CN202204479U - Virtual optical extensometer - Google Patents

Abstract

The utility model discloses a virtual optical extensometer comprising a computer, wherein the input end of the computer is connected with an image acquisition device; the output end of the computer is successively connected with an image processing device, a data processing device and a result output device; the image acquisition device is a CCD camera; and the CCD camera is connected with the computer through a USB interface. The virtual optical extensometer has the advantages of being convenient for mounting, saving space, having unlimited measuring range, and preventing the extensometer from damaging due to change of the measuring range. The right and wrong contact measurement has no additional influence on the measured objects. The virtual optical extensometer also has the advantages of good linearity, high measurement accuracy, no cumulative errors and the like.

Description

The virtual optics extensometer

Technical field

The utility model relates to a kind of a kind of instrument and measuring method thereof of measuring average line strain between member and other objects at 2, be specifically related to a kind of need not to contact with test specimen can measure and virtual optics extensometer easy to use, with low cost.

Background technology

Extensometer (extensometer) is a kind of instrument of measuring average line strain between member and other objects at 2, is made up of sensor, amplifier and register three parts usually.In measuring process, the direct and tested member of sensor contacts.

At present, the most frequently used extending counted the resistance-type extensometer.When measuring strain with this kind extensometer; Extensometer is installed on test specimen, and edge contacts with test specimen and test piece deformation is delivered to distortion and transmits on the bar, further is delivered to flexible member again; Transmit bar perception strain through distortion; Stick on the foil gauge perception strain on the flexible member and be converted into resistance change, export after converting resistance variations into change in voltage with suitable measurement amplifying circuit, finally obtain the size of strain through the voltage signal values that records.This kind extensometer need contact test specimen when measuring, need lead to draw signal, therefore when measuring flexible material deformation or in particular surroundings (like HTHP), measure, has difficulties.In addition, because the foil gauge range is limited, and between the edge of a knife certain distance will be arranged, this kind extensometer can not be measured the distortion of small test specimen, when measuring large deformation, also has problems.

The utility model content

Goal of the invention: in order to overcome the deficiency that exists in the prior art, the utility model provide a kind of need not to contact with test specimen can measure and measuring accuracy is high, the linearity is good and the virtual optics extensometer of saving installing space.

Technical scheme: for realizing above-mentioned purpose; The virtual optics extensometer of the utility model; Comprise computing machine, image collecting device, image processing apparatus, data processing equipment and output unit as a result; Said input end and computer is connected with image collecting device, and the output terminal of computing machine and image processing apparatus, data processing equipment are connected with output unit as a result successively.

Said image collector is changed to the CCD camera, and the CCD camera is connected through USB interface with computing machine.

Said image collecting device is sent to computing machine with the original image that collects; Said image processing apparatus extracts the gauge point image from original image, said data processing equipment calculates according to the gauge point image that is extracted and the result is exported through output unit as a result.

The CCD camera is for to use camera with charge-coupled image sensor as the remote sensing of light sensor and photoelectric commutator.The English full name of CCD is Charge-coupled Device, and Chinese full name is a charge coupled cell, can be called ccd image sensor.CCD is a kind of semiconductor devices, can be converted into digital signal to optical image.The small photoactive substance of the last implantation of CCD is called pixel (Pixel).The pixel count that comprises on CCD is many more, and its screen resolution that provides is also just high more; The effect of CCD is just as film, but it is to convert image pixel to digital signal; The last electric capacity that many marshallings are arranged of CCD can be responded to light, and image is transformed into digital signal; Via the control of external circuit, each little electric capacity can with its with electric charge be given to its adjacent electric capacity.

Beneficial effect: compared with prior art, the virtual extensometer of the utility model is easy for installation, and it is little to take up space; Range is unrestricted, can extensometer destroyed; Be non-cpntact measurement, measured object is not had any added influence; Have that the linearity is good, measuring accuracy is high and do not have advantages such as cumulative errors.

Description of drawings

Fig. 1 is the structured flowchart of the virtual optics extensometer of the utility model;

Fig. 2 is the workflow diagram of image collecting device;

Fig. 3 is the measured real-time displacement time curve;

Fig. 4 is a measured real-time strain time curve;

Fig. 5 is the correlation curve of measured value and actual displacement;

Fig. 6 is the orientation synoptic diagram of peach test specimen marked point;

Fig. 7 is a stress-strain diagram.

Embodiment

Below in conjunction with accompanying drawing the utility model is done explanation further.

As shown in Figure 1; The virtual optics extensometer of the utility model; Comprise computing machine 1, image collecting device 2, image processing apparatus 3, data processing equipment 4 and output unit 5 as a result; The input end of computing machine 1 is connected with image collecting device 2, and the output terminal of computing machine 1 and image processing apparatus 3, data processing equipment 4 are connected with output unit 5 as a result successively.

In the present embodiment, image collecting device 2 is the CCD camera, and CCD camera and computing machine 1 are connected through USB interface.

The utility model in use; Image collecting device 2 is sent to computing machine 1 with the original image that collects; Image processing apparatus 3 extracts the gauge point image from original image, data processing equipment 4 calculates according to the gauge point image that is extracted and the result is exported through output unit 5 as a result.

The virtual optics extensometer of the utility model specifically may further comprise the steps in use:

1) at two gauge point of test specimen marked to be measured; 2) image collecting device 2 real time record comprise the picture of these two gauge points, and are input in the computing machine 1; 3) image processing apparatus 3 through filtering, binaryzation cut apart, the zone is selected and the gauge point image is extracted in the closed operation of image from original image; 4) data processing equipment 4 at first calculates the grey scale centre of gravity coordinate of gauge point according to following formula,

$x=\frac{\underset{1}{\overset{M\×N}{\mathrm{\Σ}}}i\·I(i,j)}{\underset{1}{\overset{M\×N}{\mathrm{\Σ}}}I(i,j)},$ $y=\frac{\underset{1}{\overset{M\×N}{\mathrm{\Σ}}}j\·I(i,j)}{\underset{1}{\overset{M\×N}{\mathrm{\Σ}}}I(i,j)}$

Wherein, M is the total pixel number on the picture horizontal direction, and N is the total pixel number on the picture vertical direction, and I is the gray-scale value of pixel, and i is the horizontal ordinate of pixel, and j is the ordinate of pixel;

Then, be translated into actual displacement value and draw displacement time curve; 5) output unit 5 is exported The above results with numeral or analog quantity as a result.

In practical operation, the shape of identification point can be arbitrarily, as long as make things convenient for the calculating of mathematical method.Such as the identification point of cross symbol, can be through the intersection point that arrives of fitting a straight line; The identification point of annular can match obtain circular curve, thereby obtain central coordinate of circle; The shape of gauge point can also be rectangle.

The virtual optics extensometer of the utility model is the basis with the Digital Image Processing, and adopts a kind of special gauge point recognition methods, i.e. grey scale centre of gravity method.

The grey scale centre of gravity method is the gray scale of Pixel of Digital Image to be regarded as " density " of digital picture, thinks that be centered close to its " center of gravity " of a spot located in the image.For a planar object, establish it and occupy the closed region D on the xoy face, point (x, the surface density of y) locating be ρ (x, y), its barycentric coordinates are:

$x_o=\frac{\∫\underset{D}{\∫}x\·\mathrm{\ρ}(x,y)\mathrm{dxy}}{\∫\underset{D}{\∫}\mathrm{\ρ}(x,y)\mathrm{dxy}},$ $y_o=\frac{\∫\underset{D}{\∫}y\·\mathrm{\ρ}(x,y)\mathrm{dxy}}{\∫\underset{D}{\∫}\mathrm{\ρ}(x,y)\mathrm{dxy}}---\left(1\right)$

Correspondingly, be the gray level image zone of M * N for width of cloth size, the gray-scale value of establishing its each pixel be I (i, j), (grey scale centre of gravity coordinate that can defined label point is for 0＜i＜m, 0＜j＜n):

$x=\frac{\underset{1}{\overset{M\×N}{\mathrm{\Σ}}}i\·I(i,j)}{\underset{1}{\overset{M\×N}{\mathrm{\Σ}}}I(i,j)},$ $y=\frac{\underset{1}{\overset{M\×N}{\mathrm{\Σ}}}j\·I(i,j)}{\underset{1}{\overset{M\×N}{\mathrm{\Σ}}}I(i,j)}---\left(2\right)$

Wherein, M is the total pixel number on the picture horizontal direction, and N is the total pixel number on the picture vertical direction, and I is the gray-scale value of pixel, and i is the horizontal ordinate of pixel, and j is the ordinate of pixel;

When measuring, the position of different time charts can be detected, the displacement of this gauge point can be obtained according to same gauge point position with respect to the variation of initial time as marked point with formula (2).The relative displacement of two gauge points then is the strain that will measure with the ratio of the distance L of initial time two gauge points.

The virtual instrument development platform LabVIEW of American National instrument (NI) company; The programming of use graphical programming language, friendly interface, easy to learn; Supporting image processing software bag can provide abundant image to handle and the analytical algorithm function; Be very easy to the user, make make up Flame Image Process and analytic system easily, flexible, portability of program is good, has shortened system development cycle greatly.Releasing on the basis of application software, NI company has released image pick-up card again, can carry out seamless link with LabVIEW for the image pick-up card of NI company, but shortcoming is to cost an arm and a leg and be difficult for carrying out low level development.Nowadays along with the progress of science and technology, the camera of band USB interface is more and more universal, and price is more and more cheaper, and performance is also increasingly high, adopts digital camera+USB pattern just more and more to receive visual item slip-stick artist's favor.Because adopt digital camera+USB pattern not only to practice thrift design cost, but also greatly improved the dirigibility of system.The utility model just is being based on such purpose, has proposed the method that a kind of CCD camera DH-HV-1303UM. that under the LabVIEW environment, cooperates Daheng's USB interface realizes virtual extensometer.

Company of Daheng is VC++; The software development application configuration that VB etc. are commonly used the corresponding driving program; But the corresponding driving program but is not provided under the LabVIEW platform, so we need use the method for calling dynamic link library to realize the operation to camera in labview.At first under vc++6.0, develop a dynamic link library, the function to the collection of the setting of camera and image is accomplished in this dynamic chain storehouse; Call this dynamic link library through Call Library Function device among the LabVIEW then; Thereby accomplish the collection to image through labview.Note setting the content such as title, type of data, rreturn value of title and path, the function of dynamic link library when in labview, calling dynamic link library.In labview, call the function among the DLL, maximum difficulty is that the data type of function parameter is mapped as the data type among the corresponding labview.

As shown in Figure 2, step 2) in, the picture that image collecting device 2 real time record comprise these two gauge points comprises following flow process: 1. open camera and initialization; 2. resolution is set; 3. drainage pattern is set; 4. gain is set; 5. time shutter is set; 6. output.

Below 3 flow processs of image processing apparatus in the step 3) are specified.

Filtering: because the influence of external environment and the influence of camera internal noise; Inevitably there is noise in the image of gathering; In serious time,, institute's information of interest was flooded by noise, and therefore the characteristic according to institute's images acquired is provided with rational wave filter, strengthens interested characteristic; Weaken The noise, be absolutely necessary.In the VISION of LabVIEW device, wide variety of different types of filtering algorithms are provided, existing linear filter method also has the nonlinear filtering method; Existing spatial domain filter method also has the frequency filtering method.Can be according to the rational filtering parameter of the image setting that reality is gathered.The utility model adopts the convolutional filtering method that smoothing effect is arranged.

Binaryzation is cut apart: next behind the image process smoothing denoising is binaryzation, i.e. gray level threshold segmentation.Image binaryzation promptly to given threshold interval (T1, T2), between T1, it is 0 that gray-scale value between the T2 is composed with pixel gray-scale value in the entire image, it is 1 that other pixel gray-scale value is composed, thereby the black white image of 0 and 1 two gray-scale value is only arranged.The mathematical form of gray level image binaryzation is following:

In order farthest to strengthen the contrast of gauge point and background image, the main part that so just will detect is effectively separated with its background image.

The zone is selected: after the surveyed area image binaryzation is cut apart; Inevitably there are some other noise spot similar or zones with the detected object pixel value; Can hinder the characteristic of extracting detected object, need do some necessary forms processing and as much as possible it filtered out.When extracting gauge point; Adjacent with gauge point and close with gauge point pixel value interference region also is extracted out; This has brought inconvenience to subsequent treatment; Must eliminate, the utility model is taked closed region area method for screening, promptly in Flame Image Process, only keeps and the similar zone of gauge point area size; To reject with the zone that the gauge point area differs bigger.According to mark point and be provided with between rational reservation area, this type of interference just can be eliminated.

The closed operation of image: image is being carried out binaryzation when cutting apart, owing to receive the influence of illumination variation and camera internal noise, gauge point is inner can the irregular noise spot of appearance.These noise spots must be got rid of, because these noise spots will have bigger influence to calculating grey scale centre of gravity.These noise spots can be eliminated through the closed operation of image.Closure operation can be filled and led up hole tiny in the profile, makes the small gap on the profile up, and the total position and the shape invariance of target signature.

In the step 4), what try to achieve is pixel value, need this pixel value be changed into actual displacement value according to the result of rating test gained.

In the step 5), output unit 5 is exported institute's result of calculation through data collecting card with numeral or analog quantity as a result.

Before on-test, need know the relation between the pixel and actual displacement in certain visual field earlier.Prepare the demarcation target of two circular gauge points that comprise, the actual center gravity distance of this circular gauge point is known, tries to achieve its grey scale centre of gravity distance through system then, notices that the distance of being asked this moment is to be unit with the pixel.Through it is compared with the actual center gravity distance.Can obtain the relation between pixel and the actual displacement.In experiment, the gauge point radius is got 0.5cm, and the distance in its center of circle is 2cm, and these two gauge points fixedly are placed on the flat board place on the translation stage, the adjustment camera lens, and images acquired is clear until institute; Open the grey scale centre of gravity distance (pixel) that computing machine 1 calculates two gauge points, suppose that the various distortion of camera lens and camera are all eliminated, the move left and right translation stage is measured 20 groups of data and is got its mean value then, and obtaining its mean value is 500.025; Can be regarded as this moment to such an extent that the actual displacement of pixel representative is 20000/500.25=39.998um.Therefore can think that a pixel is with respect to 0.04mm in this visual field.

Still with this translation stage the utility model is demarcated, kept viewing field of camera and enlargement factor constant, place two flat boards that comprise gauge point on the translation stage, gauge point and background gray levels differ more greatly and get final product.One of them is fixed, and one is moved with translation stage.The distance of two marks can change when translation stage moved, and came the stretching experiment of imitation specimen with this.The resolution 10 μ m of translation stage wherein, the resolution 1280 * 1024 of CCD camera,, frame per second: SXGA (1280 * 1024): 15 frame/seconds, the adjustable focus scope of zoom lens is 24～85mm.

Regulate the enlargement ratio of camera, make a pixel be equivalent to 1mm.Translation stage is moved the interval get 10 μ m, 20 μ m, 40 μ m, 50 μ m, 60 μ m (being scaled image resolution ratio is 0.01,0.02,0.04,0.05 and 0.06 pixel) respectively.Move at interval mobile respectively accurate translation stage 20 times for every kind.The result who measures is compared with actual value.The result who measures is compared with actual value.Its result is as shown in the table:

The speed of translation stage with 10um/s is moved, and Measuring Time is 20 seconds.Fig. 3 is the measured real-time displacement time curve, and Fig. 4 is a measured real-time strain time curve.

Translation stage is moved the interval get 500um, move 20 times, the correlation curve of the actual displacement of the utility model institute's result calculated and translation stage is as shown in Figure 5.

Can be found out by experimental data in the table, be that systematic error is bigger under 0.01 pixel, 0.02 pixel in image resolution ratio.In image resolution ratio is that systematic error does not have significant change under 0.05 pixel, 0.06 pixel, and maximum relative error is less than 3.5%, and average error is less than 1.5 μ m.So desirable 0.05 pixel of resolution of system.With the resolution of CCD camera is 1024 * 1024 to calculate.The strain measurement resolution of this extensometer is about 48 μ ε (0.05/1024=48 μ ε).If the resolution of CCD camera is 2048 * 2048, then the strain measurement resolution of this extensometer is about 24 μ ε.And can find out that by Fig. 5 the utility model can keep better linearity degree and measuring accuracy in the wide range scope.

For example, when peach is measured, will do last four gauge points by the test specimen that fresh peaches is processed into, as shown in Figure 6, gauge point 1 and 3 line are horizontal direction, and gauge point 2 and 4 line are vertical direction.Strain on the test specimen x direction can calculate from gauge point 1 and gauge point 3, and the strain on the test specimen y direction then can be calculated from gauge point 2 and gauge point 4.This test specimen is clamped on the MTS testing machine does tension test, use based on the extensometer real time record test specimen x direction of virtual instrument and the strain on the y direction.The related data of compatibility test machine obtains stress-strain diagram shown in Figure 7, and solid line is the stress-strain diagram of x direction among the figure, and dotted line is the stress-strain diagram of y direction.Can find out that from curve map the elastic modulus of peach is approximately 2.75MPa, the Poisson ratio of peach is approximately 0.3877.

The above only is the preferred implementation of the utility model; Be noted that for those skilled in the art; Under the prerequisite that does not break away from the utility model principle; Can also make some improvement and retouching, these improvement and retouching also should be regarded as the protection domain of the utility model.

Claims (2)

1. virtual optics extensometer; It is characterized in that: comprise computing machine (1), image collecting device (2), image processing apparatus (3), data processing equipment (4) and output unit (5) as a result; The input end of said computing machine (1) is connected with image collecting device (2), and the output terminal of computing machine (1) and image processing apparatus (3), data processing equipment (4) are connected with output unit (5) as a result successively.

2. virtual optics extensometer according to claim 1 is characterized in that: said image collecting device (2) is the CCD camera, and the CCD camera is connected through USB interface with computing machine (1).

Images