JP4831703B2 - Object displacement measurement method - Google Patents

Description

The present invention, measurement of the deformation stress distribution of various materials and design phase or bridges in use, building large structures of the displacement amount of deflection distribution measurement etc. suitably displaced ones body Ru can be performed, such as such as about the measurement how.

  Conventionally, various measuring devices such as displacement gauges and strain gauges have been used to measure the displacement and stress of objects such as materials and structures, but displacement gauges and strain gauges can only measure displacement gauges. It is possible to obtain only information on the place where the strain gauge is installed or where the strain gauge is attached by bonding or the like. Therefore, to obtain the displacement distribution and stress distribution of the entire object, it is necessary to use a large number of displacement gauges and strain gauges, but it is not practical due to problems such as labor and cost, and solves such problems. For this purpose, the following prior art has been proposed.

  In Patent Document 1, a measurement tape having a plurality of anisotropic piezoelectric materials attached thereto is adhered to the surface of a structural member, and the surface potential of the anisotropic piezoelectric material is measured using a surface potentiometer. It has been proposed to determine the stress distribution.

  In the prior art of Patent Document 1, since measurement is basically performed with a measurement probe one by one, there is a problem that it takes a lot of time to obtain a deformation stress distribution. In addition, since the anisotropic piezoelectric material itself has a standard size, there are many limitations on the distance and size that can be measured, and there is a problem that the usability is poor. Furthermore, since anisotropic piezoelectric materials are expensive, when it is necessary to measure many points, for example, one million points, there is a problem that extremely high cost and much labor are required. Further, in a structure, there is a problem that it is difficult to perform multipoint measurement using this surface electrometer in a place where it is difficult to move the surface electrometer after being fixed to the structure as a measuring device. .

  Further, Patent Document 2 proposes a technique capable of calculating a deformation amount using a digital image correlation method and obtaining a displacement amount of the entire visual field with an inexpensive device.

  The prior art of Patent Document 2 has a problem that the calculation result cannot be seen in real time because it takes much calculation time to calculate the displacement amount. Moreover, although it is necessary to affix a random pattern on the surface of a structure, there exists a problem that it is not easy to change the magnitude | size of a random pattern and it is difficult to make a random pattern according to a measurement precision. Furthermore, the measurement accuracy is about 1/10 pixel to 1/20 pixel, and there is a problem that high measurement accuracy cannot be obtained. Furthermore, since this prior art uses the luminance value of the photographed image, there is a problem that it is easily affected by ambient light.

JP 2003-185507 A JP 2006-329628 A

An object of the present invention is to provide a displacement measuring how things body Ru can detect an object displacement, such as materials and structures with a simple configuration with high accuracy.

The present invention forms a regular pattern on a surface of an object by sticking a two-dimensional sheet body on which a lattice pattern having a rectangular waveform or a sinusoidal waveform with a pitch interval corresponding to the accuracy to be measured is drawn. a step of, by the regular pattern as a reference grid, a respective predetermined areas before and after the application of force, shooting with a plurality of optical cameras installed at different positions,
A moiré fringe image is obtained by performing sampling for each pixel at equal intervals on a photographed image a plurality of times, three or more times while changing the starting pixel, and interpolating the thinned image obtained by this sampling processing. And obtaining each phase distribution in the horizontal and vertical directions of the moire fringes obtained by the phase shift method;
Calculating the displacement distribution of the object from the phase difference distribution obtained from the phase distribution before and after deformation ;
And calculating a strain or stress distribution based on the displacement amount distribution .

  According to the present invention, the adhesive tape or the like of the present invention is pasted on the portion to be measured, and is photographed with an optical photographing device such as a CCD camera or a CMOS camera. The photographed image is sampled at equal intervals. , By performing the multiple times three or more times while changing the starting pixel, generating a moire fringe image by interpolating the thinned image obtained by this sampling process, obtaining the phase distribution of the moire fringes by the phase shift method, From the phase difference distribution obtained from the phase distribution before and after the deformation, the displacement stress distribution can be obtained easily and with high accuracy to detect the displacement of the object.

Also , the pitch interval and size can be determined according to the accuracy and range to be measured. In other words, if it is desired to measure a minute deformation, a fine grid may be used. Further, when it is desired to measure a large deformation, a grating having a large pitch interval may be used, and a grating interval corresponding to the required accuracy can be arbitrarily selected and used.

In addition , since a lattice pattern sheet body having a rectangular waveform or a sinusoidal brightness distribution with a pitch interval of 10 to 1000 times the accuracy to be measured is used, phase analysis can be performed with high accuracy, and therefore object with high accuracy can be obtained. Can be obtained.

In addition , since a time-consuming process such as searching is not performed, the calculation time is faster than the conventional method, and the deformation of the object can be detected with high accuracy in a short time .

In addition , if the equipment necessary for the measurement is a general-purpose personal computer, an existing optical camera, and a sheet body, it is possible to measure the phase distribution in the horizontal direction and the vertical direction and perform highly accurate displacement measurement. The displacement of the object can be measured with high accuracy at a low cost, and a highly versatile measurement technique can be provided.

In addition , since the lattice pattern is formed on the surface of the object by pasting the sheet body on the surface of the object, when the lattice pattern is formed on the measurement target surface, the restrictions on the properties of the object surface are relaxed, and the sheet body is pasted. The present invention can be carried out on a wide range of objects.

Furthermore, the present invention is characterized in that the displacement of the object is detected three-dimensionally.
The present invention uses the two-dimensional phase distribution, and detecting the corresponding points in the captured image of the plurality of cameras.

  Furthermore, the present invention is characterized in that the three-dimensional shape of the object is measured using the detection results of corresponding points in the captured images of the plurality of cameras.

  Furthermore, the present invention is characterized in that a strain distribution on the surface of the object is obtained using the measured three-dimensional shape of the object.

  According to the present invention, the displacement of an object can be obtained as a parameter or information representing a displacement such as a displacement stress distribution in a simple and non-contact manner without contact.

  Further, according to the present invention, the lattice spacing and size can be arbitrarily determined according to the accuracy and size desired to be measured, and the displacement of the object can be obtained with the measurement accuracy corresponding to the measurement object.

  Furthermore, according to the present invention, since only a sheet body on which a lattice pattern is drawn, a general-purpose optical camera, and a personal computer are required for the measurement, information on displacement such as displacement stress distribution is measured in a short time at a low cost. be able to.

  Furthermore, according to the present invention, the present invention can be applied to displacement measurement of objects such as many materials and structures having different materials and shapes, and the present invention can be widely implemented.

  FIG. 1 is a perspective view showing an appearance of a measuring tape 1 used in the object deformation measuring method according to an embodiment of the present invention, and FIG. 2 is a lattice pattern 2 displayed on the measuring tape 1 shown in FIG. FIG. In FIG. 2, FIG. 2 (a) shows a vertical black-and-white binary grid, FIG. 2 (b) shows a horizontal black-and-white binary grid, FIG. 2 (c) shows a vertical black-and-white binary grid, 2 (d) shows a grid having a brightness change in a horizontal sine waveform or cosine waveform, FIG. 2 (e) shows a black and white square grid in the vertical and horizontal directions, and FIG. 2 (f) shows a vertical and horizontal grid. FIG. 2 (g) shows a black and white dot-like grid in the vertical and horizontal directions, and FIG. 2 (h) shows a black and white dot-like grid in the vertical and horizontal directions (brightness inversion). FIG. 2 (i) shows a grating having brightness changes in sine or cosine waveforms in the vertical and horizontal directions.

  The measuring tape 1 is obtained by printing a reference grid serving as a scale for measuring the amount of displacement on the surface of a conventional plain adhesive tape. As shown in FIGS. 2A to 2I, the printed lattice pattern is a pattern having a regular pattern at regular intervals, and is a black and white lattice having black squares “■” and black circles “●”. This is realized by a grating having a brightness change of a pattern and a sine waveform or a cosine waveform. The arrangement is two-dimensionally arranged in the vertical direction, the horizontal direction, or the vertical and horizontal directions. When measuring the deformation of an object to be measured, the measuring tape 1 having such a reference lattice can be directly attached to the surface of the place to be measured.

  The accuracy with which the displacement can be detected using the measuring tape 1 is about 1/1000 of the lattice spacing, but in actuality, the accuracy is practically approximately the lattice spacing due to the influence of ambient light, camera noise, and the like. 1/100. Therefore, for example, when it is desired to detect a displacement of 10 μm, a grid with a 1 mm interval may be used.

  Since such a measuring tape 1 only needs to be directly attached to an object to be measured, it is easy to use on the site, is inexpensive, and can be measured quickly to perform image analysis. As a method of image analysis, a sampling moire method is used, and its principle will be described next.

  FIG. 3 is a diagram for explaining the principle of the phase shift moire method. Sampling moire method is a method of taking a four-pixel pitch grid pattern as a reference grid with a CCD camera, generating four phase-shifted moire patterns based on the obtained image, and generating the phase of moire fringes. This is a technique for obtaining the distribution. When the reference grating shown in FIG. 3A is attached to the surface of the object and photographed by the CCD camera, adjustment is made so that one period of the grating is four pixels of the CCD camera. In FIG. 3A, the black dot “·” indicates the center point of sampling of the CCD camera, FIG. 3B shows the reference grid, and FIG. 3C shows the reference grid of FIG. The obtained image is shown.

  In the CCD camera, since the amount of light incident on the square area of the image pitch is detected as an integral value, gray (grayscale) data d1, d2,. At this stage, moire fringes cannot be observed, but if thinning processing, that is, sampling processing is performed every four pixels, moire can be seen as shown in FIGS. 3 (d) to 3 (g). 3D is thinned out every four pixels from the first sampling point from the left, while FIGS. 3E to 3G are thinned out from the second, third, and fourth sampling points, respectively. The phase shift process is performed. Further, if the previous data is duplicated to the luminance value of three pixels with no remaining data, an image as shown in FIGS. 3 (h) to 3 (k) is obtained, and the moire fringes are further enhanced.

  First, an image of an object to be measured is obtained by setting and photographing an optical camera so that the interval of the reference grid is 3 pixels or more of the camera. At this time, it is desirable to install the camera so that the direction of the optical axis of the camera is as perpendicular as possible to the imaging surface of the object.

  The image taken by the camera is thinned out with the number of pixels N (N ≧ 3) corresponding to the interval of the reference grid. Here, N is an integer, but there is a method of analysis even if N is not an integer. Here, as a thinning-out method, data representing luminance is extracted for each N pixel based on the first pixel from either the top, bottom, left, or right direction of the image. Next, in the same manner from the second pixel to the (N−1) th, data representing luminance is extracted for each N pixel. The luminance distribution of the nth moire fringe in the total N obtained in this way can be obtained by the following equation (1).

Here, _Ia and _Ib are the luminance amplitude and the background luminance, respectively. θ is the phase of moire fringes.

  FIG. 4 is a diagram for explaining image processing necessary to obtain a plurality of phase-shifted moire fringe images. Here, a case where the thinning-out number is 3 is shown. 4A shows the original luminance value of the photographed image, and FIG. 4B shows the luminance value obtained by thinning out every third pixel from the first start point from the left. FIG. 4C shows luminance values obtained by thinning out every third pixel from the second start point from the left, and FIG. 4D shows thinning out every third pixel from the third start point from the left. The obtained luminance value is shown.

  Originally, it is most ideal that the brightness distribution of the reference grid is a sine waveform or a cosine waveform as shown in FIGS. 2 (c), 2 (d), and 2 (i). However, even if a black and white binary lattice such as FIG. 2 (a), FIG. 2 (b), FIG. 2 (e), FIG. 2 (f), FIG. 2 (g), or FIG. By performing the conversion processing or the like, it can be approximately expressed as in Expression (1). If the thinning process is performed, the pixel size is reduced by N times, but if the missing image data is linearly interpolated as shown in FIG. 4D, an image having the same resolution as the original image can be obtained. An interpolation method other than linear interpolation can be used if the same effect can be obtained.

  For example, in FIG. 4D, by thinning out, B and C data missing between the pixel A and the pixel D can be obtained from the data of the pixels A and D by linear interpolation.

  The phase θ of the moire fringes can be obtained by the following equation (2) using the plurality of phase-shifted images.

  In particular, when N = 4, it can be easily obtained from equation (3).

Next, consider a case where the sample with the reference grid is displaced in-plane. Since the phase of the moire fringe changes when the entire sample is displaced, the phase distributions in the x direction and y direction before displacement are φ _x0 and φ _y0 , respectively, and the phase distributions in the x direction and y direction after displacement are φ, respectively. _x1, When phi _y1, the phase difference of the longitudinal displacement x and y directions, a Δφ _x = φ _x1 -φ _x0 respectively, becomes φ _y = φ _y1 -φ _y0.

The relationship between the phase differences Δφ _x , Δφ _y before and after the displacement and the displacement amounts u, v in the x direction and the y direction is as shown in Expression (4). Note that p is a lattice interval, and is determined according to the specification of the measurement tape 1 to be used and the size of the captured lattice.

Here, as shown in the equation (5), the obtained displacement distributions u and v are differentiated in the respective directions to obtain the distributions of strains ε _x and ε _{y in} the x direction and the y direction.

Furthermore, the stress distribution can be obtained by multiplying Equation (5) by a material constant such as Young's modulus.
FIG. 5 is a diagram for explaining the principle of measuring the displacement stress distribution. The left side shows an image, and the right side shows cross-sectional data of one horizontal line. First, FIG. 5A shows respective lattice images before and after displacement. It is slightly displaced to the right compared to before deformation. FIG. 5B shows four phase-shifted moire fringe images obtained by performing a thinning process when the thinning number N is 4 and linearly interpolating the missing data. Similarly, FIG. 5C shows four phase-shifted moire fringe images after displacement. FIG. 5D shows the phase distribution φ ₀ before displacement of FIG. 4B, and FIG. 5E shows the phase distribution φ ₁ after displacement. It can be seen that the phase value has changed. FIG. 5F shows the phase difference distribution Δφ before and after the displacement. By substituting this value into Equation (4) and Equation (5), a displacement distribution and a strain stress distribution are obtained.

Example 1
6 and 7 show the optical system of the experimental apparatus. Here, the deflection of a plate subjected to simple three-point bending was measured. The test piece 5 is an aluminum plate having a length of 1000 mm, a width of 30 mm, and a thickness of 2.0 mm. A measuring tape 1 on which a grid with a spacing of 2 mm was printed was pasted on the surface. This test piece 5 was placed on fulcrums 6 and 7 having a distance between fulcrums of 800 mm, and a weight 8 was placed on the central portion to apply a load. In order to know the actual amount of displacement, a laser displacement sensor 9 was installed immediately below the place where the load was applied, and the amount of displacement before and after the load was applied was measured. In addition, as shown in the plan view of FIG. 7, when it is difficult to install the camera 11 in front of the object to be measured (in this embodiment, the test piece 5), refer to the arrangement of the tilt system using the shift lens 12. An experiment may be performed by installing the camera 11 as indicated by a reference numeral 11a. If it is still difficult, shooting may be performed with the camera installed obliquely.

  FIG. 8 shows the measurement results. The left figure in Fig. 8 (a) is an untransformed image. When the image is thinned out every four pixels in the y direction and the missing data part is interpolated, the phase is shifted as shown in the left figure in Fig. 8 (b). Four moiré fringe images can be obtained. The result of obtaining the phase distribution of moire fringes before deformation from these four images is shown in the left diagram of FIG.

  Similarly, the right diagram in FIG. 8A is a deformed image. When the image is thinned out every four pixels in the y direction and the missing data portion is interpolated, the right diagram in FIG. 8B is obtained. It is possible to obtain four moire fringe images with a phase shift.

  The result of obtaining the phase distribution of the moire fringes after deformation from these four images is shown in the right figure of FIG. Each phase difference distribution obtained from the left diagram in FIG. 8C and the right diagram in FIG. 8C is shown in FIG. With respect to this phase difference distribution, the displacement distribution of FIG. 8E can be obtained from the equation (4).

  FIG. 8 shows cross-sectional data of one horizontal line in FIG. From this figure, it can be seen that the plate of the test piece 5 is bent. When measured with the laser displacement sensor 9 at the same place, there was a displacement of 299.7 μm, whereas in this method, a displacement of 301.2 μm was detected, so the error was 1.5 μm and the accuracy was high. It was confirmed that the displacement was detected. Therefore, even when applied to large structures such as actual bridges and high-rise buildings, it is considered that the displacement distribution can be obtained similarly.

(Example 2)
FIG. 10 shows an experimental apparatus. To confirm the in-plane displacement detection accuracy, a sample with a 150 mm x 150 mm reference grid (lattice spacing of 1.016 mm) is placed on a uniaxial moving stage and moved 20 times by 1 μm in the x direction. Images were taken at each position. The moving resolution of the moving stage is 0.1 μm.

  In the experiment, a CCD camera with a Sony digital interface (XCD-X700) was used to obtain an 8-bit grayscale image of 1024 pixels × 768 pixels. At this time, the position of the camera was adjusted while observing the moire fringes so that the 1.016 mm interval, which is the lattice interval, was 4 pixels of the camera. That is, at this time, the sampling interval of the CCD camera in photographing is 0.254 mm. After the thinning process and the image data interpolation process were performed on the obtained image to obtain the phase distribution at each position, the displacement amount was calculated from the phase difference distribution before and after the displacement according to Expression (4).

  FIG. 11 shows the relationship between the amount of displacement in the x direction detected by this method and the amount of displacement applied to the actual moving stage. A black circle “●” is an average value of 120 × 120 pixels in the center, and a white circle “◯” is a value at one center. From FIG. 10, when the average value was used, the average error was 0.31 μm and the maximum error was 0.71 μm with respect to the lattice having a spacing of 1.016 mm. The average error of 0.31 μm corresponds to 0.03% of the lattice spacing used, and is 0.0012 pixels when converted to the number of pixels.

  From the experimental results, it was confirmed that displacement can be detected with very high accuracy. On the other hand, the error is slightly large in the case of one point. As a cause thereof, it is conceivable that in the case of one point compared to the average value, it is easily affected by random noise of the CCD camera. In such a case, improvement in accuracy can be expected by acquiring a plurality of images continuously for the same image and performing image averaging.

(Example 3)
In the past, measurement was performed after directly applying a measuring tape to the surface of the measurement object. However, if there is a regular pattern on the surface of the measurement object, the pattern can be used well. As an example, a method for detecting shaking of a high-rise building is described.

  FIG. 12 shows a photograph of a high-rise building. Here, in the target portion (FIG. 13A) in the center frame, the pattern of the member made of the building structural material can be regarded as the reference lattice. At this time, the interval of one period of the pattern on the structure corresponds to approximately 7 pixels on the camera. Therefore, if the image thinning process is performed for every seven pixels in the x direction, seven moire fringe images having a phase shift as shown in FIG. 13B can be obtained. The phase distribution of moire fringes as shown in (c) can be obtained. When the building shakes (there is displacement), this phase distribution also changes, so that the building shake can be detected from the phase difference distribution before and after the building shake.

[Sampling moire method]
Next, a phase analysis method using the sampling moire method of the present invention will be described. The characteristics of the sampling moire method can be analyzed with a single image, both shooting and analysis can be performed at high speed, and the accuracy of 1/100 to 1/1000 of the lattice spacing can be improved. It can be realized at low cost.

(Phase analysis method of original grid image)
A method for obtaining the phase distribution of the lattice image in the x and y directions from the phase distribution of the moire pattern in the x and y directions obtained by the phase shift moire method will be described. The moiré fringes obtained by sampling every N pixels correspond to the moiré fringes obtained by superimposing the captured lattice image and the lattice image with one interval of N pixels, and the phase value is before and after the deformation. It is obtained from the difference of the phase values of the gratings. Therefore, the phase distribution of the moire pattern can be obtained from the difference between the phase distribution of the captured grid image and the phase distribution of the grid that forms one interval with N pixels.

As shown in FIG. 14, a phase distribution image in which one interval is N pixels (in this case, four pixels are taken as an example), that is, N pixels form one cycle, is used as a reference image. Assuming that the phase distribution of the reference image is θ _R , the phase distribution of the moire pattern is θ _m , and the phase distribution of the captured grating is θ _g , the relationship shown in Expression (6) is established based on the moire generation principle.
θ _m = θ _g −θ _R (6)

This equation (6) is modified to obtain the following equation (7). Thereby, the phase distribution of the grating is obtained. Even in the case of a two-dimensional grating, when a reference image is created in each of the x and y directions and is calculated with each moire pattern image, phase distributions in the x and y directions of the grating can be obtained.
θ _g = θ _m + θ _R (7)

From equation (7), if the phase values of the original grating before and after deformation are θ _g0 and θ _g1 , the displacement is θ _g1 −θ _g0 ,
θ _g1 −θ _g0 = (θ _m1 −θ _R ) − (θ _m0 −θ _R )
= Θ _m1 −θ _m0 (8)

  From the equation (8), the deformation amount of the original lattice can be obtained directly from the moire phase difference distribution as a result, and the displacement distribution can be obtained correctly even when the image is taken from an oblique direction.

(Multi-camera shape measurement method)
By using the detection results of corresponding points in the captured images of a plurality of cameras, the three-dimensional shape of the object and the strain distribution on the surface of the object can be obtained as follows.

  FIG. 15 shows the relationship between the lattice image, the phase of the lattice, and the phase of the phase-connected lattice. Serial numbers are assigned to the grids in order. The phase of the grating is obtained as a value in the range of 2π. In the case of FIG. 15, it repeats in a sawtooth waveform between −π and π. A range in which the phase continuously changes is regarded as a range of the same lattice, and phase connection can be performed for each pixel by performing an operation of adding a value obtained by multiplying the serial number of the lattice by 2π to the original phase value. Dividing the phase connected phase by 2π results in a distribution of fractionalized lattice numbers. That is, by performing this operation, the positions of the pixels and the grid can be precisely associated.

FIG. 16 schematically shows the phases of the x-direction and y-direction gratings phase-connected to the two-dimensional grating. In the case of a two-dimensional grating, when a phase distribution phase-connected in each direction is obtained, a two-dimensional phase value (φ _x , φ _y ) can be obtained for each pixel. In the case of a two-dimensional lattice, one point on the two-dimensional lattice can be specified from the two-dimensional phase values (φ _x , φ _y ).

  FIG. 17 shows an arrangement of an object to which a two-dimensional grid is attached and two cameras. FIG. 18 shows an image obtained by photographing an object to which a two-dimensional lattice is attached with two cameras. Since each camera captures a two-dimensional lattice image, the correspondence between one point on the object and the pixel in the image captured by each camera can be determined by obtaining a two-dimensional phase value for each pixel. . If the corresponding points in the images taken by the left and right cameras are known, the three-dimensional coordinates of one point on the object can be obtained by calibrating the two cameras in advance. By performing it for many points on the object, the shape of the object can be obtained.

  Further, when the measurement target object is deformed, the above-described shape measurement is performed before and after the deformation. In this shape measurement method, the three-dimensional coordinates at the corresponding points of the lattice pattern fixed on the object surface are obtained. Therefore, by calculating how the distance between two points on the object surface changes before and after the deformation, Surface strain can be calculated. Strain distribution can be obtained by calculating strain at many points.

(Projected shape measurement method)
A reference plane with z = 0 is R ₀ , and a reference plane moved by a distance z in the z direction from the reference plane is R ₁ . The xy plane is parallel to the reference plane, and the height direction of the object to be measured is the z-axis. The spatial coordinates are represented by (x, y, z). First, _R 0, a point on the _{R 1 (x 0, y 0} , 0) of any point (x, y, z) and are captured in the same pixel of the camera on the _object, (x _{1, y} 1, The phase value of z ₁ ) is obtained using a phase analysis method. Thereafter, the phase value of (x, y, z) is obtained. From the relationship when the two reference planes R ₀ and R ₁ shown in FIG. 19 are used, the following equation (9) is obtained.

  When formula (9) is arranged with respect to z, the following formula (10) is obtained.

From the equation (10), the coordinates of the point z are obtained. However, Expressions (9) and (10) hold when the distance h between the camera and the reference plane R ₀ is sufficiently large.

(Height distribution of measurement object)
A procedure for actually measuring the object to be measured shown in FIG. 19 using the above analysis principle will be described. In this experiment, the distance h between the camera and the projection surface was 960 mm, the distance between the projector and the projection surface was 650 mm, and the reference plate interval was 2.0 mm. The experimental environment is the same as in FIG. An object to be measured is a first plate having a vertical width a1 = 50 mm, a horizontal width b1 = 40 mm, a thickness t1 = 1 mm, a vertical plate a2 = 100 mm, a horizontal width b2 = 80 mm, and a second plate having a thickness t2 = 1 mm. A third plate with a3 = 150 mm, width b3 = 120 mm, thickness t3 = 1 mm, and a fourth plate with length a4 = 200 mm, width b4 = 150 mm, thickness t4 = 3 mm are stacked concentrically in this order. It is a configuration.

  FIG. 20 shows the result of photographing the projected image with a CCD camera. FIG. 21 shows a moire pattern image obtained by performing the thinning process using the image of FIG. FIG. 22 shows a phase distribution image of a moire pattern. FIG. 23 shows a height distribution image of the measurement object. FIG. 24 shows a cross-sectional shape at a height where one horizontal line of A in FIG. 23 is extracted. FIG. 25 shows a cross-sectional shape of a height obtained by extracting one vertical line of B in FIG.

  Table 1 shows an average value and standard deviation when one horizontal line A is extracted, and Table 2 shows an error obtained from an actual value obtained from Table 1 and a measured height. Table 2 shows the average value and standard deviation when one vertical line of B is extracted. In addition, the actual measurement value was measured with a height gauge. Both the average error and the standard deviation were less than 0.1 mm and could be measured with high accuracy.

It is a perspective view which shows the external appearance of the measuring tape 1 used with the deformation | transformation measuring method of the object of one Embodiment of this invention. It is a figure which shows the lattice pattern 2 displayed on the measuring tape 1 shown in FIG. 1, FIG. 2 (a) shows the black-and-white binary lattice of a vertical direction, FIG.2 (b) shows the black-and-white binary lattice of a horizontal direction. 2 (c) shows a black-and-white binary grid in the vertical direction, FIG. 2 (d) shows a grid with a brightness change in the horizontal sine waveform or cosine waveform, and FIG. FIG. 2 (f) shows a black and white square lattice (intensity inversion) in the vertical and horizontal directions, FIG. 2 (g) shows a black and white dot lattice in the vertical and horizontal directions, and FIG. 2 (h) shows a vertical and horizontal direction. FIG. 2 (i) shows a grid having a brightness change of a sine waveform or cosine waveform in the vertical and horizontal directions. FIGS. 3A and 3B are diagrams for explaining the principle of the phase shift moire method, in which FIG. 3A shows the center point of the CCD camera sampling by pasting the reference grating on the surface of the object, and FIG. 3B shows the reference grating. FIG. 3C shows an image obtained by photographing the reference grid of FIG. FIG. 3C shows gray (grayscale) data d1, d2,..., Which are detected by using the amount of light incident on the square area of the image pitch as an integral value. g) shows an image when the thinning process is performed every four pixels, and FIGS. 3 (h) to 3 (k) show moire fringes obtained by duplicating the previous data to the luminance values of three pixels without data. Show. It is a figure for demonstrating the image processing procedure for obtaining the several moire fringe image by which the phase shift was carried out, and shows the case where the thinning number is set to 3. 3A shows the original luminance value of the photographed image, FIG. 3B shows the luminance value obtained by thinning out every third pixel from the first start point from the left, and FIG. c) shows a luminance value obtained by thinning out every third pixel from the second start point from the left, and FIG. 3D was obtained by thinning out every third pixel from the third start point from the left. Indicates the luminance value. FIG. 5A is a diagram showing the principle of displacement stress distribution measurement, FIG. 5A shows a lattice image before and after displacement, FIG. 5B shows a moire fringe image before displacement, and FIG. FIG. 5D shows a phase distribution before displacement, FIG. 5E shows a phase distribution after displacement, and FIG. 5F shows a phase difference distribution before and after displacement. It is a side view which shows the optical layout of the deformation distribution measurement of an aluminum plate. It is a top view which shows the optical layout of the deformation distribution measurement of an aluminum plate. FIGS. 8A and 8B show measurement results of the displacement distribution in FIG. 4, FIG. 8A shows an image before deformation, FIG. 8B shows an image after deformation, and FIG. 8C shows a phase before deformation. FIG. 8D shows a phase-shifted moire fringe image after deformation, FIG. 8E shows a phase distribution of moire fringes before deformation, and FIG. 8F shows a shifted moire fringe image. Shows the phase distribution of the moire fringes after deformation, FIG. 8 (g) shows the phase difference distribution between the phase distribution before deformation and the phase distribution after deformation of FIG. 8 (c), and FIG. 8 (e) shows the displacement. Show the distribution. It is a figure which shows the bending shape of the aluminum plate in the line B of FIG. It is a figure which shows the photograph of the optical system and apparatus of a confirmation experiment of the detection accuracy of a displacement. It is a figure which shows the accuracy comparison of an experimental result and an actual displacement. It is a photograph of a high-rise building. FIG. 13A shows the analysis result of the central region in FIG. 10, FIG. 13A shows the captured image, FIG. 13B shows the moire fringe image obtained by thinning out every 7 pixels, and FIG. The phase distribution obtained from seven phase-shifted images of 13 (b) is shown. It is a figure which shows the reference image of the phase distribution in which 1 space | interval comprises 1 pixel and 1 period. It is a figure which shows the relationship between the grating | lattice image, the phase of a grating | lattice, and the phase of the phase-connected grating | lattice. It is a figure which shows typically the phase of the grating | lattice of the x direction and y direction phase-connected with the two-dimensional grating | lattice. It is a figure which shows arrangement | positioning of the object on which the two-dimensional lattice was stuck, and two cameras. It is a figure which shows the image which image | photographed the object with which the two-dimensional lattice was stuck with two cameras. It is a figure which shows the shape of a measuring object. It is a figure which shows the result of having image | photographed the projection image with the CCD camera.

The moire pattern image obtained by performing the thinning process using the image of FIG. 20 is shown. The phase distribution image of a moiré pattern is shown. The height distribution image of a measurement object is shown. The phase value which extracted one horizontal line of A of FIG. 23 is shown. The phase value which extracted one vertical line of B of FIG. 23 is shown.

Explanation of symbols

1 Measuring Tape 2 Grid Pattern 3 Object Surface

Claims (5)

A regular pattern is formed by attaching a two-dimensional sheet body on which a lattice pattern having a rectangular waveform or a sinusoidal waveform brightness distribution with a pitch interval corresponding to the accuracy to be measured is applied to the surface of the object, and the regular pattern pattern as the reference grid and the respective predetermined regions before and after the application of force, a step of photographing by a plurality of optical cameras installed at different positions,
A moiré fringe image is obtained by performing sampling for each pixel at equal intervals on a photographed image a plurality of times, three or more times while changing the starting pixel, and interpolating the thinned image obtained by this sampling processing. And obtaining each phase distribution in the horizontal and vertical directions of the moire fringes obtained by the phase shift method;
Calculating the displacement distribution of the object from the phase difference distribution obtained from the phase distribution before and after deformation ;
And calculating a strain or stress distribution based on the displacement amount distribution . The object displacement measuring method according to claim 1 , wherein the displacement of the object is detected three-dimensionally. The object displacement measuring method according to claim 2 , wherein corresponding points in the captured images of the plurality of cameras are detected using a two-dimensional phase distribution. 4. The object displacement measuring method according to claim 3 , wherein a three-dimensional shape of the object is measured using detection results of corresponding points in the captured images of the plurality of cameras. 5. The object displacement measuring method according to claim 4 , wherein a strain distribution on the surface of the object is obtained using the measured three-dimensional shape of the object.