JP3996560B2 - Object shape measuring device - Google Patents

Description

  The present invention relates to an object shape measurement apparatus that performs measurement of a shape of a measurement object in a non-contact manner by irradiating a measurement object with an irradiation pattern and imaging a reflection imaging pattern reflected from the measurement object.

Irradiate the measurement object with the irradiation light pattern from the projector, capture the reflected light pattern from the measurement object with a camera, and measure the shape of the measurement object in a non-contact manner based on the displacement of the reflected light pattern from the irradiation light pattern. Three-dimensional measurement by the pattern projection method is known.
As the pattern projection method, a fringe scanning method using a fringe pattern having a sinusoidal light amount distribution with different phases as the irradiation light pattern is often used. The principle of the fringe scanning method is described in Non-Patent Document 1 to be described later. Explained.
According to the description of Non-Patent Document 1, when the measurement object is irradiated with an irradiation pattern in which the illuminance change I (x) in the x-axis direction is I (x) = A (x) + B {φ (x) + φo}, The light quantity distribution of the reflected imaging pattern from the measurement object imaged by the camera with the irradiation pattern corresponding to φo = 0, π / 2, 3π / 2 is Io, I1, I2, I3, and the inverse function of the function tan is atan. The phase distribution φ (x) is expressed by equation (1).

φ (x) = atan {(I3 (x) −I1 (x)) / (Io (x) −I2 (x)}
... (1)

  In the equation (1), φ (x) corresponds to the irradiation angle of the projector. However, since φ takes a value from 0 to 2π for each region, it cannot be associated with the irradiation angle in a one-to-one manner. Assuming that the region number is Lid (1, 2,..., M−1) and the global phase in the measurement region is φg, the global phase φg and the irradiation angle can be associated with each other using equation (2).

        φg = φ + 2πLid (2)

  The irradiation angle α from the projector can be known for each pixel of the reflection imaging pattern from φg in the expression (2), the incident angle β of the incident light for each pixel of the camera, the preset camera and projector Based on the distance L between the optical centers, the distance z between the measurement object surface and the optical center of the camera for each pixel of the captured image of the measurement object is measured by the equation (3) from the principle of triangulation.

        z = L / (tan α + tan β) (3)

  In the fringe scanning method, the region number Lid cannot be known from the imaging pattern, and therefore a process called unwrapping is performed to obtain φg from φ having an indefinite phase. As this unwrapping process, methods such as attaching markers to specific fringes, limiting the spatial frequency of fringes and using Fourier transform, and assuming phase continuity between adjacent periods are adopted. Yes.

By the way, when measuring the shape of an object by the pattern projection method, as shown in FIG. 7, the irradiation light from the projector 2 is irradiated onto the measurement object 1, and the reflection imaging pattern from the point A of the measurement object 1 is obtained. In the case of imaging, for example, the irradiation light from the projector 2 is multiple-reflected at a point B near the point A of the measurement object 1, and the original irradiation light Iio and the reflected light at the point B are reflected at the point A of the measurement object 1. Is incident as multiple scattered light Iim. For this reason, from the point A of the measurement object 1, the reflected imaging pattern light Iro + Irm based on the irradiation light Iio and the multiple scattered light Iim is incident on the camera 8, and a captured image based on this reflected imaging pattern light is obtained.
In this way, when multiple scattered light obtained by integrating the reflected light around the measurement point is included, the captured image distribution is smoothed and functions as a kind of low-pass filter related to spatial frequency. When the irradiation pattern consisting of the above is irradiated, the higher the spatial frequency is, the easier it is to reduce. The contrast of the captured image deteriorates.

By the way, as already explained by the equation (1), the light quantity distribution Pa excluding the background light is Pa = I3−I1 and Pb = Io−I2, and the irradiation angle α is calculated from φ (x) = atan (Pa / Pb). However, when the contrast of the sine wave decreases, the absolute values of Pa and Pb decrease.
On the other hand, since the light quantity sensitivity of the camera is finite, Pa and Pb are quantized and measured. When the absolute values of Pa and Pb are reduced, the information amount is reduced, and the reliability of the phase distribution φ (x) is lowered, and the measurement is performed. The measurement error of the object increases.
In order to solve this problem, Japanese Patent Application Laid-Open No. H10-228561, which will be described later, shows that when a spatial coding method is used as a pattern projection method, a spatial code assigned to a measurement space changes under a certain rule. Based on the above, there is disclosed a three-dimensional shape measurement method for detecting the influence of multiple reflections by predicting a spatial code in a space adjacent to a certain spatial code.
In the method of Patent Document 1, by providing a spatial code increase / decrease extraction step, a spatial code measured by deviating from a predetermined increase / decrease beyond a predetermined value is removed as being affected by multiple reflections. .
Similarly, in Patent Document 2 to be described later, a difference image obtained by using a spatial encoding method as a pattern projection method and taking a light amount difference between a pattern of a captured image obtained and a predetermined reference pattern is set in advance. An image noise detection apparatus that determines light amount noise based on whether a threshold value is exceeded is disclosed.

The three-dimensional shape measurement method disclosed in Patent Document 1 described above is a pattern projection other than the spatial code method that predicts a code in the spatial coding method and detects the influence of multiple reflection based on deviation from the prediction. It cannot be applied to the law.
Further, in the disclosure of Patent Document 2 described above, the difference image indicates the light amount, and the absolute value of the light amount itself depends greatly on measurement conditions such as the reflectance of the measurement object, and is based on the influence of multiple reflection. It is not suitable for the judgment.
Edited by Toru Yoshizawa “Three Dimensional Measurement” New Technology Communications JP 2000-193438 A Japanese Patent Laid-Open No. 7-225834

  The present invention has been made in view of the current state of operation of an object shape measuring apparatus using this kind of pattern projection method as described above, and its purpose is to measure the shape of a measurement object measured under the influence of multiple reflections. An object of the present invention is to provide an object shape measuring apparatus capable of detecting the reliability of measurement data with high accuracy in a state in which the reliability of measurement data is appropriately compatible with various pattern projection methods.

  In order to achieve the object, the invention according to claim 1 is provided with an irradiation unit that irradiates a measurement object with an irradiation pattern composed of a repeated light pattern, and the irradiation pattern is disposed at a position different from the irradiation unit. An object shape measuring apparatus comprising: an imaging unit that captures a reflected imaging pattern obtained by reflection by an object; and a shape calculating unit that calculates a surface shape of the measurement object based on the reflected imaging pattern captured by the imaging unit , The lower limit value of the light amount distribution of the irradiation pattern is Plb, the upper limit value of the light amount distribution of the irradiation pattern is Pub, the light amount of the irradiation pattern is Ia, and the contrast ratio R indicating the measurement reliability of the measurement object is Ia / (Pub-Plb) reliability calculation means, surface shape data calculated by the shape calculation means and the reliability calculation means It is characterized in that it has a data output means for outputting the contrast ratio data.

  According to such means, the irradiation object is irradiated with an irradiation pattern composed of a repeated light pattern by the irradiation means, and the irradiation pattern is obtained by being reflected by the measurement object by the imaging means arranged at a position different from the irradiation means. The reflection imaging pattern is imaged, and the shape calculation means calculates the surface shape of the measurement object based on the reflection imaging pattern imaged by the imaging means. The lower limit value of the light distribution of the irradiation pattern is Plb, and the irradiation is performed. Assuming that the upper limit value of the light quantity distribution of the pattern is Pub and the light quantity of the irradiation pattern is Ia, the reliability calculation means calculates the contrast ratio R indicating the measurement reliability of the measurement object based on Ia / (Pub−Plb). Surface shape data calculated by the shape calculation means and contrast ratio data calculated by the reliability calculation means by the data output means Therefore, the reliability of the shape measurement data of the measurement object measured under the influence of multiple reflections is detected with high accuracy in a state that corresponds widely and accurately to various pattern projection methods, and the captured image Accurately grasp for each pixel.

  Similarly, in order to achieve the object, the invention according to claim 2 is the irradiation according to claim 1, wherein A, B, and φc are constants, φ is a phase in a range of 0 to 2π, and an initial phase is φo. The light quantity distribution I of the pattern is I = A + Bcos (φ + φo), the lower limit value Plb of the light quantity distribution of the irradiation pattern is Plb = A−B, the upper limit value Pub of the light quantity distribution of the irradiation pattern is Pub = A + B, and φo = 0. , Π / 2, π, 3π / 2, the light quantity distribution I of the irradiation pattern is Ppc, Pns, Pnc, Pps, the inverse function of the tan function is antan, the measure function is φ, and the measurement object by the shape calculation means Is calculated based on φ = atan {(Pps−Pns) / (Ppc−Pnc)} + φc, and the calculation of the contrast ratio R by the reliability calculation means is R = {cos (φ + φ ) (And is characterized in Ppc-Pnc) + sin (φ + φo) (Pps-Pns)} / (be performed based on Pub-Plb).

  According to such means, the measurement object 1 is irradiated with four kinds of repeated patterns corresponding to the light quantity distribution I = I + A + Bcos (φ + φo) and φo of 0, π / 2, π, 3π / 2 as irradiation patterns. By using a large number of irradiation patterns, the measurement object is measured with high accuracy, and the operation of the invention according to claim 1 is executed.

  Similarly, in order to achieve the object, the invention according to claim 3 is the irradiation according to claim 1, wherein A, B, and φc are constants, φ is a phase in the range of 0 to 2π, and an initial phase is φo. The light quantity distribution I of the pattern is I = A + Bcos (φ + φo), the lower limit value Plb of the light quantity distribution of the irradiation pattern is Plb = A−B, the upper limit value Pub of the light quantity distribution of the irradiation pattern is Pub = A + B, and φo = 0. , Π / 2, π of the irradiation pattern light quantity distribution I is Ppc, Pns, Pnc respectively, the inverse function of the tan function is antan, the measure function is φ, and the shape calculation means calculates the surface shape of the measurement object. , Φ = atan {(Pnc−Pns) / (Ppc−Pns)} + φc, and the calculation of the contrast ratio R by the reliability calculation means is R = [cos (φ + φo) (Ppc−Pn). c) + 2sin (φ + φo) {(Pub + Plb) / 2Pns] / (Pub−Plb).

  According to such means, the measurement object 1 is irradiated with three types of repeating patterns corresponding to the light quantity distribution I = I + A + Bcos (φ + φo) and φo of 0, π / 2, and π, as claimed in claim 2. By using a smaller number of irradiation patterns than in the described invention, the operation of the invention described in claim 1 is performed with high-accuracy measurement on the measurement object while reducing the necessary measurement time.

  Similarly, in order to achieve the object, the invention according to claim 4 is the irradiation according to claim 1, wherein A, B, and φc are constants, φ is a phase in a range of 0 to 2π, and an initial phase is φo. The light quantity distribution I of the pattern is I = A + Bcos (φ + φo), the lower limit value Plb of the light quantity distribution of the irradiation pattern is Plb = A−B, the upper limit value Pub of the light quantity distribution of the irradiation pattern is Pub = A + B, and φo = 0. , Π / 2, the light quantity distribution I of the irradiation pattern is Ppc and Pns, the inverse function of the tan function is antan, the measure function is φ, and the calculation of the surface shape of the measurement object by the shape calculation means is φ = atan [{(Pub + Plb) / 2Pns} / {Ppc− (Pub + Plb) / 2}]] + φc, and the calculation of the contrast ratio R by the reliability calculation means is R = 2 [cos ( φ + φo) {Ppc (Pub + Plb) / 2} + sin (φ + φo) {(Pub + Plb) / 2} −Pns] / (Pub−Plb).

  According to such means, the measurement object 1 is irradiated as the two repetitive patterns corresponding to the light amount distribution I = I + A + Bcos (φ + φo), φo = 0, and π / 2. By using a smaller number of irradiation patterns than that described in the invention, the measurement time can be further shortened for highly accurate measurement on the measurement object, and the operation of the invention described in claim 1 is executed.

  Similarly, in order to achieve the above object, according to a fifth aspect of the present invention, in the first aspect of the present invention, the irradiation pattern is obtained by optical scanning with respect to a plurality of slits, and A and B are constants, and the light quantity of the irradiation pattern. Is the lower limit value Plb = A−B of the light amount distribution of the irradiation pattern in the portion without the slit, and the upper limit value Pub = A + B of the light amount distribution of the irradiation pattern in the slit portion, and the calculation of the surface shape of the measurement object by the shape calculating means Is performed based on the light quantity distribution Ps with respect to the slit pattern, and the calculation of the contrast ratio R by the reliability calculation means is performed based on R = {2Ps− (Pub + Plb)} / (Pub−Plb). To do.

  According to such a means, the irradiation pattern is obtained by optical scanning with respect to a plurality of slits, and the light quantity of the irradiation pattern with A and B as constants is the lower limit value Plb = A−B of the light distribution of the irradiation pattern in the portion without the slit. The upper limit value Pub = A + B of the light distribution of the irradiation pattern at the slit portion, and the calculation of the surface shape of the measurement object by the shape calculating means is performed based on the designated slit position, and the contrast ratio R is calculated by the reliability calculating means. Is performed based on R = {2Ps− (Pub + Plb)} / (Pub−Plb), and the amount of light is two values of brightness and darkness, and is not easily affected by multiple reflections. The action is executed.

  Similarly, in order to achieve the object, the invention according to claim 6 is the invention according to claim 1, in which the light quantity distribution of the irradiation pattern is A-B in which A and B used in the spatial coding method are constants. And A + B binary rectangular wave distribution, the lower limit value Plb of the light quantity distribution of the irradiation pattern is A−B, the upper limit value Pub of the light quantity distribution of the irradiation pattern is A + B, and the surface shape of the measurement object by the shape calculating means Is calculated based on the spatial code obtained from the light quantity distribution Psc for the 1-level portion and the 0-level portion of the irradiation pattern, and the contrast ratio R is calculated by the reliability calculating means R = {2Psc− (Pub + Plb)} / (Pub−Plb) is performed.

  According to such means, the light quantity distribution of the irradiation pattern is a binary rectangular wave distribution of AB and A + B with A and B used as constants in the spatial coding method, and the lower limit value of the light distribution of the irradiation pattern Plb is A−B, the upper limit value Pub of the light amount distribution of the irradiation pattern is A + B, and the calculation of the surface shape of the measurement object by the shape calculating means is obtained from the light amount distribution Psc for the 1 level portion and the 0 level portion of the irradiation pattern. The contrast ratio R is calculated based on the space code, and the contrast ratio R is calculated based on R = {2Psc− (Pub + Plb)} / (Pub−Plb). Then, the operation according to the first aspect of the present invention is performed in a state in which it is hardly affected by the multiple reflection.

  Similarly, in order to achieve the above object, according to a seventh aspect of the invention, in the first aspect of the invention, the data output means has a contrast ratio calculated by the reliability calculation means lower than a preset reference value. For measurement coordinates, the output of contrast ratio data corresponding to the surface shape data calculated by the shape calculation means is prohibited.

  According to such means, the data output means corresponds to the surface shape data calculated by the shape calculating means for the measurement coordinates in which the contrast ratio calculated by the reliability calculating means falls below a preset reference value. Since the output of the contrast ratio data is prohibited, highly reliable shape measurement data of the measurement object is extracted and detected.

  Similarly, in order to achieve the object, the invention according to claim 8 is the invention according to claim 1, in which the data output means calculates the contrast ratio data calculated by the reliability calculation means and the surface shape calculated by the shape calculation means. When sampling data is output to the data, the coordinates of the sampling points are the coordinates obtained by weighting and averaging the coordinates of the neighboring points with the contrast ratio, and the contrast ratio of the sampling points is the average value of the contrast ratio of the neighboring coordinates. It is.

  According to such means, when the data output means performs sampling output on the contrast ratio data calculated by the reliability calculation means and the surface shape data calculated by the shape calculation means, the coordinates of the sampling points are set to the neighboring coordinates. Since the contrast ratio at the sampling point is the average value of the contrast ratio of the neighboring coordinates, the measurement reliability between the sampling point and the neighboring point is used when sampling the shape measurement data of the measurement object. Based on the calculation, the measurement reliability of the sampling output is increased.

  According to the first aspect of the invention, the reliability of the shape measurement data of the measurement object measured under the influence of multiple reflection is detected with high accuracy in a state that corresponds widely and accurately to various pattern projection methods, It becomes possible to accurately grasp each pixel of the captured image.

  According to the second aspect of the present invention, four types of repetitive patterns corresponding to the light amount distribution I of I = A + Bcos (φ + φo) and φo of 0, π / 2, π, 3π / 2 are applied to the measurement object 1 as irradiation patterns. Since it is irradiated, by using a large number of irradiation patterns, it becomes possible to measure the measurement object with high accuracy and realize the effect obtained by the invention of claim 1.

  According to the third aspect of the present invention, the measurement object 1 is irradiated with three kinds of repeated patterns corresponding to the light amount distribution I = I + A + Bcos (φ + φo) and φo of 0, π / 2, and π as the irradiation pattern. By using a smaller number of irradiation patterns than the invention according to claim 2, the effect obtained by the invention according to claim 1 can be realized by reducing the measurement time required for highly accurate measurement on the measurement object. Is possible.

  According to the fourth aspect of the present invention, since the two repetitive patterns corresponding to the light quantity distribution I of I = A + Bcos (φ + φo) and φo of 0, π / 2 are irradiated to the measurement object 1 as the irradiation pattern, By using a smaller number of irradiation patterns than in the invention described in Item 3, it is possible to further reduce the measurement time for highly accurate measurement on the measurement object and realize the effect obtained by the invention described in Item 1. Become.

  According to the invention described in claim 5, the irradiation pattern is obtained by optical scanning with respect to a plurality of slits, and the light quantity of the irradiation pattern with A and B as constants is the lower limit value Plb = A -B, the upper limit value Pub = A + B of the light distribution of the irradiation pattern at the slit portion, and the calculation of the surface shape of the measurement object by the shape calculating means is performed based on the designated slit position, and the contrast ratio R by the reliability calculating means The calculation according to claim 1 is performed based on R = {2Ps− (Pub + Plb)} / (Pub−Plb), and the amount of light is light and dark so that it is not easily affected by multiple reflections. The effect obtained by the invention can be realized.

  According to the invention of claim 6, the light quantity distribution of the irradiation pattern is a binary rectangular wave distribution of AB and A + B with A and B used as constants in the spatial coding method, and the light quantity distribution of the irradiation pattern The lower limit value Plb is A−B, the upper limit value Pub of the light distribution of the irradiation pattern is A + B, and the calculation of the surface shape of the measurement object by the shape calculating means is based on the light amount distribution Psc for the 1 level portion and the 0 level portion of the irradiation pattern. The contrast ratio R is calculated based on the obtained space code, and the contrast ratio R is calculated based on R = {2Ps− (Pub + Plb)} / (Pub−Plb). For this reason, it is possible to realize the effect obtained by the invention according to the first aspect in a state in which it is hardly affected by the multiple reflection.

  According to the seventh aspect of the present invention, the surface shape data calculated by the shape calculating means is applied to the measurement coordinates in which the contrast ratio calculated by the reliability calculating means falls below a preset reference value. Since the output of the corresponding contrast ratio data is prohibited, it is possible to extract and detect shape measurement data with high reliability of the measurement object.

  According to the invention described in claim 8, when the data output means performs sampling output on the contrast ratio data calculated by the reliability calculation means and the surface shape data calculated by the shape calculation means, the coordinates of the sampling points are set. Since the coordinates obtained by weighting and averaging the neighboring coordinates with the contrast ratio are used, and the contrast ratio of the sampling points is the average value of the contrast ratios of the neighboring coordinates, the measurement reliability between the sampling points and the neighboring points is used when sampling the shape measurement data of the measurement object. The calculation based on the degree makes it possible to increase the measurement reliability of the sampling output.

(First embodiment)
Example 1
A first embodiment of the present invention will be described with reference to FIGS.
FIG. 1 is a block diagram showing the overall configuration of this embodiment, FIG. 2 is a characteristic diagram showing the characteristics of the light quantity distribution and contrast ratio of the imaging light of this embodiment, and FIG. 3 is an image obtained by shape measurement of this embodiment. It is an image.

In the present embodiment, as shown in FIG. 1, the measurement object 1 is provided with a light source 3, a liquid crystal panel 5, and an imaging lens 6 so as to face the measurement object 1 to be measured. A projector 2 that irradiates, and a camera 8 that includes an imaging lens 10 and a CCD 11 and that captures a reflected imaging pattern from the measurement object 1 are arranged with the distance between the optical centers set to L.
The projector 2 uses the liquid crystal panel 5 to convert the light from the light source 3 into a dynamic irradiation pattern composed of repeated light patterns and irradiate the measurement object 1, and the upper limit pattern of the maximum uniform light quantity and the minimum uniform of the repeated light pattern. It has a function of irradiating the measurement object 1 with the lower limit pattern of the amount of light. In addition, the camera 8 has a function of taking a reflected imaging pattern from the measurement object 1 from the imaging lens 10, performing photoelectric conversion by the CCD 11, and outputting it as an imaging signal.
Further, in FIG. 1, the coordinate origin is set at the optical center of the imaging lens 10 of the camera 8, and the optical center of the imaging lens 6 of the projector 2 is located on the x-axis of this coordinate.

A projector control unit 7 that controls the operation of the projector 2 is connected to the projector 2, a camera control unit 12 that controls the operation of the camera 8 and extracts an imaging signal is connected to the camera 8, and the camera control unit 12 A shape calculation unit 13 that calculates the shape of the measurement object 1 from the imaging signal and a reliability calculation unit 14 that calculates a contrast ratio indicating the reliability from the imaging signal are connected. The data calculation unit 13 and the reliability calculation unit 14 take in the output signal of the shape calculation unit 13 and the output signal of the reliability calculation unit 14, and perform data processing as necessary to output the data. The data output unit 15 is connected to a memory 16 in which output data from the data output unit 15 is stored as a file 17.
Further, in this embodiment, sine wave patterns having different phases by π / 2 are used as irradiation patterns, and A, B, and φc are constants, φ is a phase in the range of 0 to 2π, and an initial phase is φo. In the light quantity distribution I, four types of irradiation patterns of φo = 0, π / 2, π, and 3π / 2 are used as I = A + Bcos (φ + φo).

The shape measuring operation of the measuring object 1 of the present embodiment having such a configuration will be described.
The projector 2 irradiates the measurement object 1 with a sinusoidal wave pattern having a light intensity represented by the following expression with respect to the irradiation angle φ, with Sa and Sb as constants.

Spc = Sa + Sbcos (φ + φo) (4a)
Sns = Sa−Sbsin (φ + φo) (4b)
Snc = Sa−Sbcos (φ + φo) (4c)
Sps = Sa + Sbsin (φ + φo) (4d)

  On the other hand, the light amounts of the lower limit pattern LB and the upper limit pattern UB, which serve as a reference for the light amount amplitude, are expressed by the following equations with uniform light amounts equal to the lower limit value and the upper limit value of the light amount of the sine wave pattern, respectively.

Slb = Sa−Sb (5a)
Sub = Sa + Sb (5b)

  When the reflected light from the measurement object 1 having the irradiation patterns shown in (4a) to (4d) is imaged by the camera 8, the imaging light amount is A for the background light, the reflectance of the measurement object 1, the camera 8, the measurement object 1, and A coefficient based on the geometric arrangement of the projector 2 is B, and a contrast coefficient of the effect of lowering the light amount amplitude due to multiple reflection due to the spatial frequency of the irradiation pattern is C (0 ≦ C ≦ 1).

Ppc = A + B · C2 cos (φ + φo) (6a)
Pns = A−B · C2sin (φ + φo) (6b)
Pnc = A−B · C2 cos (φ + φo) (6c)
Pps = A + B · C2sin (φ + φo) (6d)

  Further, since the lower limit pattern LB and the upper limit pattern UB with uniform light quantity have different spatial frequencies, they are expressed by the following equations using a contrast coefficient C1 different from the sine wave pattern.

Plb = A−B · C1 (7a)
Pub = A + B · C1 (7b)

  On the other hand, the phase distribution φ for obtaining the irradiation angle is given by the following equation.

φ = atan {(Pps−Pns) / (Ppc−Pnc)} + φc
... (8)

  In the present embodiment, the global phase is obtained from the local phase obtained by the equation (8) by the already described unwrapping, and the corresponding irradiation angle α is determined. Then, based on the imaging angle β obtained from the imaging position of the imaging pattern and the distance L between the optical centers of the camera and the projector, the shape calculation unit 13 captures an image of the measurement object 1 based on the already described equation (3). A distance z between the measurement object surface for each pixel of the image and the optical center of the camera 8 is calculated as z = L / (tan α + tan β), and the calculation data is input to the data output unit 15.

  By the way, the amplitude of the pattern observed by imaging corresponds to the product of the coefficient B and the contrast coefficient C, and is expressed as the following equation.

B2a = B · C2 = (Ppc−Pnc) / 2cos (φ + φo)
... (9a)
B2b = B · C2 = (Pps−Pns) / 2sin (φ + φo)
... (9b)
B1 = B · C1 = (Pub−Plb) / 2 (10)

  In Equations (9a) and (9b), the amplitudes B2a and B2b for the sine wave pattern are the same amount. However, in this state, depending on the values of cos and sin in the denominator, the values may be indefinite, so this is avoided. And the sum of weights of {cos (φ + φo)} ∩2 and {sin (φ + φo)} ∩2 for B2a and B2b, respectively, is changed to B2 To do.

B2 = B2a {cos (φ + φo)} ∩2 + B2b {sin (φ + φo) ∩2
= Cos (φ + φo) (Ppc−Pnc) / 2
+ Sin (φ + φo) (Pps−Pns) / 2
(11)

  In this case, B1 and B2 are equal if there is no amount of multiple reflections, but B1> B2 due to the influence of multiple reflections. Therefore, the ratio of B1 and B2 is contrast R = B2 / B1, and is given by the following equation.

R = {cos (φ + φo) (Ppc−Pnc) + sin (φ + φo) (Pps
-Pns)} / (Pub-Plb) (12)

Ppc, Pns, Pnc, Pps, Plb, and Pub with respect to the imaging angle of the camera 8 (image forming position on the CCD 11) are as shown in FIG. 2 (a), but Pub and Plb are measured to have a uniform irradiation light amount. Since the image of the object 1 is multiplied by the reflectance, it is not originally a straight line, but in the figure, for simplicity, Pub and Plb are straight lines. Further, Pub and Plb are less affected by multiple reflection, but the irradiation pattern of the irradiation pattern decreases as the spatial frequency increases due to the influence of multiple reflection. When viewed from the camera 8, as shown in FIG. The width between the envelopes Ena1 and Enb1 indicated by is smaller than the width between Pub and Plb.
The reliability calculation unit 14 calculates the contrast ratio R by the equation (12), and the calculation data is input to the data output unit 15. The calculated contrast ratio R is as shown in FIG. 2 (b). In this case, the influence of multiple reflection is large at a portion where the imaging angle is large, and the contrast R is small.

In this example, the measurement object is a white polystyrene foam face model, and the captured image IM obtained by imaging with a white plate as a background and the result of calculating the contrast ratio are as shown in FIG. Corresponds to a contrast ratio of 1 and black corresponds to a contrast ratio of 0. The completely black portion is a portion where the amount of reflected light is small and cannot be measured, and the number of cells u and v indicate the pixel position of the CCD 11. The fringes seen in the period of the irradiation pattern in FIG. 3 show a deviation between the irradiated sine wave pattern and the true sine wave pattern, and it is possible to correct the fringes so that they do not occur.
In FIG. 3, since the influence of multiple reflections is large around the eyes, under the nose, upper jaw, and under the throat, the captured image is dark, and the contrast ratio is set for each of the spatial coordinates obtained in the captured image. The reliability can be determined by calculation.

  From the data output unit 15 of FIG. 1, the x, y, z coordinate data of each pixel calculated by the shape calculation unit 13 based on the equation (3), and the reliability calculation unit 14 based on the equation (12). Point cloud data to which the calculated contrast ratio data is added is output to the memory 16. In this case, when the user selects as necessary, a predetermined threshold value is preset for the contrast ratio, and only x, y, z coordinate data having a contrast ratio exceeding the threshold value is output from the data output unit 15. . For example, in the case of the measurement of the measurement object shown in FIG. 3, it was determined that the condition was satisfied if the threshold value was set to 0.2, according to the comparison between the required accuracy and the actually measured shape.

Also, when sampling output is performed from all point cloud data, the number of sampling points to be output and the number N of neighboring points for weighting are designated, and the sampling position of the point cloud data is selected closely in the portion where the shape curvature of the measuring object 1 is large. The initial sampling position (coordinate number) is determined by selecting coarsely in a portion having a small shape curvature. Next, N coordinates in the vicinity of the sampling position are subjected to weighted averaging processing with a contrast ratio based on equation (13), and sampling output is performed.
That is, assuming that the kth measurement coordinate (a vector having three components of x, y, and z coordinates) is r_k, the contrast ratio of the kth measurement coordinate is R_k, the position number of the sampling point is i, and the position number near i. A weighted averaging operation is performed on j based on the following equation, and the accurate sampling coordinates are shifted from r_i to the coordinate side with high reliability.

        r_i = (Σr_jR_j) / (ΣR_j) (13)

  In this way, according to the present embodiment, the projector 2 irradiates the light amount distribution I consisting of four repetitive patterns corresponding to I = A + Bcos (φ + φo) and φo of 0, π / 2, π, 3π / 2. A pattern is irradiated onto the measurement object 1 and the light quantity distribution of the imaging pattern picked up by the camera 8 is Ppc, Pns, Pnc, Pps, the upper limit value of the irradiation pattern light quantity distribution is Pub, the lower limit value is Plb, and the phase distribution φ Is atan {(Pps−Pns) / (Ppc−Pnc)} + φc, and the distance z between the measurement object surface and the camera 8 is calculated by z = L / (tan α + tan β). The ratio R is calculated based on R = {cos (φ + φo) (Ppc−Pnc) + sin (φ + φo) (Pps−Pns)} / (Pub−Plb). Therefore, the reliability of the shape measurement data of the measurement object measured under the influence of multiple reflections is detected with high accuracy in a state that is widely and accurately compatible with various pattern projection methods, and is suitable for each pixel of the captured image. It becomes possible to grasp accurately.

(Second Embodiment)
Example 2
A second embodiment of the present invention will be described.
In this embodiment, referring to FIG. 1, the projector 2 emits three types of repetitive patterns corresponding to light amount distribution I = I + A + Bcos (φ + φo) and φo of 0, π / 2, and π. The measurement object 1 is irradiated as a pattern.
Since the configuration of the other parts of the present embodiment is the same as that of the first embodiment already described, duplicate description will not be given.

In the present embodiment, the light amount distributions Ppc, Pns, and Pnc of the imaging pattern obtained from the three types of repeated patterns are the same as those in the first embodiment, as described above with 6 (a), 6 (b), and 6 (c). The lower limit pattern LB and the upper limit pattern UB are equal to 7 (a) and 7 (b), respectively.
In this embodiment, the phase distribution φ is given by the following equation as φc = (π / 4) −φo.

φ = atan {(Pnc−Pns) / (Ppc−Pns)} + φc
(14)

  Further, the amplitude of the pattern observed by imaging corresponds to the product of the coefficient B and the contrast coefficient C, and is expressed as the following equation.

B2a = B · C2 = (Ppc−Pnc) / 2cos (φ + φo)
... (15a)
B2c = B · C2 = {(Pub + Plb) / 2 · Pns} / sin (φ + φo)
... (15b)
B1 = B · C1 = (Pub−Plb) / 2 (16)

  In order to perform stable calculation, the following equation is obtained by using B2 as the sum of B2a and B2c weighted by {cos (φ + φo)} ∩2 and {sin (φ + φo)} ∩2, respectively.

B2 = B ・ C2
= B2a {cos (φ + φo)} ∩2 + B2b {sin (φ + φo)} ∩2
= Cos (φ + φo) (Ppc−Pnc) / 2
+ Sin (φ + φo) (A−Pns) / 2 (17)

  The contrast ratio R = B2 / B1 is given by the following equation.

R = [cos (φ + φo) (Ppc−Pnc) +2 sin (φ + φo) {(Pub
+ Plb)} / 2Pns] / (Pub-Plb)
... (18)

Thus, according to the present embodiment, the calculation of the surface shape of the measurement object by the shape calculation means is performed based on φ = atan {(Pnc−Pns) / (Ppc−Pns)} + φc, and the reliability calculation is performed. The calculation of the contrast ratio R by the means is performed based on R = [cos (φ + φo) (Ppc−Pnc) +2 sin (φ + φo) {(Pub + Plb) / 2Pns] / (Pub−Plb).
The other operations of the present embodiment are the same as those of the first embodiment already described, and therefore will not be described repeatedly.

According to the present embodiment, by using a smaller number of irradiation patterns than the first embodiment already described, in addition to the effect obtained in the first embodiment, the shape measurement of the measurement object is performed with a reduced measurement time, It becomes possible to realize highly accurate shape measurement for the measurement object 1.
The other effects of the present embodiment are the same as those of the first embodiment already described, and therefore will not be described repeatedly.

(Third embodiment)
Example 3
A third embodiment of the present invention will be described.
In this embodiment, FIG. 1 will be used to explain. From the projector 2, two types of repetitive patterns corresponding to the light amount distribution I = I = A + Bcos (φ + φo), φo = 0, and π / 2 are used as irradiation patterns. The measurement object 1 is irradiated.
Since the configuration of the other parts of the present embodiment is the same as that of the first embodiment already described, duplicate description will not be given.

In the present embodiment, the light quantity distributions Ppc and Pns of the imaging pattern obtained from the two types of repeated patterns are equal to 6 (a) and 6 (b) already described in the same manner as in the first embodiment, and the lower limit pattern LB, The upper limit pattern UB is equal to 7 (a) and 7 (b).
In this embodiment, the phase distribution φ is given by the following equation as φc = −φo.

φ = atan [{(Pub + Plb) / 2Pns} / {Ppc− (Pub
+ Plb) / 2}] + φc
... (19)

  Further, the amplitude of the pattern observed by imaging corresponds to the product of the coefficient B and the contrast coefficient C, and is expressed as the following equation.

B2d = B · C2 = {Ppc− (Pub + Plb) / 2} / cos (φ + φo)
... (20a)
B2c = B · C2 = {(Pub + Plb) / 2−Pns} / sin (φ + φo) (20b)
B1 = B · C1 = (Pub−Plb) / 2 (21)

  In order to perform stable calculation, the following equation is obtained by setting B2d and B2c as weighted sums of {cos (φ + φo)} ∩2 and {sin (φ + φo)} ∩2, respectively, as B2.

B2 = BC2
= B2a {cos (φ + φo)} ∩2 + B2b {sin (φ + φo)} ∩2
= Cos (φ + φo) {Ppc− (Pub + Plb) / 2}
+ Sin (φ + φo) {(Pub + Plb) / 2−Pns}
(22)

  The contrast ratio R is given by the following equation.

R = 2 [cos (φ + φo) {Ppc− (Pub + Plb) / 2} + sin (φ + φo) {(Pub + Plb) / 2} −Pns] / (Pub−Plb)
(23)

Thus, according to the present embodiment, the calculation of the surface shape of the measurement object by the shape calculation means is based on φ = atan [{(Pub + Plb) / 2Pns} / {Ppc− (Pub + Plb) / 2}] + φc. When the contrast ratio R is calculated by the reliability calculation means, R = 2 [cos (φ + φo) {Ppc− (Pub + Plb) / 2} + sin (φ + φo) {(Pub + Plb) / 2} −Pns] / (Pub− Plb).
The other operations of the present embodiment are the same as those of the first embodiment already described, and therefore will not be described repeatedly.

According to the present embodiment, by using a smaller number of irradiation patterns than the second embodiment already described, in addition to the effect obtained in the first embodiment, the measurement of the shape of the measurement object can be performed with a further reduction in the measurement time. It becomes possible to realize highly accurate shape measurement for the measurement object 1.
The other effects of the present embodiment are the same as those of the first embodiment already described, and therefore will not be described repeatedly.

(Fourth embodiment)
Example 4
A fourth embodiment of the present invention will be described with reference to FIG.
FIG. 4 is a characteristic diagram showing the characteristics between the light quantity distribution of the imaging light and the contrast ratio in this embodiment.

In the present embodiment, the irradiation pattern is obtained by optical scanning with respect to a plurality of slits, and the light amount of the irradiation pattern is a lower limit value Plb = AB of the irradiation pattern in a portion where there is no slit, where A and B are constants. The upper limit value Pub = A + B of the light amount distribution of the irradiation pattern at the slit portion, and the calculation of the surface shape of the measurement object by the shape calculating means is performed based on the designated slit position.
In this embodiment, since a plurality of slits are used, it is necessary to specify a slit number for a plurality of slits imaged on the CCD 11. For this reason, the range of the surface height of the measurement object is limited to correspond to the slit width, so that the overlapping of the imaging range does not occur in each slit during imaging.
Since the configuration of the other parts of the present embodiment is the same as that of the first embodiment already described, duplicate description will not be given.

  The imaging light amount distribution of the irradiation pattern by the slit of this embodiment is expressed by the equation (24a) in the slit portion and (24b) in the portions other than the slit portion.

Ps = A + B · C2 (24a)
Ps = A−B · C2 (24b)

Further, the lower limit pattern LB and the upper limit pattern UB are expressed by the expressions (7a) and (7b) described above, respectively.
Furthermore, in the present embodiment, the amplitude of the imaging pattern is expressed by the equation (25a) for the slit portion of the slit pattern and by (25b) for the uniform pattern.

B2 = B · C2 = Ps− (Pub + Plb) / 2
... (25a)
B1 = B · C1 = (Pub−Plb) / 2 (25b)

  The contrast ratio R is expressed by the following equation.

R = B2 / B1 = {2Ps− (Pub + Plb)} / (Pub−Plb)
... (26)

In this embodiment, as shown in FIG. 4A, the envelope Ena2 corresponding to the slit position that becomes the peak portion of the imaging light amount distribution Ps of the irradiation pattern and the slit that becomes the valley portion of the imaging light amount distribution Ps of the irradiation pattern. The difference from the envelope Enb2 corresponding to the other portion is the light amount amplitude of the irradiation pattern. As shown in FIG. 4B, the contrast ratio R calculated at the slit position according to the equation (26) is equal to the envelope Ena2, but since the high frequency component of the spatial frequency is small in portions other than the slit, the envelope Enb2 is Since the shape measurement is performed only at the slit position in this case, the contrast ratio R is calculated by the equation (26).
The other operations of the present embodiment are the same as those of the first embodiment already described, and therefore will not be described repeatedly.

According to the present embodiment, the irradiation pattern is obtained by optical scanning with respect to a plurality of slits, and by performing shape measurement only at the slit positions, the light quantity becomes a binary value of brightness and darkness, and the measurement object is not easily affected by multiple reflections. It becomes possible to perform shape measurement with high reliability.
The other effects of the present embodiment are the same as those of the first embodiment already described, and therefore will not be described repeatedly.

(Fifth embodiment)
Example 5
A fifth embodiment of the present invention will be described with reference to FIGS.
FIG. 5 is a characteristic diagram showing the characteristics of the light quantity distribution of the irradiation light and the space number in this embodiment, and FIG. 6 is a characteristic chart showing the characteristics of the light quantity distribution of the imaging light and the contrast ratio in this embodiment.

In this embodiment, the light amount distribution of the irradiation pattern is a binary rectangular wave distribution of AB and A + B with A and B used as constants in the spatial coding method, and the lower limit value Plb of the light amount distribution of the irradiation pattern is A-B, the upper limit value Pub of the light amount distribution of the irradiation pattern is A + B, and the calculation of the surface shape of the measurement object by the shape calculating means is obtained from the light amount distribution Psc for the 1 level portion and the 0 level portion of the irradiation pattern It is comprised so that it may be performed based on.
Since the configuration of the other parts of the present embodiment is the same as that of the first embodiment already described, duplicate description will not be given.

In this embodiment, as shown in FIGS. 5A and 5B, a spatially coded pattern having a light quantity distribution obtained by binarizing the spatial code with respect to the irradiation angle is irradiated. Associate space numbers Lid up to 15. In the case of FIG. 5A, Igo is the lowest bit position and Ig3 is the highest bit position. However, if the number of coding bits is increased, finer spatial resolution can be obtained.
In the present embodiment, an Igo pattern at the lowest bit position having the highest spatial frequency is imaged, and the contrast ratio is calculated therefrom. The light amount distribution with respect to the imaging angle at which the lowest bit pattern is imaged is as shown in FIG. 6A, and the pattern light amount at the lowest bit position is located between the upper limit pattern and the lower limit pattern due to the influence of multiple reflection. To do.

  The imaging light quantity distribution Psc of the spatial code pattern of this embodiment is given by the equation (27a) at the 1st level portion and by the equation (27b) at the 0th level portion.

Psc = A + B · C2 (27a)
Psc = A−B · C2 (27b)

When n-bit coding is performed, the number of spatial coding patterns is n, the number of discriminable spatial codes is 2 to the nth power, and when n = 8, the number of spatial coding patterns is 8, and the number of spatial codes is 256. Thus, the measurement range can be divided into 256 parts.
Further, the lower limit pattern LB and the upper limit pattern UB are respectively expressed by the expressions (7a) and (7b) already described.
Further, the amplitude of the pattern observed by imaging is expressed by the following equation with respect to the slit portion of the slit pattern and the uniform pattern.

B2 = BC2 = Psc− (Pub + Plb) / 2
... (28)

  And contrast ratio R in a slit part is represented by following Formula.

R = B2 / B1 = {2Psc- (Pub + Plb)} / (Pub-Plb)
... (29)

  The contrast ratio R calculated based on the equation (29) is as shown in FIG. 6B, and the contrast ratio R deteriorates where the imaging angle is large, so the reliability of the spatial code decreases. Corresponding to the boundary position where the amount of light in FIG. 6 (a) changes greatly, the contrast ratio R in FIG. 6 (b) is lowered. This is due to the blurring of the edge of the irradiation pattern. It is not due to the occurrence. The contrast ratio indicated by the envelope Enc in FIG. 6B indicates the influence of the occurrence of multiple reflection.

According to the present embodiment, the irradiation pattern is based on the spatial encoding method, the light amount distribution is a binary rectangular wave distribution, the light amount is light and dark binary, and is not easily affected by multiple reflections, and the shape of the measurement object Measurement can be performed with high reliability.
Other effects of the present embodiment are the same as those of the first embodiment already described, and therefore, a duplicate description will not be given.

  The present invention relates to measurement of human body shape in the medical field such as orthopedics and oral surgery, automatic measurement in the men's and women's clothing industry, measurement of three-dimensional CAD data from gray models such as automobiles, cameras and electrical products, It can be used for 3D.

It is a block diagram which shows the whole structure of Example 1 of this invention. It is a characteristic view which shows the characteristic of light quantity distribution and contrast ratio of the imaging light of the Example. It is the captured image obtained by the shape measurement in the same Example. It is a characteristic view which shows the characteristic of the light quantity distribution and contrast ratio of the imaging light of Example 4 of this invention. It is a characteristic view which shows the characteristic of light quantity distribution and space number of irradiation light of Example 5 of this invention. It is a characteristic view which shows the characteristic of the light quantity distribution and contrast ratio of the imaging light of the Example. It is explanatory drawing which shows the principle of the conventional pattern measuring method.

Explanation of symbols

1 Measurement Object 2 Projector 7 Projector Control Unit 8 Camera 11 CCD
12 Camera Control Unit 13 Shape Calculation Unit 14 Reliability Calculation Unit 15 Data Output Unit 16 Memory

Claims (8)

An irradiating unit that irradiates the measurement object with an irradiation pattern composed of a repeated light pattern; and an imaging unit that is disposed at a position different from the irradiating unit and that captures a reflected imaging pattern obtained by reflecting the irradiation pattern on the measurement object; An object shape measuring apparatus comprising shape calculating means for calculating a surface shape of the measurement object based on a reflection image pickup pattern picked up by the image pickup means, wherein the lower limit value of the light amount distribution of the irradiation pattern is Plb, and the irradiation A reliability calculation means for calculating the contrast ratio R indicating the measurement reliability of the measurement object by Ia / (Pub−Plb), where the upper limit value of the light distribution of the pattern is Pub, the light amount of the irradiation pattern is Ia,
An object shape measuring apparatus comprising: data output means for outputting surface shape data calculated by the shape calculating means and contrast ratio data calculated by the reliability calculating means. A, B, and φc are constants, φ is a phase in the range of 0 to 2π, and the initial phase is φo, the light distribution I of the irradiation pattern is I = A + Bcos (φ + φo), and the lower limit Plb of the light distribution of the irradiation pattern is Plb = A−B, the upper limit value Pub of the light amount distribution of the irradiation pattern is Pub = A + B,
Shape calculation means with the light intensity distribution I of the irradiation pattern of φo = 0, π / 2, π, 3π / 2 as Ppc, Pns, Pnc, Pps, the inverse function of the tan function as atan, and the measure function as φ, respectively Is calculated based on φ = atan {(Pps−Pns) / (Ppc−Pnc)} + φc,
The contrast ratio R is calculated by the reliability calculating means based on R = {cos (φ + φo) (Ppc−Pnc) + sin (φ + φo) (Pps−Pns)} / (Pub−Plb). The object shape measuring apparatus according to claim 1. A, B, and φc are constants, φ is a phase in the range of 0 to 2π, and the initial phase is φo, the light distribution I of the irradiation pattern is I = A + Bcos (φ + φo), and the lower limit Plb of the light distribution of the irradiation pattern is Plb = A−B, the upper limit value Pub of the light amount distribution of the irradiation pattern is Pub = A + B,
The surface shape of the object to be measured by the shape calculation means, where the light quantity distribution I of the irradiation pattern of φo = 0, π / 2, π is Ppc, Pns, Pnc, the inverse function of the tan function is atan, and the measure function is φ. Is calculated based on φ = atan {(Pnc−Pns) / (Ppc−Pns)} + φc,
The contrast ratio R is calculated by the reliability calculating means based on R = [cos (φ + φo) (Ppc−Pnc) +2 sin (φ + φo) {(Pub + Plb) / 2Pns] / (Pub−Plb). The object shape measuring apparatus according to claim 1. A, B, and φc are constants, φ is a phase in the range of 0 to 2π, and the initial phase is φo, the light distribution I of the irradiation pattern is I = A + Bcos (φ + φo), and the lower limit Plb of the light distribution of the irradiation pattern is Plb = A−B, the upper limit value Pub of the light amount distribution of the irradiation pattern is Pub = A + B,
The calculation of the surface shape of the measurement object by the shape calculation means with the light quantity distribution I of the irradiation pattern of φo = 0, π / 2 as Ppc and Pns, the inverse function of the tan function as atan, and the measure function as φ, φ = atan [{(Pub + Plb) / 2Pns} / {Ppc− (Pub + Plb) / 2}] + φc,
The calculation of the contrast ratio R by the reliability calculation means is R = 2 [cos (φ + φo) {Ppc− (Pub + Plb) / 2} + sin (φ + φo) {(Pub + Plb) / 2} −Pns] / (Pub−Plb) The object shape measuring apparatus according to claim 1, wherein the object shape measuring apparatus is performed based on the above.   An irradiation pattern is obtained by optical scanning with respect to a plurality of slits, and A and B are constants, and the light amount of the irradiation pattern is a lower limit value Plb = A−B of the light amount distribution of the irradiation pattern in a portion without the slit, and the slit portion The upper limit value Pub = A + B of the light quantity distribution of the irradiation pattern is obtained, the calculation of the surface shape of the measurement object by the shape calculating means is performed based on the light quantity distribution Ps with respect to the slit pattern, and the contrast ratio R is calculated by the reliability calculating means. 2. The object shape measuring apparatus according to claim 1, which is performed based on R = {2Ps− (Pub + Plb)} / (Pub−Plb).   The light quantity distribution of the irradiation pattern is a binary rectangular wave distribution of AB and A + B with A and B as constants used in the spatial coding method, and the lower limit value Plb of the light quantity distribution of the irradiation pattern is AB. The upper limit value Pub of the light amount distribution of the irradiation pattern is A + B, and the calculation of the surface shape of the measurement object by the shape calculating means is based on the spatial code obtained from the light amount distribution Psc for the 1 level portion and the 0 level portion of the irradiation pattern. 2. The object shape measuring apparatus according to claim 1, wherein the calculation of the contrast ratio R by the reliability calculation means is performed based on R = {2Psc- (Pub + Plb)} / (Pub-Plb).   The data output means prohibits the output of contrast ratio data corresponding to the surface shape data calculated by the shape calculating means for measurement coordinates where the contrast ratio calculated by the reliability calculating means falls below a preset reference value. The object shape measuring apparatus according to claim 1. When the data output means performs sampling output on the contrast ratio data calculated by the reliability calculation means and the surface shape data calculated by the shape calculation means, the coordinates of the sampling points are weighted and averaged by the contrast ratio with the coordinates of the sampling points. The object shape measuring apparatus according to claim 1, wherein the coordinates are used, and the contrast ratio of the sampling points is an average value of the contrast ratios of neighboring coordinates.

Images