 4 is fixed on the camera branch
On support 3, the CMOS camera 4 is configured with zoom lens 5, and the host computer 1 is via network connection (cable network or wireless network
Network) the regulation CMOS camera 4 parameter, the target film target 9 is pasted on 11 surface of the first LED area light source, institute
It states the first strong magnets 10 and is pasted on 11 back side of the first LED area light source, it is described to be pasted on described second with reference to film target
LED area light source surface, second strong magnets are pasted on second LED area light source back side, and said supply unit 2 is for giving
First LED area light source 11, the second LED area light source, the CMOS camera 4 and the host computer 1 are powered, the target light source
Target module 6 and the reference light source target module 8 are viscous by first strong magnets 10, second strong magnets respectively
It is affixed on metal component to be measured, the CMOS camera 4 is set to the target light source target module 6 and the reference light source target
8 front of module.
Specifically, in measurement, target light source target module 6 is placed on the position to be measured of metal component to be measured,
Reference light source target module 8 is placed on any metal component to be measured, and adjusts camera tripod 3, makes 4 face mesh of CMOS camera
It marks 9 module of film target and is in CMOS camera with reference to film target module, target film target 9 and with reference to film target module
It in 4 visual field, and is imaged as close to the middle position of image, the image information that CMOS camera 4 acquires is transmitted to upper
In machine 1, and handled, by seeking the displacement of 9 module of target film target, all cross jitter elimination algorithm, obtain to
Survey the deflection of metal component.
In the present embodiment, the target film target 9 and the film target that refers to are provided in round, the target film
On target 9 and the concentric loop for being provided with unlike material with reference to film target, it is preferable that concentric loop is in chequered with black and white
Setting, there are two fan-shaped and two fan-shaped central angles in the target for the setting of the annulus at 9 middle part of target film target
The center point of film target 9, there are two fan-shaped and two fan-shaped centers of circle for the setting with reference to the annulus in the middle part of film target
Angle is in the center point with reference to film target, as same transformation, also on target film target 9 and can will refer to film target
It is set as the symmetrical shape such as diamond shape.
In the present embodiment, the light that first LED area light source 11 and second LED area light source issue is 1000-
10000K natural light is respectively provided on the light-emitting surface of first LED area light source 11 and on the light-emitting surface of second LED area light source
There is high light transmission PWMA panel.
In the present embodiment, first LED area light source 11 and second LED area light source are disposed as circle, and described
One LED area light source, 11 outer end and second LED area light source outer end are provided with frame, and frame is made of black material, institute
The width for stating 11 frame of the first LED area light source and the second LED area light source frame is 0.5-2cm, first LED area light source
11 areas and the area of second LED area light source are all larger than the area and the reference film target of the target film target 9
Target area, the target film target 9 fits in 11 middle part of the first LED area light source, described to be bonded with reference to film target
In the middle part of second LED area light source.
A kind of measurement method of the metal structure deformation based on machine vision, steps are as follows:
Step 1: 9 module of target film target is affixed on metal component to be measured by the first strong magnets 10, will refer to
Film target module is affixed on metal component to be measured by the second strong magnets, is adjusted camera tripod 3, is made CMOS camera 4 just
CMOS is in 9 module of target film target and with reference to film target module, target film target 9 and with reference to film target module
In the visual field of camera 4, and it is imaged as close to the middle position of image；
Step 2: opening the first LED area light source 11, the focal length and aperture of the zoom lens 5 of CMOS camera 4, acquisition are adjusted
9 module of target film target and the image with reference to film target module on metal component to be measured, and image data is passed through into GigE
Agreement is passed to host computer 1；
Step 3: being split using Threshold Segmentation Algorithm and prior information to image on host computer 1, target is obtained
9 module of film target and reference film target module region seek 9 module of target film target respectively and refer to film target
Mark coordinate (Xbeg_1, Ybeg_1), (Xbeg_2, Ybeg_2) of the module region starting point in original image；
Step 4: designing customized threshold using target film target 9 and with reference to the existing physical size information of film target
Value-based algorithm carries out customized Threshold segmentation processing to target film target 9 and with reference to film target region, and it is luxuriant and rich with fragrance to obtain target
Woods target 9 and the pixel edge round with reference to film target outermost, and obtain the 9 elliptical horizontal picture in outside of target film target
Vegetarian refreshments number xPiXcels, vertical pixel number yPiXcels；
Step 5: the Pixel-level side justified to the outermost of target film target 9 and reference film target that the 4th step obtains
Edge obtains sub-pixel edge using sub-pix detection and correcting algorithm, is then fitted to sub-pixel edge, obtains target
Film target 9 and reference film target are in centre coordinate (Xti_1, Yti_1), (Xti_2, the Yti_2) for dividing region, most
(Xbeg_1, Ybeg_1), (Xbeg_2, the Ybeg_2) sought afterwards using third step is obtained target film target 9 and refers to the film
Centre coordinate (Xi_1, Yi_1), (Xi_2, Yi_2) of the target in original image；
Step 6: according to xPiXcels, yPiXcels that the 4th step obtains, 9 outermost edges of combining target film target
The parameter of round physical dimensional parameters and camera determines the horizontal transformation factor kx from image coordinate system to world coordinate system, erects
Straight transformation factor ky；
Step 7: centre coordinate (Xi_1, the Yi_ of the target film target 9 and reference film target obtained using the 5th step
1), coordinate transform factor kx, ky that (Xi_2, Yi_2) and the 6th step obtain calculates each frame target film target 9 and with reference to luxuriant and rich with fragrance
World coordinates (Xwi_1, Ywi_1), (Xwi_2, the Ywi_2) at woods target image center, and seek target film target 9 and reference
The undulate quantity (Δ Xwi_1, Δ Ywi_1) at film target image center, (Δ Xwi_2, Δ Ywi_2)；
Step 8: the center undulate quantity of 9 image of target film target is subtracted the fluctuation with reference to film target image center
Amount obtains displacement variable (Δ Xwi, Δ Ywi), then draws out the position with initial position world coordinates difference Δ Xwi, Δ Ywi
Move curvilinear motion figure.
In the present embodiment, customized Threshold Segmentation Algorithm described in the 4th step, specific step is as follows, and the target film is arranged
The initial threshold of target image 9 is 0, carries out Threshold segmentation to target film target 9, and it is luxuriant and rich with fragrance then to seek the target after threshold value separates
9 image dimension information of woods target is compared with the physical size of target film target 9, when inconsistent, carries out adding 1 to threshold value
Operation, when 9 image dimension information of target film target after separating is consistent with the physical size information of target film target 9
Until, current threshold value is final threshold value.
In the present embodiment, sub-pix described in the 5th step detection and correcting algorithm obtain sub-pixel edge, fitting, in
Specific step is as follows for heart coordinate, according to the pixel edge for 9 outermost of the target film target circle that the 4th step obtains, using space
Moment Methods obtain sub-pixel edge, the systematic error eliminating this method and generating then are corrected to pixel edge, later
Ellipse fitting is carried out to sub-pix point using least square method, obtains target film target 9 in the image coordinate of cut zone
(Xti, Yti) finally seeks 9 center of target film target in original image according to (Xbeg, the Ybeg) that third step obtains
Coordinate (Xi, Yi).
In the present embodiment, believed described in the 6th step according to the physical size of the outermost edges of target film target 9 circle
Breath determines transformation factor kx, ky from image coordinate system to world coordinate system, transformation relation are as follows:
Embodiment two
In the present embodiment, the quantity of the target light source target module 6 have it is multiple, by multiple target light source target modules 6
It is pasted on the different detection positions of mechanism to be detected by strong magnets respectively, it can multiple metal components to be measured simultaneously
Difference is detected, and testing principle is the same as example 1, herein not burdensome narration.
The above is only present pre-ferred embodiments, is not intended to limit the present invention in any form, although
The present invention is disclosed as above with preferred embodiment, and however, it is not intended to limit the invention, any person skilled in the art,
It does not depart within the scope of technical solution of the present invention, when the technology contents using the disclosure above make a little change or are modified to equivalent change
The equivalent embodiment of change, but without departing from the technical solutions of the present invention, technology refers to above embodiments according to the present invention
Made any simple modification, equivalent change and modification, belong in the range of technical solution of the present invention.
Claims (9)
1. a kind of measuring device of the metal structure deformation based on machine vision, it is characterised in that: including host computer, camera support
Frame, CMOS camera, target light source target module, reference light source target module and power supply unit, the target light source target module
Including the first strong magnets, the first LED area light source and target film target, the reference light source target module includes the second strength
Magnet, the second LED area light source and reference film target, the CMOS camera are fixed on the camera tripod, the CMOS
Camera configuration has zoom lens, and the host computer regulates and controls the CMOS camera via network connection, and the target film target is viscous
It is affixed on first LED area light source surface, first strong magnets are pasted on first LED area light source back side, the ginseng
It examines film target and is pasted on second LED area light source surface, second strong magnets are pasted on second LED area light source
The back side, said supply unit are used for first LED area light source, the second LED area light source, the CMOS camera and described upper
Machine power supply, the target light source target module and the reference light source target module pass through first strong magnets, institute respectively
The second strong magnets are stated to be pasted on metal component to be measured.
2. a kind of measuring device of metal structure deformation based on machine vision according to claim 1, it is characterised in that:
The target film target and the reference film target are provided in round, the target film target and the reference film target
Mark is provided with the concentric loop of unlike material, and there are two sector and two for the setting of the annulus in the middle part of the target film target
Fan-shaped central angle is in the center point of the target film target, and there are two the settings with reference to the annulus in the middle part of film target
Fan-shaped and two fan-shaped central angles are in the center point with reference to film target.
3. a kind of measuring device of metal structure deformation based on machine vision according to claim 1, it is characterised in that:
The light that first LED area light source and second LED area light source issue is 1000-10000K natural light, the first LED
On the light-emitting surface of area source and high light transmission PWMA panel is provided on the light-emitting surface of second LED area light source.
4. a kind of measuring device of metal structure deformation based on machine vision according to claim 1, it is characterised in that:
First LED area light source and second LED area light source are disposed as circle, first LED area light source outer end and described
Second LED area light source outer end is provided with frame, the side of the frame of first LED area light source and second LED area light source
The width of frame is 0.5-2cm, and the area of the first LED area light source area and second LED area light source is all larger than the mesh
The area and the area with reference to film target of film target are marked, the target film target fits in the first face LED
It is described to be fitted in the middle part of second LED area light source with reference to film target in the middle part of light source.
5. a kind of measuring device of metal structure deformation based on machine vision according to claim 1, it is characterised in that:
The quantity of the target light source target module is multiple.
6. a kind of measurement method of the metal structure deformation based on machine vision, which is characterized in that steps are as follows:
Step 1: target film target module is affixed on metal component to be measured by the first strong magnets, film target will be referred to
Mark module is affixed at the endpoint of metal component to be measured by the second strong magnets, is adjusted camera tripod, is made CMOS camera face
Target film target module and reference film target module, target film target and reference film target module are in CMOS camera
Visual field in, and be imaged as close to the middle position of image；
Step 2: opening the first LED area light source, the focal length and aperture of the zoom lens of CMOS camera are adjusted, metal to be measured is acquired
Target film target module and the image with reference to film target module on component, and image data is passed to by GigE agreement
Host computer；
Step 3: being split using Threshold Segmentation Algorithm and prior information to image on host computer, target film target is obtained
It marks module and refers to film target module region, seek target film target module respectively and refer to film target module institute
In coordinate (Xbeg_1, Ybeg_1), (Xbeg_2, Ybeg_2) of the area starting point in original image；
Step 4: designing customized threshold value using target film target and with reference to the existing physical size information of film target and calculating
Method carries out customized Threshold segmentation processing to target film target and with reference to film target region, obtains target film target
The round pixel edge with reference film target outermost, and obtain elliptical horizontal pixel points on the outside of target film target
XPiXcels, vertical pixel number yPiXcels；
Step 5: the pixel edge use of the target film target and the outermost circle with reference to film target obtained to the 4th step
Sub-pix detection and correcting algorithm obtain sub-pixel edge, are then fitted to sub-pixel edge, obtain target film target
Mark and reference film target are finally utilized in centre coordinate (Xti_1, Yti_1), (Xti_2, the Yti_2) of segmentation region
(Xbeg_1, Ybeg_1), (Xbeg_2, the Ybeg_2) that third step is sought obtain target film target and exist with reference to film target
Centre coordinate (Xi_1, Yi_1), (Xi_2, Yi_2) in original image；
Step 6: according to xPiXcels, yPiXcels that the 4th step obtains, the object of combining target film target outermost edges circle
The parameter of dimensional parameters and camera is managed, the horizontal transformation factor kx from image coordinate system to world coordinate system is determined, converts vertically
Factor ky；
Step 7: the target film target that is obtained using the 5th step and with reference to film target centre coordinate (Xi_1, Yi_1),
Coordinate transform factor kx, ky that (Xi_2, Yi_2) and the 6th step obtain calculates each frame target film target and refers to film target
Mark on a map world coordinates (Xwi_1, Ywi_1), (Xwi_2, the Ywi_2) of inconocenter, and seeks target film target and with reference to the film
The undulate quantity (Δ Xwi_1, Δ Ywi_1) at target image center, (Δ Xwi_2, Δ Ywi_2)；
Step 8: the center undulate quantity of target film target image is subtracted the undulate quantity with reference to film target image center, obtain
Displacement variable (Δ Xwi, Δ Ywi) is taken, the displacement curve with initial position world coordinates difference Δ Xwi, Δ Ywi is then drawn out
Variation diagram.
7. a kind of measurement method of metal structure deformation based on machine vision according to claim 6, it is characterised in that:
Customized Threshold Segmentation Algorithm described in 4th step, specific step is as follows, and the initial threshold of target film target image is arranged
It is 0, Threshold segmentation is carried out to target film target, then seeks the target film target image dimension information after threshold value separates,
It is compared with the physical size of target film target, when inconsistent, threshold value is carried out plus 1 operates, the target after separating
Until when film target image dimension information is consistent with the physical size information of target film target, current threshold value is as final
Threshold value.
8. a kind of measurement method of metal structure deformation based on machine vision according to claim 6, it is characterised in that:
The detection of sub-pix described in 5th step and correcting algorithm acquisition sub-pixel edge, fitting, specific step is as follows for centre coordinate,
The pixel edge of the target film target outermost circle obtained according to the 4th step obtains sub-pixel using space Moment Methods
Then edge is corrected the systematic error eliminating this method and generating, later using least square method to Asia to pixel edge
Pixel carries out ellipse fitting, obtains target film target in the image coordinate (Xti, Yti) of cut zone, finally according to third
It walks obtained (Xbeg, Ybeg) and seeks coordinate (Xi, Yi) of the target film target center in original image.
9. a kind of measurement method of metal structure deformation based on machine vision according to claim 6, it is characterised in that:
Described in 6th step according to the outermost edges of the target film target circle physical size information, determine from image coordinate system to
Transformation factor kx, ky of world coordinate system, transformation relation are as follows:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910511836.0A CN110207605A (en) | 2019-06-13 | 2019-06-13 | A kind of measuring device and method of the metal structure deformation based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910511836.0A CN110207605A (en) | 2019-06-13 | 2019-06-13 | A kind of measuring device and method of the metal structure deformation based on machine vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110207605A true CN110207605A (en) | 2019-09-06 |
Family
ID=67792555
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910511836.0A Pending CN110207605A (en) | 2019-06-13 | 2019-06-13 | A kind of measuring device and method of the metal structure deformation based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110207605A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114440776A (en) * | 2022-01-28 | 2022-05-06 | 上海交途科技有限公司 | Automatic displacement measuring method and system based on machine vision |
| WO2023028955A1 (en) * | 2021-09-02 | 2023-03-09 | Telefonaktiebolaget Lm Ericsson (Publ) | Method and apparatus for measurement |
| CN116309418A (en) * | 2023-03-09 | 2023-06-23 | 中建铁路投资建设集团有限公司 | Intelligent monitoring method and device for deformation of girder in bridge cantilever construction |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104359461A (en) * | 2014-11-06 | 2015-02-18 | 中国人民解放军装备学院 | Binocular vision measuring system having variable structure and parameter determining method |
| US20150138349A1 (en) * | 2012-07-04 | 2015-05-21 | Creaform Inc. | 3-d scanning and positioning system |
| CN104897062A (en) * | 2015-06-26 | 2015-09-09 | 北方工业大学 | Visual measurement method and device for shape and position deviation of part non-coplanar parallel holes |
| CN106871787A (en) * | 2017-04-13 | 2017-06-20 | 中国航空工业集团公司北京长城航空测控技术研究所 | Large space line scanning imagery method for three-dimensional measurement |
| CN106989669A (en) * | 2017-02-16 | 2017-07-28 | 上海大学 | Big visual field high-precision vision system calibrating method based on virtual three-dimensional target |
| CN107014293A (en) * | 2017-04-13 | 2017-08-04 | 中国航空工业集团公司北京长城航空测控技术研究所 | A kind of photogrammetric survey method of camera scanning imaging |
| CN107121109A (en) * | 2017-06-12 | 2017-09-01 | 北京航空航天大学 | A kind of structure light parameter calibration device and method based on preceding plated film level crossing |
| CN108458671A (en) * | 2018-03-08 | 2018-08-28 | 西安知微传感技术有限公司 | A kind of scaling method of line-structured light three-dimension measuring system |
| CN208155323U (en) * | 2018-04-16 | 2018-11-27 | 西安微普光电技术有限公司 | A kind of more optical axis automated calibration systems |
| US20190101383A1 (en) * | 2014-12-22 | 2019-04-04 | Bombardier Inc. | Reference system for online vision inspection |
| CN210154538U (en) * | 2019-06-13 | 2020-03-17 | 广东省特种设备检测研究院东莞检测院 | Metal structure deformation measuring device based on machine vision |
-
2019
- 2019-06-13 CN CN201910511836.0A patent/CN110207605A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150138349A1 (en) * | 2012-07-04 | 2015-05-21 | Creaform Inc. | 3-d scanning and positioning system |
| CN104359461A (en) * | 2014-11-06 | 2015-02-18 | 中国人民解放军装备学院 | Binocular vision measuring system having variable structure and parameter determining method |
| US20190101383A1 (en) * | 2014-12-22 | 2019-04-04 | Bombardier Inc. | Reference system for online vision inspection |
| CN104897062A (en) * | 2015-06-26 | 2015-09-09 | 北方工业大学 | Visual measurement method and device for shape and position deviation of part non-coplanar parallel holes |
| CN106989669A (en) * | 2017-02-16 | 2017-07-28 | 上海大学 | Big visual field high-precision vision system calibrating method based on virtual three-dimensional target |
| CN106871787A (en) * | 2017-04-13 | 2017-06-20 | 中国航空工业集团公司北京长城航空测控技术研究所 | Large space line scanning imagery method for three-dimensional measurement |
| CN107014293A (en) * | 2017-04-13 | 2017-08-04 | 中国航空工业集团公司北京长城航空测控技术研究所 | A kind of photogrammetric survey method of camera scanning imaging |
| CN107121109A (en) * | 2017-06-12 | 2017-09-01 | 北京航空航天大学 | A kind of structure light parameter calibration device and method based on preceding plated film level crossing |
| CN108458671A (en) * | 2018-03-08 | 2018-08-28 | 西安知微传感技术有限公司 | A kind of scaling method of line-structured light three-dimension measuring system |
| CN208155323U (en) * | 2018-04-16 | 2018-11-27 | 西安微普光电技术有限公司 | A kind of more optical axis automated calibration systems |
| CN210154538U (en) * | 2019-06-13 | 2020-03-17 | 广东省特种设备检测研究院东莞检测院 | Metal structure deformation measuring device based on machine vision |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023028955A1 (en) * | 2021-09-02 | 2023-03-09 | Telefonaktiebolaget Lm Ericsson (Publ) | Method and apparatus for measurement |
| CN114440776A (en) * | 2022-01-28 | 2022-05-06 | 上海交途科技有限公司 | Automatic displacement measuring method and system based on machine vision |
| CN116309418A (en) * | 2023-03-09 | 2023-06-23 | 中建铁路投资建设集团有限公司 | Intelligent monitoring method and device for deformation of girder in bridge cantilever construction |
| CN116309418B (en) * | 2023-03-09 | 2024-05-07 | 中建铁路投资建设集团有限公司 | Intelligent monitoring method and device for deformation of girder in bridge cantilever construction |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110207605A (en) | A kind of measuring device and method of the metal structure deformation based on machine vision | |
| CN107144241B (en) | A kind of binocular vision high-precision measuring method based on depth of field compensation | |
| US20220091486A1 (en) | Tripod for 3d modeling and method for determining camera capture parameters | |
| CN107401976B (en) | A kind of large scale vision measurement system and its scaling method based on monocular camera | |
| CN208795188U (en) | A kind of structured light binocular vision detection system | |
| CN105487555B (en) | A kind of station keeping method and device of unmanned plane | |
| CN106441163B (en) | A kind of contactless column measuring for verticality method and device | |
| CN109632085A (en) | A kind of low-frequency vibration calibration method based on monocular vision | |
| CN102788572B (en) | Method, device and system for measuring attitude of lifting hook of engineering machinery | |
| CN104574388B (en) | A kind of camera calibration system and its 3D scaling methods | |
| CN106441810B (en) | The detection device and detection method of the delay time of VR equipment | |
| CN105025290A (en) | Method for automatically adjusting inclination between image photographing module group sensor and lens | |
| CN208043364U (en) | A kind of crane girder Static stiffness detector | |
| CN102136140B (en) | Rectangular pattern-based video image distance detecting method | |
| CN106504287B (en) | Monocular vision object space positioning system based on template | |
| CN103075999B (en) | Real-time multi-target position detection method and system based on image | |
| CN106651957B (en) | Monocular vision object space localization method based on template | |
| CN107167088B (en) | The measurement method and device of glass deformation amount | |
| CN108637850A (en) | A kind of the milling robot system and control method of movement locus self study | |
| CN108805940B (en) | Method for tracking and positioning zoom camera in zooming process | |
| CN109212497A (en) | A kind of measurement of space six degree of freedom vehicle radar antenna pose deviation and interconnection method | |
| CN101782370A (en) | Measurement positioning method based on universal serial bus (USB) camera and method for measuring movement locus of moving object | |
| CN110186383A (en) | Monocular camera deflection metrology method based on the variation of the target point elevation angle | |
| CN108789412A (en) | A kind of robot motion's Trajectory Planning System, method and device | |
| CN103443580A (en) | System and method for calibrating a vehicle measurement reference system |
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 |