JP2015165206A - Stress distribution measurement device and stress distribution measurement method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a stress distribution measurement device capable of removing noise even under a condition where the repeated loading is not executed and lock-in processing cannot be applied.SOLUTION: A stress distribution measurement device 1 for estimating stress distribution of an object to be measured, includes: a projector 2 for projecting a predetermined pattern onto the object to be measured; imaging means for imaging the object to be measured on which the pattern is projected; an infrared camera 3 for obtaining temperature change images of the object to be measured when no load is applied and a load is applied; three-dimensional shape calculation means for calculating a three-dimensional shape of the object to be measured from an image of the object to be measured on which the pattern is projected, imaged by the imaging means; noise removal means for removing noise of the reference temperature image and the temperature distribution image, on the basis of height distribution of the object to be measured, obtained from acquired information of the three-dimensional shape of the object to be measured; and stress calculation means for calculating the stress distribution of the object to be measured from the difference between the reference temperature image and the temperature distribution image from which noise is removed by the noise removal means.

Description

  The present invention relates to a stress distribution measuring device and a stress distribution measuring method. More particularly, the present invention relates to a stress distribution measuring apparatus and a stress distribution measuring method for estimating a stress distribution generated in an object from a change in temperature distribution of the object subjected to a load.

  In order to optimally design a member, it is important to grasp what kind of stress is generated in the member. For this purpose, for example, a method for obtaining a stress distribution by numerical analysis such as a finite element method (FEM) or a boundary element method (BEM) and a method for measuring stress applied to a member using a strain gauge are known.

However, when the shape of the member is complicated or when the member is composed of a plurality of parts, the stress may not be accurately obtained by the above-described method.
Therefore, a method (for example, infrared stress measurement method) has been proposed in which the temperature distribution of a member that receives a load is measured and the stress distribution of the member is estimated from the change in the temperature distribution.

  The infrared stress measurement method is a method in which a change in heat (temperature) generated in a member according to a change in stress acting on the member is measured by an infrared camera. In measurement, noise other than heat generated by stress ( For example, it is affected by frictional heat, changes in ambient temperature, and the like. As a technique for removing such noise and obtaining a highly accurate stress distribution, lock-in processing has been conventionally employed (see, for example, Patent Documents 1 and 2).

  In the lock-in process in the method described in Patent Document 1, repeated load fluctuations are generated in a member, and temperature average values are respectively obtained for a first period including at least one period of the load fluctuation and a second period thereafter. The first average temperature and the second average temperature are used. Then, a temperature gradient with respect to time is obtained based on the difference between the first average temperature and the second average temperature, a temperature characteristic is determined from the temperature gradient and the initial temperature of the member, and the actual temperature distribution measured for the member The corrected temperature distribution from which noise is removed is obtained by reducing the temperature obtained based on the temperature characteristics.

JP2012-103124A Japanese Patent Laid-Open No. 2001-188028

  However, in order to perform the lock-in process, it is necessary to repeatedly apply a load to the member from the load tester, so that the application range is limited. That is, the lock-in process cannot be applied to a member to which a repeated load does not act, and therefore a highly accurate stress distribution cannot be obtained.

  The present invention has been made in view of such circumstances, and a stress distribution measuring device capable of obtaining a highly accurate stress distribution by removing noise even under non-repetitive load conditions, to which lock-in processing cannot be applied, and The object is to provide a stress distribution measurement method.

(1) The stress distribution measuring apparatus according to the present invention estimates the stress distribution of the measurement object based on the difference between the temperature image of the measurement object when no load is applied and the temperature image of the measurement object when the load is applied. A stress distribution measuring device,
A projector that projects a predetermined pattern on the object to be measured;
Imaging means for imaging an object to be measured on which a pattern is projected;
An infrared camera that acquires a reference temperature image before applying a load to the object to be measured and a temperature fluctuation image when the load is applied;
Three-dimensional shape calculation means for calculating the three-dimensional shape of the measurement object from the image of the measurement object on which the pattern is projected, which is imaged by the imaging means;
Noise removing means for removing noise of the reference temperature image and the temperature distribution image based on the height distribution of the measured object obtained from the obtained three-dimensional shape information of the measured object;
Stress calculating means for calculating the stress distribution of the object to be measured from the difference between the reference temperature image and the temperature distribution image from which noise has been removed by the noise removing means.

  In the stress distribution measuring apparatus of the present invention, the noise of the reference temperature image and the temperature distribution image is obtained based on the height distribution obtained from the three-dimensional shape information of the measurement object obtained by projecting the pattern onto the measurement object. It has been removed. For this reason, noise can be removed and a highly accurate stress distribution can be obtained without using a lock-in process in which a load is repeatedly applied to an object to be measured.

  In the present specification, the “object to be measured” is an object for measuring the stress distribution using the thermoelastic effect, and is not particularly limited in shape or size. An aluminum housing or the like with can be exemplified.

From the three-dimensional shape information of the measurement object obtained by the three-dimensional shape calculation means, further comprising a dividing means for dividing the image area of the measurement object based on the height distribution of the measurement object,
The noise removing unit selects a pixel for noise removal, selects a peripheral region around the pixel belonging to the same image region as the pixel, and a pixel value from which noise of the selected pixel is removed from the selected peripheral region Can be executed for all pixels.

  In addition, the noise removing unit selects a pixel from which noise is to be removed, and performs weighting based on the height distribution for each of the peripheral pixels around the pixel and the selected pixel, and calculates the pixel value and weight of each pixel. The process of setting the sum of products to the pixel value from which the noise of the selected pixel has been removed can be executed for all the pixels.

(2) Further, the stress distribution measuring method of the present invention calculates the stress distribution of the measured object based on the difference between the temperature image of the measured object when no load is applied and the temperature image of the measured object when loaded. An estimated stress distribution measurement method,
An imaging step of imaging an object to be measured on which a predetermined pattern is projected from a projector;
A three-dimensional shape calculation step of calculating a three-dimensional shape of the object to be measured from an image of the object to be measured, which is imaged in the imaging step and onto which a pattern is projected;
An image acquisition step of acquiring a reference temperature image before applying a load to the object to be measured and a temperature fluctuation image when applying the load by an infrared camera;
A noise removal step of removing noise of the reference temperature image and the temperature distribution image based on the height distribution of the measurement object obtained from the three-dimensional shape information of the measurement object acquired in the three-dimensional shape calculation step When,
A stress calculating step of calculating a stress distribution of the object to be measured from a difference between the reference temperature image and the temperature distribution image from which noise has been removed in the noise removing step.

  In the stress distribution measurement method of the present invention, the noise of the reference temperature image and the temperature distribution image is calculated based on the height distribution obtained from the three-dimensional shape information of the measurement object obtained by projecting the pattern onto the measurement object. It has been removed. For this reason, noise can be removed and a highly accurate stress distribution can be obtained without using a lock-in process in which a load is repeatedly applied to an object to be measured.

From the three-dimensional shape information of the measurement object acquired in the three-dimensional shape calculation step, further includes a dividing step of dividing the image area of the measurement object based on the height distribution of the measurement object.
In the noise removal step, a pixel to be denoised is selected, a peripheral region around the pixel belonging to the same image region as the pixel is selected, and a pixel value from which noise of the selected pixel is removed from the selected peripheral region Can be executed for all pixels.

  In the noise removal step, a pixel to be denoised is selected, and a peripheral pixel around the pixel and the selected pixel are weighted based on a height distribution, and a pixel value and a weight of each pixel are determined. The process of setting the sum of products to the pixel value from which the noise of the selected pixel has been removed can be executed for all the pixels.

  According to the stress distribution measuring apparatus and the stress distribution measuring method of the present invention, it is possible to obtain a highly accurate stress distribution by removing noise even under conditions where a lock-in process cannot be applied and the load is not repeated.

It is a mimetic diagram of one embodiment of a stress distribution measuring device of the present invention. (A) is a top view of an example of the to-be-measured object in this invention, (b) is the sectional view on the AA line. It is a figure which shows the flow of the process in one Embodiment of the stress distribution measuring method of this invention. It is explanatory drawing of an example of the noise removal method.

  Hereinafter, embodiments of a stress distribution measuring device and a stress distribution measuring method of the present invention will be described in detail with reference to the accompanying drawings.

[Stress distribution measuring device]
FIG. 1 is a schematic diagram of a stress distribution measuring apparatus 1 according to an embodiment of the present invention. The stress distribution measuring apparatus 1 is based on a difference between a temperature image of the measurement object when no load is applied to the measurement object and a temperature image of the measurement object when a load is applied to the measurement object. This is a device for estimating the stress distribution of the object to be measured. The stress distribution measuring device 1 includes an infrared projector 2, an infrared camera 3, and an image processing device 4. The image processing apparatus 4 includes a calculation unit 5 (including a storage unit 5a), a display unit 6, and an input unit 7. The stress distribution measuring device 1 according to the present embodiment is a device that measures the distribution of stress generated in the measurement object 8 when a tensile force is applied to the measurement object 8. A load load portion 9 for applying force is provided.

  The infrared projector 2 is arranged toward the object 8 to be measured, and projects a predetermined pattern such as a stripe pattern on the object 8 to be measured.

  The infrared camera 3 detects infrared rays emitted from the surface of the object, converts the infrared rays into an electrical signal by the infrared sensor, and outputs it as an image signal. The infrared camera 3 is arranged toward the object 8 to be measured, captures the object 8 on which a predetermined pattern is projected, acquires a pattern projection image, and also measures the object 8 when no load is applied. And a temperature variation image that is a temperature distribution of the measurement object 8 when a load is applied by the load loading unit 9.

  The calculation unit 5 receives the image signal from the infrared camera 3 and calculates the stress distribution of the measurement object 8 based on the distribution of temperature fluctuations of the measurement object 8 obtained by processing the image signal. The display unit 6 displays the calculation result by the calculation unit 5 on the monitor.

  The calculation unit 5 in the present embodiment is a three-dimensional shape calculation unit that calculates a three-dimensional shape of the measurement object 8 from an image of the measurement object 8 on which a pattern is projected, which is captured by the infrared camera 3 that is an imaging unit. A noise removing means for removing noise from the reference temperature image and the temperature distribution image based on the height distribution of the measured object 8 obtained from the obtained three-dimensional shape information of the measured object 8, and noise removal; It functions as a stress calculation means for calculating the stress distribution of the object to be measured from the difference between the reference temperature image and the temperature distribution image from which noise has been removed by the means. The calculation unit 5 may also be a dividing unit that divides the image area of the measurement object from the three-dimensional shape information of the measurement object obtained by the three-dimensional shape calculation unit based on the height distribution of the measurement object. Function.

[Stress distribution measurement method]
Next, the procedure of the stress distribution measurement method will be described with reference to FIGS. FIG. 3 is a diagram showing a process flow in the stress distribution measurement method according to the embodiment of the present invention, and FIG. 4 is an explanatory diagram of an example of the noise removal method.

  First, in step S1, a predetermined pattern (for example, a fringe pattern) is projected from the infrared projector 2 onto the measurement object 8, and the infrared camera 3 images the measurement object 8 on which the predetermined pattern is projected. A signal of the captured image is output to the image processing device 4 and processed in the image processing device 4. This process will be described later.

  Next, in step S <b> 2, the three-dimensional shape of the measurement object 8 is calculated from the image of the measurement object 8 onto which the pattern is projected, obtained in step S <b> 1. The calculation method of the three-dimensional shape using the pattern is not particularly limited in the present invention. For example, a general method as described in JP-A-7-63527 and JP-A-7-229723 is used. It can be adopted as appropriate.

  Next, in step S <b> 3, a reference temperature image indicating the temperature distribution of the measurement object 8 in a state in which no load (tensile load) is applied by the load application unit 9 is acquired by the infrared camera 3, and thereafter, by the load application unit 9. Temperature of the measurement object 8 in a state in which a load is applied (in this embodiment, a tensile force is applied along the longitudinal direction of the measurement object 8 by the load application unit 9. See the white arrow in FIG. 1). Similarly, an infrared camera 3 obtains a temperature fluctuation image indicating the distribution. The acquired signals of both images are output to the image processing device 4 and processed in the image processing device 4. This process will also be described later.

Next, in step S4, based on the three-dimensional shape information (position information on the three-dimensional orthogonal coordinate axis) of the measurement object 8 obtained in step S2, the measurement object is based on the distribution in the height direction of the measurement object 8. The image area of the object 8 is divided by the calculation unit 5. Specifically, the measurement object 8 in the present embodiment is a rectangular plate material in which a circular recess 8a is formed at the center, as shown in FIG. If the height direction of the object to be measured 8 is the vertical direction in FIG. 2B, the height of the region of the recess 8a is different from the other region. In this embodiment, the image area of the measurement object 8 is divided based on the distribution in the height direction of the measurement object 8 because the stress distribution varies depending on the difference in height when the height is different. ing. Specifically, in the present embodiment, the calculation unit 5 divides the image area into the area of the recess 8a and the other area.
In the above example, the object to be measured is divided into two image regions for simplicity, but the object to be measured having a complicated shape can be divided into three or more image regions.

  In step S5, the arithmetic unit 5 performs noise removal processing. The surface temperature of the measurement object 8 does not change only due to the stress applied by the load application unit 9 but is affected by noise (for example, frictional heat, ambient temperature change, etc.). Therefore, the stress distribution cannot be estimated with high accuracy unless the influence of this noise is eliminated. In the present embodiment, noise is removed from the reference temperature image and the temperature fluctuation image obtained in step S2 as follows, for example.

<Noise reduction>
Noise removal is performed on the entire image acquired by the infrared camera 3. For example, in the case of the infrared camera 3 having 100 × 100 pixels, noise removal processing is performed on 10,000 pixels.

  A region where pixels (p2 to p9) that are one outside of the target pixel p1 from which noise is to be removed exists is set as a peripheral region of the target pixel p1. In the example shown in FIG. 4, there are eight pixels p2 to p9 in the peripheral region of the pixel of interest p1 (see (a) of FIG. 4).

  Now, it is assumed that the image region to which the two pixels p4 and p6 that have been subjected to double hatching belong (the image region divided by the calculation unit 5 in step S4) is different from the image region to which the other seven pixels belong (see FIG. 4 (b)). In this case, the noise removing unit determines that pixels belonging to the same image area as the target pixel p1 and located around the target pixel p1, that is, pixels p2, p3, p5, p7, p8, and p9, a total of six pixels. Selected. And the pixel value from which the noise of the pixel of interest p1 was removed is estimated using the pixel values of these six pixels. For example, the average of the six pixel values can be used as the pixel value from which the noise of the pixel of interest p1 has been removed.

  By sequentially shifting the target pixel up and down or left and right to set a new target pixel and repeating the above-described processing, it is possible to obtain pixel values from which noise has been removed for all the pixels. This noise removal process is performed for each of the reference temperature image and the temperature fluctuation image.

  Next, in step S6, the calculation unit 5 acquires the difference between the reference temperature image and the temperature fluctuation image from which noise has been removed, and in step S7, the calculation unit 5 then measures the difference from the difference acquired in step S6. The stress distribution of the object 8 is calculated. In calculating the stress, the thermoelastic coefficient of the material constituting the measurement object 8 is used. The calculated stress distribution is displayed on the monitor of the display unit 6.

[Other variations]
It should be noted that the embodiment disclosed this time is merely an example in all respects and is not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the meanings described above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

  For example, in the above-described embodiment, the image area of the object to be measured is divided by the dividing unit based on the height distribution of the object to be measured, but noise removal is performed without performing such division processing. You can also.

Now, as shown in FIG. 4A, consider a case where there are eight peripheral pixels centered on the pixel of interest p1.
In step S2, information on the three-dimensional shape of the measurement object 8 (position information on the three-dimensional orthogonal coordinate axes) is calculated.

  Next, the pixel of interest p1 and the surrounding pixels p2 to p9 are weighted. This weighting can be performed by various methods, for example, based on the spatial distance of each pixel. In the present invention, since the three-dimensional shape of the object to be measured is measured, the relative distance of each pixel is known. That is, the “spatial distance of each pixel” that is the distance between the coordinates of a certain peripheral pixel (for example, pixel p2) and the coordinates of the pixel of interest p1 is known. Then, a process (weighting) is performed in which the weight is increased if the peripheral pixel is close to the target pixel p1, and the weight is decreased if the peripheral pixel is far from the target pixel p1. The weighting method is not particularly limited in the present invention, and a conventionally known method can be appropriately employed. For example, weighting based on a Gaussian filter is used, or the distance from the pixel of interest is greatly separated. If so, the weight can be set to zero (threshold processing).

  Then, for each of the pixels p1 to p9, the luminance value of the pixel × weight is calculated, and by calculating the sum thereof, the luminance value from which noise has been removed can be obtained for the pixel of interest p1.

  In the above-described embodiment, the pattern projection image, the reference temperature image, and the temperature fluctuation image are acquired using the infrared camera. However, the pattern projection image is acquired not by the infrared camera but by a normal camera (imaging unit). can do. That is, a pattern image projected by a general projector instead of an infrared projector can be acquired by a normal camera, and a reference temperature image and a temperature fluctuation image can be acquired by an infrared camera. However, when using two cameras (normal camera and infrared camera), it is necessary to know the coordinate system of both cameras (to know the relative relationship between both cameras, Since it is necessary to grasp the coordinate system and posture by means such as camera calibration), it is desirable to make the infrared camera function as a pattern imaging means from the viewpoint of simplifying the system.

  In the above-described embodiment, the pixel of interest and the peripheral pixels are present in a total of nine pixels of 3 × 3 pixels, but the number of unit pixels at the time of noise removal is not limited to this. The target pixel and the peripheral pixels may exist within x5 pixels and a total of 25 pixels.

1: Stress distribution measuring device, 2: Infrared projector, 3: Infrared camera, 4: Image processing device, 5: Calculation unit, 6: Display unit, 7: Input unit, 8: Object to be measured, 8a: Recess, 9 : Load application section

Claims (7)

A stress distribution measuring device for estimating a stress distribution of a measurement object based on a difference between a temperature image of the measurement object when no load is applied and a temperature image of the measurement object when a load is applied,
A projector that projects a predetermined pattern on the object to be measured;
Imaging means for imaging an object to be measured on which a pattern is projected;
An infrared camera that acquires a reference temperature image before applying a load to the object to be measured and a temperature fluctuation image when the load is applied;
Three-dimensional shape calculation means for calculating the three-dimensional shape of the measurement object from the image of the measurement object on which the pattern is projected, which is imaged by the imaging means;
Noise removing means for removing noise of the reference temperature image and the temperature distribution image based on the height distribution of the measured object obtained from the obtained three-dimensional shape information of the measured object;
A stress distribution measuring apparatus comprising: a stress calculating unit that calculates a stress distribution of the object to be measured from a difference between the reference temperature image and the temperature distribution image from which noise has been removed by the noise removing unit. From the three-dimensional shape information of the measurement object obtained by the three-dimensional shape calculation means, further comprising a dividing means for dividing the image area of the measurement object based on the height distribution of the measurement object,
The noise removing unit selects a pixel for noise removal, selects a peripheral region around the pixel belonging to the same image region as the pixel, and a pixel value from which noise of the selected pixel is removed from the selected peripheral region The stress distribution measuring apparatus according to claim 1, wherein a process for estimating the stress is executed for all pixels.   The noise removing unit selects a pixel from which noise is removed, and performs weighting based on a height distribution for each of the surrounding pixels around the pixel and the selected pixel, and the product of the pixel value and the weight of each pixel. The stress distribution measuring apparatus according to claim 1, wherein a process of setting a sum as a pixel value from which noise of the selected pixel is removed is executed for all pixels.   The stress distribution measuring apparatus according to claim 1, wherein the projector is an infrared projector, and the infrared camera also functions as the imaging unit. A stress distribution measurement method for estimating a stress distribution of an object to be measured based on a difference between a temperature image of the object to be measured when no load is applied and a temperature image of the object to be measured when a load is applied,
An imaging step of imaging an object to be measured on which a predetermined pattern is projected from a projector;
A three-dimensional shape calculation step of calculating a three-dimensional shape of the measurement object from the image of the measurement object on which a pattern is projected, which is imaged in the imaging step;
An image acquisition step of acquiring a reference temperature image before applying a load to the object to be measured and a temperature fluctuation image when applying the load by an infrared camera;
A noise removal step of removing noise of the reference temperature image and the temperature distribution image based on the height distribution of the measurement object obtained from the three-dimensional shape information of the measurement object acquired in the three-dimensional shape calculation step When,
A stress distribution measurement method comprising: a stress calculation step of calculating a stress distribution of the object to be measured from a difference between the reference temperature image and the temperature distribution image from which noise has been removed in the noise removal step. From the three-dimensional shape information of the measurement object acquired in the three-dimensional shape calculation step, further includes a dividing step of dividing the image area of the measurement object based on the height distribution of the measurement object.
In the noise removal step, a pixel to be denoised is selected, a peripheral region around the pixel belonging to the same image region as the pixel is selected, and a pixel value from which noise of the selected pixel is removed from the selected peripheral region The stress distribution measuring method according to claim 5, wherein the process for estimating the stress is executed for all pixels. In the noise removal step, a pixel to be denoised is selected, and the surrounding pixels around the pixel and the selected pixel are weighted based on the height distribution, and the product of the pixel value and the weight of each pixel is calculated. The stress distribution measuring method according to claim 5, wherein a process of setting a sum as a pixel value from which noise of the selected pixel is removed is executed for all pixels.

Images