JP6019886B2 - Shape measuring device - Google Patents

Description

  The present invention relates to a shape measuring apparatus using a phase shift method.

  2. Description of the Related Art Conventionally, a shape measuring apparatus that measures a three-dimensional shape of a surface to be measured using a phase shift method is known. In the phase shift method, a linear lattice pattern is projected onto the surface to be measured. On the surface to be measured, a deformed lattice image corresponding to the shape is generated. The deformed grid image is picked up from a direction different from the projection axis. By analyzing the degree of deformation of the captured deformed lattice image, the three-dimensional shape of the surface to be measured is calculated.

  Specifically, when the lattice fringe image is projected onto the surface to be measured, the phase of the lattice fringe image is changed. Further, a lattice fringe image changing in phase on the surface to be measured is picked up, and a change in luminance corresponding to the phase change in the image pickup means is measured. Based on the measured luminance change, the phase component φ of the luminance of the reflected light incident on a predetermined position on the imaging surface is obtained.

  It is known that the distance y from the predetermined reference surface of the reflection point (measurement point) of the reflected light on the surface to be measured and the phase φ are in a correspondence relationship. Therefore, the distance y of the measurement point can be obtained by obtaining the phase φ.

  Further, it is known that the accuracy of the calculation of the distance y increases as the spatial period of the lattice pattern to be imaged is shorter (the spatial frequency is higher). Therefore, in Patent Documents 1 to 4, when obtaining the distance y of the measurement points, the distance y of the measurement points obtained by the lattice fringes having a long spatial period is corrected by using the lattice fringes having a short spatial period.

JP-A-6-66527 JP 2001-159510 A JP 2010-203867 A JP 2010-151842 A

  By the way, a light projecting / receiving optical system that projects a lattice fringe on a surface to be measured and picks up an image thereof has a frequency transmission characteristic (MTF, Modulation Transfer Function) indicating a light transmission characteristic corresponding to the spatial period of the lattice fringe. . Due to the frequency transmission characteristics, the shorter the spatial period of the lattice fringes, the lower the transparency. Then, as illustrated in FIG. 19, when a plurality of lattice fringes having different spatial periods are used, the light reception luminance amplitude on the imaging surface becomes smaller as the spatial period of the lattice fringes becomes shorter. If the light reception luminance amplitude on the imaging surface becomes small, it becomes difficult to detect a change in the light reception luminance, and it becomes difficult to calculate the phase φ. Accordingly, an object of the present invention is to provide a shape measuring apparatus capable of suppressing a decrease in received light amplitude due to frequency transmission characteristics of an optical system.

  The present invention relates to a shape measuring apparatus. The shape measuring apparatus projects a first checkered image having a first spatial period and a first projected luminance amplitude onto a surface to be measured, a first projecting means, a second spatial period and a second projected luminance amplitude. (2) Second projection means for projecting a lattice pattern image onto the surface to be measured; phase displacement means capable of changing the phase of the lattice pattern image projected onto the surface to be measured; and grid pattern projected onto the surface to be measured An imaging unit capable of capturing an image; and an arithmetic unit that calculates a shape of a surface to be measured based on a change in received light luminance of the imaging unit corresponding to the change in phase. The first spatial period is shorter than the second spatial period, and the first projection luminance amplitude is larger than the second projection luminance amplitude.

  In the above invention, when the first checkered image projected on the surface to be measured is picked up, the light reception luminance amplitude on the image pickup surface of the imaging means is the second checkered image projected on the surface to be measured. It is preferable to control the first projection luminance amplitude and the second projection luminance amplitude so as to be substantially equal to the light reception luminance amplitude on the imaging surface when imaging.

  In the above invention, when the first checkered image projected on the surface to be measured is picked up, the received light intensity on the image pickup surface of the imaging means and the second checkered image projected on the surface to be measured are It is preferable to control the first projection luminance amplitude and the second projection luminance amplitude so that the received light luminance amplitude on the imaging surface when imaging is less than the luminance saturation level.

  In the above invention, it is preferable to control the first projection luminance amplitude and the second projection luminance amplitude while maintaining a ratio between the first projection luminance amplitude and the second projection luminance amplitude.

  In the above invention, the calculation unit calculates the shape of the surface to be measured based on a region of the imaging surface of the imaging unit in which the received light luminance falls within a predetermined range, and calculates the first projection luminance amplitude and the first 2 It is preferable to change the region where the received light luminance falls within the predetermined range by changing the projection luminance amplitude.

  In addition, the shape measuring apparatus according to the present invention projects a first checkered image having a first spatial period onto a surface to be measured, first projecting means, and a second checkered image having a second spatial period to be measured. The second projection means that projects onto the surface, the phase displacement means that can change the phase of the lattice pattern image projected onto the surface to be measured, and the image of the lattice pattern projected onto the surface to be measured can be captured The imaging means, the first exposure adjustment means for controlling the exposure time to the first exposure time when the imaging means images the first lattice fringe image projected on the measurement surface, and the measurement surface. A first exposure adjusting unit that controls an exposure time to a second exposure time when the imaged unit captures the projected second checkered image; and a light receiving luminance of the imaging unit corresponding to the phase change Calculate the shape of the surface to be measured based on the change in It includes a calculation unit. The first spatial period is shorter than the second spatial period, and the first exposure time is longer than the second exposure time.

  In the above invention, when the first checkered image projected on the surface to be measured is picked up, the first light-receiving luminance amplitude on the image pickup surface of the imaging means is the second checkered pattern projected on the surface to be measured. It is preferable to control the first exposure time and the second exposure time so as to be approximately equal to the second light receiving luminance amplitude on the imaging surface when an image is captured.

  In the above invention, when the first checkered image projected on the surface to be measured is picked up, the first light reception luminance amplitude on the image pickup surface of the imaging means and the second checkered pattern projected on the surface to be measured. It is preferable to control the first exposure time and the second exposure time so that the second light reception luminance amplitude on the imaging surface when an image is captured is less than the luminance saturation level.

  Further, in the above invention, the calculation unit calculates a shape of the measurement target surface based on a region of the imaging surface of the imaging unit where the received light luminance falls within a predetermined range, and the first exposure time and the second exposure time are calculated. It is preferable to change an area in which the received light luminance falls within the predetermined range by changing an exposure time.

  In addition, the shape measuring apparatus according to the present invention projects a first checkered image having a first spatial period onto a surface to be measured, first projecting means, and a second checkered image having a second spatial period to be measured. The second projection means that projects onto the surface, the phase displacement means that can change the phase of the lattice pattern image projected onto the surface to be measured, and the image of the lattice pattern projected onto the surface to be measured can be captured The first amplification for amplifying the received light amplitude on the imaging surface of the imaging means by a first amplification factor when the imaging means and the first lattice fringe image projected on the measurement surface are imaged by the imaging means. And a second amplifying unit that amplifies the light reception luminance amplitude on the imaging surface of the imaging unit by a second amplification factor when the imaging unit images the second lattice pattern image projected on the surface to be measured; Corresponding to the phase change, after amplification of the imaging means Based on the change in light intensity, and calculates the shape of the surface to be measured, it comprises a calculation unit, a. The first spatial period is shorter than the second spatial period, and the first amplification factor is greater than the second amplification factor.

  In addition, the shape measuring apparatus according to the present invention projects a first checkered image having a first spatial period onto a surface to be measured, first projecting means, and a second checkered image having a second spatial period to be measured. The second projection means that projects onto the surface, the phase displacement means that can change the phase of the lattice pattern image projected onto the surface to be measured, and the image of the lattice pattern projected onto the surface to be measured can be captured An imaging unit, and a calculation unit that calculates a shape of the surface to be measured based on a change in received light luminance of the imaging unit corresponding to the change in phase. The first spatial period is shorter than the second spatial period, and when the first checkered image projected on the surface to be measured is captured, the light reception luminance amplitude on the imaging surface of the imaging unit is the surface to be measured Is substantially equal to the light reception luminance amplitude on the imaging surface when the second grid pattern image projected onto the image is captured.

  According to the present invention, it is possible to suppress a decrease in received light luminance amplitude due to frequency transmission characteristics of an optical system.

It is a block diagram which shows the structure of the shape measuring apparatus in embodiment of this invention. It is a perspective view explaining the principle of the shape measurement using a phase shift method. It is a figure explaining the principle of the shape measurement using a phase shift method. It is a figure explaining how to obtain | require an absolute phase. It is a figure explaining how to obtain | require an absolute phase. It is a figure explaining the calibration of a projection luminance component. It is a figure explaining the calibration of a projection luminance component. It is a figure explaining the calibration for calculating | requiring a function y (i, j) = f ((phi) (i, j)). It is a figure explaining the calibration for calculating | requiring the function x (i, j) = f (y (i, j)) and the function z (i, j) = f (y (i, j)). It is a block diagram which shows the structure of another example of the shape measuring apparatus in embodiment of this invention. It is a block diagram which shows the structure of another example of the shape measuring apparatus in embodiment of this invention. It is a figure explaining the example of a measurement by the shape measuring apparatus in embodiment of this invention. It is a figure explaining the example of a measurement by the shape measuring apparatus in embodiment of this invention. It is a figure explaining the example of a measurement by another example of the shape measuring device in an embodiment of the invention. It is a figure explaining the example of a measurement by another example of the shape measuring device in an embodiment of the invention. It is a figure explaining the example of a measurement by another example of the shape measuring device in an embodiment of the invention. It is a figure explaining the example of a measurement by another example of the shape measuring device in an embodiment of the invention. It is a figure explaining the example of a measurement by another example of the shape measuring device in an embodiment of the invention. It is a figure explaining the frequency transmission characteristic of a light projection / reception optical system.

  FIG. 1 illustrates a shape measuring apparatus 10 according to the present embodiment. The shape measuring apparatus 10 includes a light source 12, an input signal adjustment unit 14, a phase displacement unit 16, an imaging unit 26, and a calculation unit 20. In the following description, the depth direction (distance direction) of the measurement object 50 is the y axis, the width direction is the x axis, and the height direction is the z axis. Further, the plane coordinates in the lattice pattern on the light source 12 side are represented by the u axis and the v axis. Also, the coordinates of the imaging surface 18 of the imaging means 26 are represented by the i axis and the j axis.

  The light source 12 projects a lattice fringe image on the measurement target surface 22 of the measurement object 50. The lattice fringes projected by the light source 12 may have different spatial periods (spatial periods) of the respective lattice fringes, and the light source 12 alternately places a plurality of lattice fringes having different spatial periods on the measurement surface 22. It may be a projection. Here, the light source 12 may be a projector that projects an image of a checkered pattern using a liquid crystal element or a DMD (Digital Mirror Device). Further, the light source 12 may be an illumination member such as a halogen lamp, and an actual lattice having a lattice pattern drawn on a glass plate or the like may be disposed on the optical path between the illumination member and the surface to be measured 22. .

  When a projector is used as the light source 12, a checker pattern creating unit 23 that creates a checkered image may be provided. The checkerboard creating means 23 may be incorporated in an arithmetic device 24 such as a computer that transmits an input signal to the projector.

  Further, regardless of whether the image is a lattice fringe image or a real lattice, the lattice fringe projected on the measured surface 22 has a pattern having a sinusoidal intensity distribution regardless of the spatial period. It is preferable that the lattice pattern is projected onto the screen. For example, it may be a lattice pattern having a cos wave-like transmittance distribution.

  The input signal adjusting unit 14 is an adjusting unit that adjusts an input signal to the light source 12. The input signal adjustment unit 14 may be incorporated in the arithmetic device 24. When the input signal adjusting means 14 projects a lattice fringe having a relatively short spatial period onto the surface to be measured 22, the luminance component of the input signal to the light source 12 has a relatively long spatial period. The input signal is adjusted so as to be larger than the luminance component at the time of projection. When the light source 12 is a projector, an input signal including a luminance component corresponding to the spatial period of the checkered image transmitted to the projector may be transmitted to the light source 12 together with the transmission of the checkered image data. Here, the luminance component of the input signal may be a projection luminance amplitude of the light source 12. The adjustment of the projection luminance amplitude may be performed by increasing or decreasing the luminance of the light source 12, or may be performed by increasing or decreasing the irradiation time per unit time of the light source 12. In the case of using a real grid, the input signal adjusting unit 14 adjusts the input signal to the light source 12 so that the luminance component according to the spatial period of the replaced real grid is obtained when the real grid is replaced. You may do it.

  The light source 12, the input signal adjustment unit 14, the lattice fringe creation unit 23, or the actual lattice may be collectively regarded as a projection unit. This projection means may selectively project a plurality of checkered images having different spatial frequencies and projection luminance amplitudes onto the surface to be measured 22. In addition, a projection unit may be provided independently for each checkered image. Further, as will be described later, the projection unit selectively selects the first checkered image having a predetermined spatial period and the second checkered image having a spatial period longer than that of the first checkered pattern 22 to be measured 22. , The luminance component when the first lattice fringe is projected onto the surface to be measured 22 can be greater than the luminance component when the second lattice fringe is projected onto the surface to be measured 22.

  The phase displacement unit 16 changes the phase of the lattice fringe image projected on the measurement surface 22. In the case of projecting a lattice fringe image with a projector, the phase displacement unit 16 may be incorporated in the arithmetic unit 24, and the lattice fringe image with the phase changed may be transmitted to the projector. Further, when the lattice fringes are real lattices, the phase displacement means 16 may be stage means for moving the real lattice.

  The imaging unit 26 captures a lattice fringe image projected on the measurement surface 22. The imaging means 26 may be a CCD camera or a CMOS camera. Further, the imaging means 26 may include an imaging surface 18. On the imaging surface 18, a plurality of imaging elements may be arranged (two-dimensionally) in the row direction and the column direction.

  The computing unit 20 calculates the shape of the surface to be measured 22 based on the light reception luminance received by the imaging unit 26. Specifically, the calculation unit 20 calculates the shape of the measurement target surface 22 based on each change in received light luminance of the imaging unit 26 corresponding to each change in phase of lattice fringes having different spatial periods. The calculation unit 20 may be incorporated in the calculation device 24. The computing unit 20 may include a relative phase calculation unit 28, an absolute phase calculation unit 30, a phase-y coordinate calculation unit 32, and a y coordinate-xz coordinate calculation unit 34 as subsystems. These actions will be described later.

Next, the principle of shape measurement using the phase shift method will be described. As shown in FIG. 2, the light source 12 projects an image of the checkered pattern 36 toward the measurement surface 22. At this time, the transmittance P _k (u, v) at an arbitrary coordinate (u, v) on the lattice fringe 36 with respect to the lattice fringe 36 on the light source 12 side can be expressed as Equation 1 below.

  Here, A is the offset, and B is the projection amplitude of the lattice fringes 36. Also, λ is the spatial period of the lattice fringes 36. Further, k is an arbitrary real number and is a value representing the displacement of the checkered pattern 36. For example, when the phase of the grating pattern 36 is shifted by π / 2 from the initial phase (k = 0), k = 1.

The luminance I _k (i, j) of the reflected light from the surface to be measured 22 received by the imaging device at an arbitrary coordinate (i, j) on the imaging surface 18 according to Equation 1 is as shown in the following (Equation 2). Can be expressed as

Here, A (i, j) and B (i, j) are values determined by ambient light and the reflectance of the object to be measured. φ (i, j) represents the phase of the reflected light at the coordinates (i, j).

Assuming that k in Equation 1 is 1, 2, and 3, as shown in FIG. 3, the phase of the lattice fringes 36 is sequentially shifted by π / 2, π, and 3 / 2π from the initial phase (k = 0). At this time, the luminances I ₀ (i, j), I ₁ (i, j), I ₂ (i, j), I ₃ (i, j) of the image sensor at the coordinates (i, j) including the initial phase are included. ) Can be derived from Equation 2 below.

  From Equation 3, the phase φ (i, j) can be obtained. Here, the distance y between the reflection point (measurement point) on the measured surface 22 of the reflected light received by the imaging device at coordinates (i, j) and an arbitrary reference surface, and the phase φ ( i, j) are known to correspond one-to-one. That is, y (i, j) = f (φ (i, j)). Therefore, the relationship between the reference plane y and the phase φ is obtained in advance using a standard sample or the like, and the function y (i, j) = f (φ (i, j)) is obtained. According to the phase φ (i, j) obtained in step 3, the distance y (i, j) of the measurement point can be obtained.

  As shown in FIG. 4, the phase φ (i, j) on the line of sight from the image sensor at the coordinates (i, j) repeatedly changes beyond 2π when the distance y changes greatly. For example, if the phase changes by 2nπ in the measurement region in the y direction, there are n y coordinate candidates. FIG. 4 illustrates a state where five y-coordinate candidates having the phase φ are generated in the measurement region in the y direction.

In order to narrow down y-coordinate candidates, lattice fringes 36 having different spatial periods may be used. First, as shown in FIG. 5, the phase φ _a ′ (i, j) is obtained using a lattice fringe 36A (coarse lattice fringe) having a relatively long spatial period (low spatial frequency). Here, so that only one phase is calculated in the measurement region in the y direction, in other words, in the measurement region in the y direction, the spatial period is such that the possible range of the phase is 0 to 2π. It is preferable to use the lattice stripe 36A. At this time, in the y direction of the measurement region, one phase φ _a '(i, j) corresponds to one of the y-coordinate y _a' (i, j) can be obtained.

Next, the phase φ _b ′ (i, j) is obtained using the fine lattice fringes 36B having a shorter spatial period than the coarse lattice fringes 36A. At this time, a plurality of y coordinate candidate values y _bk ′ (i, j) corresponding to the phase φ _b ′ (i, j) can be obtained within the measurement region in the y direction. Further, the y-coordinate y _b ′ (i, j) closest to the y-coordinate y _a ′ (i, j) calculated using the coarse grid pattern 36A is extracted from the plurality of candidate values y _bk ′ (i, j). To do. For example, the y coordinate y _b ′ (i, j) may be extracted by calculating the one having the smallest difference ε _b from y _a ′ (i, j).

Next, the phase φ _c ′ (i, j) is obtained using the fine lattice fringe 36C having a shorter spatial period than the lattice fringe 36B. At this time, a plurality of y coordinate candidate values y _ck ′ (i, j) corresponding to the phase φ _c ′ (i, j) can be obtained within the measurement region in the y direction. Further, the y coordinate y _c ′ (i, j) closest to the y coordinate y _b ′ (i, j) calculated using the fine grid stripe 36B is extracted from the plurality of candidate values y _ck ′ (i, j). To do. The extracted y _c ′ (i, j) is set as the y coordinate of the measurement point (y _c ′ (i, j) = y (i, j)). For example, 'by calculating what (i, j) the difference epsilon _c is the smallest and, y-coordinate _{y c' y b (i,} j) may be extracted.

  Further, the x coordinate x (i, j) and the z coordinate z (i, j) may be obtained from the obtained y coordinate y (i, j). For reasons such as the aperture angle of the image pickup means 26 being narrowed, the incident angle of the reflected light incident on the image pickup device is predetermined. Therefore, if the y coordinate on the optical path of the reflected light incident on the image sensor is specified, the x coordinate and the z coordinate corresponding to the y coordinate can also be obtained. For example, the y coordinate and the corresponding x coordinate and z coordinate may be obtained in advance using a standard sample or the like. Further, the functions x (i, j) = f (y (i, j)) and z (i, j) = f (y (i, j)) representing the relationship between the y coordinate, the x coordinate, and the z coordinate are expressed. You may ask for it. From these functions and the y-coordinate y (i, j) extracted above, the x-coordinate x (i, j) and z-coordinate z (i, j) of the measurement point can be extracted.

  Next, the flow of shape measurement based on the above principle will be described. The shape measurement can be divided into two types of processes: (1) calibration process and (2) actual measurement process.

  First, the (1) calibration process in the previous stage of measuring the surface to be measured will be described. The calibration process includes (1-1) calibration of the projection luminance component, (1-2) calibration for obtaining the function y (i, j) = f (φ (i, j)), and (1- 3) The function x (i, j) = f (y (i, j)) and the calibration for obtaining z (i, j) = f (y (i, j)) may be included. .

(1-1) Calibration of Projected Luminance Component In using the lattice fringes 36A to 36C having different spatial periods, the luminance component (projected luminance component) of the light source 12 is adjusted. As shown in FIG. 6, imaging by the imaging unit 26 is performed according to the spatial period of the lattice fringe 36 by the frequency transmission characteristic (MTF) of the light projecting / receiving optical system including the light source 12, the imaging unit 26, the arithmetic unit 24, and the like. The luminance amplitude (light reception luminance amplitude) on the surface 18 changes. Specifically, as illustrated in the upper part of FIG. 6, the light reception luminance amplitude on the imaging surface 18 decreases as the spatial period of the lattice pattern 36 becomes shorter. In order to suppress a decrease in the light reception luminance amplitude on the imaging surface 18, the projection luminance component of the light source 12 is adjusted in advance according to the spatial period of the lattice fringes prior to the measurement of the measurement surface 22.

  First, from the light source 12, lattice fringes having different spatial periods are alternately projected onto a standard plate such as a ceramic plate. A lattice fringe image projected on a flat plate and whose phase is changed is picked up by the image pickup means 26. The relationship between the light reception luminance amplitude on the imaging surface 18 and the spatial period of the imaged lattice fringes is obtained by actual measurement.

_Assuming that the spatial period of the coarse grating fringe 36A is λ _L , the spatial period of the fine grating fringe 36B is λ _M , and the spatial period of the fine grating fringe 36C is λ _H , the spatial period of the grating fringe 36 is λ _L , λ _M , And λ _H , respectively, the received light intensity amplitudes R _L , R _M , and R _H are obtained. Here, as shown in the upper part of FIG. 6, the received light luminance amplitudes R _L , R _M , and R _H are all relative values, and are all set to 1 or less.

From the relationship between the spatial periods λ _L , λ _M , and λ _H and the received light luminance amplitudes R _L , R _M , and R _H , the input signal adjusting unit 14 uses the projection luminance component of the light source 12 corresponding to each of the lattice fringes 36A to 36C. Adjust. Here, when adjusting the projection brightness of the light source 12, it is preferable that the brightness of the light source is excessively amplified and does not exceed the brightness saturation level of the image sensor.

  Furthermore, when each image of the lattice fringes 36A to 36C is picked up, the sensitivity is good in any of the lattice fringes 36A to 36C as long as the light reception luminance amplitude is maximum within a range in which the luminance level is not saturated. Measurements can be made. Therefore, it is preferable to adjust the projection luminance component of the light source 12 so that the light reception luminance amplitude for each of the lattice fringes 36A to 36C is the same and the maximum value within a range not exceeding the saturation level. is there.

  First, in order to suppress the light reception luminance to a luminance saturation level or less, a state in which the light reception luminance amplitude on the imaging surface 18 is maximized is set, and the projection luminance of the light source 12 is adjusted. For example, the standard plate is disposed closest to the imaging means 26 side in the measurement region in the y direction, and the reflected light that is specularly reflected from the standard plate is received by the imaging surface 18.

First, the input signal adjusting unit 14 determines the amplitude setting value (projection) when projecting the lattice fringe image from the relationship between the spatial periods λ _L , λ _M , and λ _H and the received light luminance amplitudes R _L , R _M , and R _H. Luminance amplitude) ST is determined. Specifically, as shown in the lower part of FIG. 6, projecting the projection luminance amplitude ST _L of spatial period lambda _L R _H, the projection luminance amplitude ST _M of spatial period lambda _M of R _H / R _M, spatial period lambda _H and 1 luminance amplitude ST _H. At this time, the light reception luminance amplitude on the imaging surface 18 when the lattice fringe image of each period is projected on the surface to be measured is illustrated in the upper part of FIG. In any spatial period, the received light amplitude is equal to _RH .

Further, the input signal adjusting unit 14 amplifies the projection luminance amplitude so that the received light luminance becomes a maximum value within a range not exceeding the saturation level. Specifically, as shown in the lower part of FIG. 7, the projection luminance amplitudes ST _L , ST _M , and ST _H are multiplied by a coefficient A (> 1). In setting the coefficient A, it is preferable to adjust by actual measurement or the like so that the luminance level of the image sensor is not saturated. The projection luminance amplitudes ST _L ′ = A · ST _L , ST _M ′ = A · ST _M , and ST _H ′ = A · ST _H multiplied by the coefficient A are respectively stored in a storage unit (not shown) of the arithmetic unit 24. .

(1-2) Calibration for obtaining function y (i, j) = f (φ (i, j)) Next, the relationship between the y coordinate y (i, j) and φ (i, j) Ask. As shown in FIG. 8, a standard plate 40 such as a ceramic flat plate is disposed on the y stage 42. The y stage 42 can move the standard plate 40 on the y axis. The standard plate 40 is moved by the y stage 42 over the entire measurement area in the y direction or an arbitrary range. Each time the y coordinate changes, a grid pattern is projected from the light source 12 onto the standard plate 40. During this projection, the phase of the lattice pattern is changed. Furthermore, the imaged means 26 is made to image the projected lattice pattern image. The phase φ is obtained based on the change in the light receiving luminance on the imaging surface 18 accompanying the change in the phase of the lattice fringe image. Since the y coordinate of the standard plate at this time is known from the position setting value of the y stage 42, the phase φ and the coordinate y are associated with each other. This association is performed with respect to all coordinates (all imaging elements) on the imaging surface 18 or arbitrary coordinates. Further, the correlation between the phase φ and the coordinate y is performed over the entire measurement area in the y direction or over an arbitrary range. The obtained correspondence relationship between the phase φ and the coordinate y may be approximated to the function y (i, j) = f (φ (i, j)) and stored in a storage unit (not shown) of the arithmetic unit 24. The function y may be calculated for each lattice stripes 36 a - 36 c, corresponding to each lattice stripes, function _{y a (i, j) =} f a (φ (i, j)), y b (i, _{j) = f b (φ (} i, j)), and _{y c (i, j) =} f c (φ (i, j)) may be calculated and stored in the storage unit. Alternatively, the relationship between the phase φ and the coordinate y in these functions may be stored in the storage unit in the form of a lookup table or the like.

(1-3) Calibration for obtaining functions x (i, j) = f (y (i, j)) and z (i, j) = f (y (i, j)) As shown in FIG. In addition, the grid grating 44 is arranged on the y stage 42. The grid lattice 44 may be a real lattice whose x and z coordinates of each grid are known. An image of the grid grating 44 when the grid grating 44 is positioned at a predetermined y coordinate y ₀ is picked up by the image pickup means 26. At this time, the x and z coordinates of the measurement points on the grid grating 44 corresponding to the coordinates (i, j) on the imaging surface 18 can be obtained from the known x and z coordinates of the surrounding grid by linear interpolation or the like. it can. Therefore, the x and z coordinates when the y coordinate of the measurement point on the measurement surface corresponding to the coordinate (i, j) on the imaging surface 18 is y ₀ are obtained. This calculation is performed on all coordinates (all image sensors) on the imaging surface 18 or arbitrary coordinates. Further, the coordinates (i, j) and y coordinate on the imaging surface 18 are associated with the x coordinate and z coordinate over the entire measurement area in the y direction or over an arbitrary range. The correspondence between the obtained coordinates (i, j) and y coordinates and the x and z coordinates is expressed as follows: functions x (i, j) = f (y (i, j)) and z (i, j) = It may be approximated to f (y (i, j)) and stored in a storage unit (not shown) of the arithmetic unit 24. Alternatively, the coordinates (i, j) and y coordinates in the function x (i, j) = f (y (i, j)) and z (i, j) = f (y (i, j)) and x The correspondence relationship with the coordinate and the z coordinate may be stored in a format such as a lookup table.

(2) Actual measurement process As shown in FIG. 1, the measuring object 50 is arrange | positioned on the optical path of the light source 12 (S10). Next, the grid pattern creating means 23 transmits image data of the coarse grid pattern 36A to the light source 12 (S12). At this time, the input signal adjusting means 14 transmits an input signal of a luminance component suitable for the projection luminance amplitude A · ST _L corresponding to the coarse lattice pattern 36A to the light source 12 (S14). The light source 12 projects an image of the coarse grid pattern 36A having a projection luminance amplitude A · ST _L onto the measurement target surface 22 of the measurement object 50 (S16).

The phase displacing means 16 changes the phase of the image of the coarse grating fringe 36A projected on the measurement surface 22 (S18). For example, the phase is changed in the order of π / 2, π, 3π / 2. The image pickup means 26 picks up an image of the displaced coarse lattice fringes 36A projected on the measurement surface 22 (S20). The relative phase calculation unit 28 obtains the phase φ _a ′ (i, j) based on the image of the coarse lattice pattern 36A captured by the imaging unit 26 (S22). The calculation of the phase φ _a ′ (i, j) is performed for all coordinates (all imaging elements) on the imaging surface 18 or arbitrary coordinates.

Next, the grid pattern creating means 23 transmits the image data of the fine grid pattern 36B to the light source 12 (S24). At this time, the input signal adjusting means 14 transmits an input signal of a luminance component suitable for the projection luminance amplitude A · ST _M corresponding to the fine grid stripes 36B to the light source 12 (S26). The light source 12 projects an image of the fine grid stripes 36B having the projection luminance amplitude A · ST _M onto the measurement target surface 22 of the measurement object 50 (S28). Thereafter, the same processing as in steps (S18), (S20), and (S22) is performed, and the relative phase φ _bk ′ (i, j) is obtained based on the image of the fine lattice stripes 36B.

Next, the grid pattern creating means 23 transmits the image data of the fine grid pattern 36C to the light source 12 (S30). At this time, the input signal adjusting unit 14 transmits an input signal of a luminance component suitable for the projection luminance amplitude A · ST _H corresponding to the fine grid stripes 36C to the light source 12 (S32). The light source 12 projects an image of the fine grid pattern 36C having the projection luminance amplitude A · ST _H onto the measurement target surface 22 of the measurement object 50 (S34). Thereafter, the same processing as in steps (S18), (S20), and (S22) is performed, and the relative phase φ _ck ′ (i, j) is obtained based on the image of the fine lattice fringes 36C.

Next, the absolute phase calculation means 30 uses the phase φ _a ′ (i, j), the relative phase φ _bk ′ (i, j), and the relative phase φ _ck ′ (i, j) to calculate the absolute phase φ ( (i, j) is narrowed down (S36). In parallel with this, the phase-y coordinate calculation means 32 derives the y coordinate corresponding to the absolute phase φ (i, j) from the relationship obtained in (1-2) (S38).

Specifically, the phase φ _a '(i, j) as a function _{y a (i, j) =} f a (φ (i, j)) from, y-coordinate y _a' is calculated (i, j) a. Next, a y-coordinate candidate y _bk ′ (i, j) is obtained from the relative phase φ _bk ′ (i, j) and the function y _b (i, j) = f _b (φ (i, j)). _{Further, y bk '(i, j} ) among the most y _a' extracts (i, j) to be close to y _b '(i, j) as a. Next, to obtain the relative phase φ _ck '(i, j) as a function _{y c (i, j) =} f c (φ (i, j)) from, y coordinate candidate y _ck' a (i, j). Among y _ck ′ (i, j), the one closest to y _b ′ (i, j) is set as the y coordinate of the measurement point (y _c ′ (i, j) = y (i, j)). Further, the phase φ _c ′ (i, j) corresponding to y _c ′ (i, j) is defined as the absolute phase φ (i, j).

  Next, the y coordinate-xz coordinate calculation means 34 derives x (i, j) and z (i, j) corresponding to y (i, j) from the relationship obtained in (1-3) above. (S40). When the x, y, and z coordinates of the measurement points on the corresponding measurement target surface are obtained for all or arbitrary coordinates (imaging element) on the imaging surface 18, the calculation unit 20 connects these coordinates. A three-dimensional shape of the surface to be measured is created (S42).

  Note that, for example, when the surface to be measured is a glossy surface, a part of the received light luminance on the imaging surface 18 may exceed the saturation level. On the other hand, a part of the received light luminance on the imaging surface 18 may be too small to obtain the phase φ. In such a case, the phase φ may be obtained by excluding a region where the saturation level is exceeded or a region where the received light luminance is too small. In other words, the shape of the surface to be measured 22 may be calculated based on a region of the imaging surface 18 in which the received light luminance falls within a predetermined allowable range.

  Furthermore, after the calculation of the shape of the measurement target surface 22 in the region is completed, the region of the imaging surface 18 in which the received light luminance falls within the allowable range may be changed. For example, the region where the received light luminance falls within the allowable range may be changed by changing the projection luminance amplitude of the light source 12.

Further, when changing the projection luminance amplitude, it is preferable to change each projection luminance amplitude while maintaining the ratio of the projection luminance amplitudes of the lattice fringe images having different spatial periods. By doing so, it is possible to make the light reception luminance amplitude the same between the respective checkered images. For example, (1-1) coefficients of projection luminance amplitudes ST _L ′ = A · ST _L , ST _M ′ = A · ST _M , and ST _H ′ = A · ST _H set in the calibration of the projection luminance component By adjusting A, each projection luminance amplitude may be changed.

  In the above embodiment, the decrease in the light reception luminance amplitude due to the MTF of the light projecting / receiving optical system is compensated by the input signal adjusting means 14 as in (1-1) calibration of the projection luminance component. It is not restricted to this form. For example, as illustrated in FIG. 10, an exposure adjustment unit 52 that determines an exposure time for the imaging surface 18 may be provided in the imaging unit 26 or the arithmetic unit 24. The exposure adjustment unit 52 may have a function of adjusting the shutter speed of the imaging unit 26.

  The exposure adjusting unit 52 projects the lattice fringe 36 having a relatively long spatial period on the surface to be measured 22 when the lattice fringe 36 having a relatively short spatial cycle is projected on the surface to be measured 22. It is preferable to determine the exposure time for the imaging surface 18 so as to be longer than the exposure time for the imaging surface 18 at that time. In other words, when the first lattice fringe image having a predetermined spatial period and the second lattice fringe image having a spatial period longer than the first lattice fringe image are selectively projected on the measurement target surface 22, The exposure time when projecting the lattice fringe image onto the surface to be measured 22 may be set longer than the exposure time when projecting the second lattice fringe image onto the surface to be measured 22. With such a configuration, it is possible to compensate for a decrease in received light amplitude due to the MTF of the light projecting / receiving optical system.

  Here, as described above, there is a case where a part of the light receiving luminance on the imaging surface 18 exceeds the saturation level, or a part of the light receiving luminance on the imaging surface 18 is too small to obtain the phase φ. is there. In such a case, the shape of the surface to be measured 22 may be calculated based on a region of the imaging surface 18 in which the received light luminance falls within a predetermined allowable range. Further, by changing the exposure time with respect to the imaging surface 18, the region of the imaging surface 18 in which the received light luminance falls within the allowable range may be changed. In addition, the exposure time may be changed after performing control to change each projection luminance amplitude while maintaining the ratio of the projection luminance amplitudes of the lattice fringe images having different spatial periods. By doing so, it is possible to more effectively prevent the saturation level of the received light luminance from being exceeded as compared with the case where only the exposure time is changed or only the projection luminance amplitude is changed.

  Further, in place of the calibration of the luminance component and the adjustment of the exposure time, an amplification means 54 for amplifying the received light luminance amplitude on the imaging surface 18 may be provided as illustrated in FIG. The amplifying unit 54 measures the lattice fringe 36 whose amplification factor of the received light intensity on the imaging surface 18 has a relatively long spatial period when the lattice fringe 36 having a relatively short spatial period is projected onto the surface to be measured 22. It is preferable that the amplification factor is determined so as to be larger than the amplification factor of the light reception luminance amplitude on the imaging surface 18 when projected onto the surface 22. The input signal adjusting unit 14, the exposure adjusting unit 52, and the amplifying unit 54 may be used in appropriate combination. In other words, when the first lattice fringe image having a predetermined spatial period and the second lattice fringe image having a spatial period longer than the first lattice fringe image are selectively projected on the measurement target surface 22, The amplification factor when projecting the lattice fringe image onto the surface to be measured 22 may be set to be larger than the amplification factor when projecting the second lattice fringe image onto the surface to be measured 22.

<Specific example 1>
Next, a specific example of three-dimensional shape measurement using the above-described calibration and measurement principle will be described. In this specific example, a projector capable of projecting a checkerboard image is used as the light source 12. The settable brightness range of the projector was 8 bits (0 to 255).

First, shape measurement was performed using lattice fringes 36A to 36C having three different cosine waveforms. The periods of the lattice fringes 36A to 36C were λ _H = 1.1 mm, λ _M = 5.5 mm, and λ _L = 44 mm, respectively.

  The calibration of the (1-1) projection luminance component was performed on the lattice fringes 36A to 36C. First, a ceramic flat plate having a known shape is placed on the optical path of the light source 12. The lattice fringe creating means 23 sequentially projects the lattice fringes 36A to 36C having the density of the cosine wave from the light source 12. Here, regarding the projection luminance amplitude of the light source 12, the offset was set to 125 and the amplitude was set to 95.

  Next, images of the lattice fringes 36 </ b> A to 36 </ b> C projected in order on the ceramic flat plate are picked up by the image pickup means 26. Next, the light reception luminance amplitude for each of the lattice fringes 36A to 36C was obtained from the cosine wave image obtained at each period of the lattice fringes 36A to 36C.

FIG. 12 shows the relationship between the spatial period λ of the lattice fringes 36A to 36C and the light reception luminance amplitude R. From this figure, R _L = 1, R _M = 0.84, and R _H = 0.35 are obtained. From these values, ST _L = R _H = 0.35, ST _M = R _H / R _M = 0.42, and ST _H = 1 are obtained. The projection brightness setting value of the light source 12 was determined from these values and the two conditions that the brightness level of the image sensor does not saturate at the closest point of the measurement range. First, with respect to all spatial period offsets, 125 out of the luminance of 8 bits (0 to 255) was set. Furthermore, the amplitude of λ _H is 115, the amplitude of λ _M is 48, and the amplitude of λ _L is 40. As described above, since the amplitudes of the lattice fringes 36A to 36C when the MTF characteristics are not considered are all 95, the amplitude (115) of λ _H can be increased. Based on these values, a total of 12 types (= 3 periods × 4 phases) of cosine waveform lattice fringes having phases of λ _H , λ _M , and λ _L , each having a phase difference of π / 2, are created. I took an image.

  FIG. 13 shows the result of measuring the position of the ceramic flat plate placed on the precision stage and moving in the y direction. In FIG. 13, the difference between the y coordinate obtained from the phase φ and the actual y coordinate on the precision stage is shown for each stage position. A white circle plot (◯) is a measurement result when the projection luminance amplitude is changed for each checkered pattern. Also, a filled circle plot (●) is a measurement result when a constant projection luminance amplitude is set for all the lattice fringes. As shown in this figure, the error between the calculated y coordinate and the actual y coordinate is reduced by adjusting the projection luminance amplitude for each lattice pattern.

<Specific example 2>
Next, a measurement example in the case where the measured surface 22 made of a glossy surface is specularly reflected will be described. Also in this specific example, a projector having a luminance range of 8 bits (0 to 255) was used as the light source 12. In this specific example, an exposure adjusting means 52 for adjusting the exposure time for the imaging surface 18 is provided. Specifically, the exposure adjustment unit 52 is a shutter speed changing unit that adjusts the shutter speed of the imaging unit 26.

  As shown in FIG. 10, a measurement object 50 made of a gloss standard plate is placed on the optical path of the light source 12. At this time, the measurement object 50 was installed so that the inclination θ of the measured surface 22 with respect to the optical axis of the light source 12 was 15 °.

Also in this specific example, cos waveform lattice fringes 36A to 36C having three different periods λ _H = 1.1 mm, λ _M = 5.5 mm, and λ _L = 44 mm were used. Nine sheets (G1 to G9) having 60 ° glossiness of 1 to 93% and different glossiness standard plates used as the measurement object 50 were used. This gloss standard plate may be, for example, manufactured by Aiki. The glossiness was measured with a gloss meter GM-268Plus manufactured by Konica Minolta. The glossiness of each standard plate is shown in FIG.

First, the amplitude attenuation amount of the lattice fringes 36A to 36C was measured. With respect to the inclination θ of the glossiness standard plate, θ = 0 ° is set on the optical path (front side) of the light source 12, and the lattice fringes 36 </ b> A to 36 </ b> C from the light source 12 with different cosine wave shades are changed to the glossiness standard plate in order. The projected image was captured by the imaging means 26. From the cosine wave image obtained at each period of the lattice fringes 36A to 36C, the relationship between the period and the amplitude of the lattice fringes 36A to 36C was obtained. This measurement was performed on nine glossy plates (G1 to G9). Based on this measurement result, the relationship between the period and the amplitude as shown in FIG. 15 was obtained. In addition, the average value of the amplitude obtained from the measurement of nine glossy plates is used for the amplitude. The result of FIG. 15 is different from the result of FIG. 12 in the specific example 1, but is considered to be due to the difference in the surface reflection characteristics. In addition, since the glossiness standard plate was small and could project only up to the lattice fringes with a period λ = 30 mm, the amplitude value of λ _L was extrapolated.

From FIG. 15, R _L = 1.02, R _M = 0.78, and R _H = 0.7 were obtained. Further, from these values, ST _L = 0.69, ST _M = 0.88, and ST _H = 1 were obtained. The brightness setting value of the light source 12 was set on the condition that the brightness level of the image sensor does not saturate at these values and the closest point of the measurement range. Specifically, the offset of all periods was 125, the amplitude of λ _H was 115, the amplitude of λ _M was 100, and the amplitude of λ _L was 78. Since the amplitude of the lattice fringes 36A to 36C when the MTF characteristics were not taken into account was 95, the amplitude of λ _H could be increased.

Based on these values, the lattice fringe creating means 23 creates a total of 12 types of cos waveform lattice fringes having phases of λ _H , λ _M , and λ _L each having a phase difference of π / 2. Here, the checkered image of the glossiness standard plate is shown in FIGS. The number of seconds in parentheses represents the shutter speed. As shown in FIGS. 16 and 17, the higher the glossiness is, the more the lattice pattern image is spotted. Using this spot portion as a measurement region, the three-dimensional coordinates were obtained.

  Further, in order to suppress the luminance saturation of the image pickup device during the image pickup of the glossiness standard plate, the exposure adjustment means 52 changes the shutter speed every 10 times from 1/15 seconds to 1/80000 seconds so that the lattice fringes are generated. I took an image.

  First, the shutter speed is set to 1/15 seconds. Then, the coarse lattice fringes 36A, the fine lattice fringes 36B, and the fine lattice fringes 36C were projected in this order, and the three-dimensional coordinates of the glossiness standard plate were obtained in the same manner as in Example 1. At this time, if there is an image sensor whose luminance is saturated in the measurement area, the image is taken again at a higher shutter speed. For example, when the image is saturated at 1/15 seconds, the image is taken again at a shutter speed of 1/120 seconds. If there is an image sensor that saturates even at this speed, the image was taken with the shutter speed increased to 1/1000 second. The measurement was performed at an increased shutter speed until no image sensor was saturated.

  From the three-dimensional coordinate measurement values obtained in this way, the three-dimensional plane of the glossiness standard plate was obtained by the method of least squares. Further, the difference between the coordinates of the actual three-dimensional plane and the measured coordinates was obtained as the position variation. The result is shown in FIG. A white diamond plot (◇) is a measurement result when the projection luminance amplitude and the shutter speed are changed for each checkered pattern. Further, the filled diamond plot (♦) is a measurement result when a certain projection luminance amplitude is set for all the lattice fringes. By projecting while changing the projection luminance amplitude value, the position variations of the standard plates G1 to G8 are smaller by 0.3 to 3.8 μm than when the projection luminance amplitude value is the same. In addition, when the projection luminance amplitude values were the same, the glossy standard plate G9 was erroneously measured, but measurement was possible by changing the projection luminance amplitude value and the shutter speed and projecting.

  DESCRIPTION OF SYMBOLS 10 Shape measuring device, 12 Light source, 14 Input signal adjustment means, 16 Phase displacement means, 18 Imaging surface, 20 Computation part, 22 Surface to be measured, 23 Plaid preparation means, 24 Arithmetic device, 26 Imaging means, 28 Relative phase calculation means , 30 absolute phase calculation means, 32 phase-y coordinate calculation means, 34 y coordinate-xz coordinate calculation means, 36 grid stripes, 40 standard plate, 42 y stage, 44 grid grating, 50 measurement object, 52 exposure adjustment means, 54 Amplification means.

Claims (10)

First projection means for projecting a first checkered image having a first spatial period and a first projected luminance amplitude onto a surface to be measured;
Second projecting means for projecting a second checkered image having a second spatial period and a second projected luminance amplitude onto a surface to be measured;
A phase displacement means capable of changing the phase of the lattice fringe image projected on the measurement surface;
An imaging means capable of imaging a checkerboard image projected on the measurement surface;
A calculation unit that calculates a shape of a surface to be measured based on a change in light receiving luminance of the imaging unit corresponding to a change in the phase;
With
The first spatial period is shorter than the second spatial period;
The first projection luminance amplitude is greater than the second projection luminance amplitude;
Shape measuring device. The shape measuring device according to claim 1,
When the first checkered image projected on the surface to be measured is picked up, the received light amplitude on the image pickup surface of the image pickup means picks up the second checkered image projected on the surface to be measured. A shape measuring apparatus capable of controlling the first projection luminance amplitude and the second projection luminance amplitude so as to be substantially equal to a light reception luminance amplitude on an imaging surface. The shape measuring device according to claim 1 or 2,
When imaging the first checkered image projected on the surface to be measured, the light reception luminance amplitude on the image pickup surface of the imaging means, and when capturing the second checkered image projected on the surface to be measured, A shape measuring apparatus capable of controlling the first projection luminance amplitude and the second projection luminance amplitude so that a light reception luminance amplitude on an imaging surface is less than a luminance saturation level. The shape measuring device according to claim 2 or 3,
A shape measuring apparatus capable of controlling the first projection luminance amplitude and the second projection luminance amplitude while maintaining a ratio between the first projection luminance amplitude and the second projection luminance amplitude. The shape measuring apparatus according to any one of claims 1 to 4,
The calculation unit calculates a shape of the surface to be measured based on an area of the imaging surface of the imaging unit where the received light luminance falls within a predetermined range,
A shape measuring apparatus that changes a region in which the received light luminance falls within the predetermined range by changing the first projection luminance amplitude and the second projection luminance amplitude. First projecting means for projecting a first checkered image having a first spatial period onto a surface to be measured;
A second projection means for projecting a second checkered image having a second spatial period onto the surface to be measured;
A phase displacement means capable of changing the phase of the lattice fringe image projected on the measurement surface;
An imaging means capable of imaging a checkerboard image projected on the measurement surface;
A first exposure adjusting means for controlling an exposure time to a first exposure time when the imaging means picks up the first checkered image projected on the measurement surface;
A first exposure adjusting means for controlling an exposure time to a second exposure time when the imaging means picks up the second checkered image projected on the measurement surface;
A calculation unit that calculates a shape of a surface to be measured based on a change in light receiving luminance of the imaging unit corresponding to a change in the phase;
With
The first spatial period is shorter than the second spatial period;
The first exposure time is longer than the second exposure time;
Shape measuring device. The shape measuring device according to claim 6,
When imaging the first checkered image projected on the surface to be measured, the first light reception luminance amplitude on the image pickup surface of the imaging means when capturing the second checkered image projected on the surface to be measured The shape measuring apparatus capable of controlling the first exposure time and the second exposure time so as to be substantially equal to the second light receiving luminance amplitude on the imaging surface. The shape measuring device according to claim 6 or 7,
When imaging the first checkered image projected on the measurement surface, the first received light amplitude on the imaging surface of the imaging means and when capturing the second checkered image projected on the measurement surface. The shape measuring apparatus capable of controlling the first exposure time and the second exposure time so that the second light reception luminance amplitude on the imaging surface is less than the luminance saturation level. The shape measuring device according to any one of claims 6 to 8,
The calculation unit calculates a shape of the surface to be measured based on an area of the imaging surface of the imaging unit where the received light luminance falls within a predetermined range,
A shape measuring apparatus that changes a region in which the received light luminance falls within the predetermined range by changing the first exposure time and the second exposure time. First projecting means for projecting a first checkered image having a first spatial period onto a surface to be measured;
A second projection means for projecting a second checkered image having a second spatial period onto the surface to be measured;
A phase displacement means capable of changing the phase of the lattice fringe image projected on the measurement surface;
An imaging means capable of imaging a checkerboard image projected on the measurement surface;
A first amplifying unit that amplifies the light reception luminance amplitude on the imaging surface of the imaging unit by a first amplification factor when the imaging unit images the first checkered image projected on the surface to be measured;
A second amplifying means for amplifying the light reception luminance amplitude on the imaging surface of the imaging means by a second amplification factor when the imaging means images the second checkered image projected on the surface to be measured;
A calculation unit that calculates a shape of the surface to be measured based on a change in received light luminance after amplification of the imaging unit corresponding to the change in the phase;
With
The first spatial period is shorter than the second spatial period;
The first amplification factor is greater than the second amplification factor;
Shape measuring device.