JP2006201037A - Shape measuring apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a precise shape measuring apparatus which can easily realize a modulation pattern at a photographing step, is easy to implement the modulation pattern and determines a distance (phase) from refection pulse light returning along the direction of an object to be measured. <P>SOLUTION: The shape measuring apparatus includes an irradiation means 3 for irradiating the object to be measured 2 with an irradiation pattern being a light pattern, and photographing means 4-6 disposed at positions optically identical to that of the irradiation means 3 to obtain a photographed pattern being a light pattern reflected by the surface of the object to be measured 2, and computes a surface shape of the object to be measured 2 from the photographed pattern. The object to be measured 2 is disposed in a prescribed range of distance from the irradiation means 3, and the irradiation means 3 irradiates the object to be measured 2 with the irradiation pattern having a certain intensity modulated with an angular frequency ω, and the photographing means 4-6 modulate reflection light which is reflected by the surface of the object to be measured 2 and returns again to the photographing means 4-6 by passing through the path identical to that of the irradiation, with the angular frequency ω and a phase ϕd, and then detect a temporal average of light intensity, thereby deriving the distance L of the object to be measured 2 from the light intensity obtained under a plurality of modulation conditions. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a non-contact type shape measuring apparatus that irradiates a measurement object with an irradiation pattern and calculates a surface shape of the measurement object from an imaging pattern reflected on the surface of the measurement object.

Conventionally, when a person or a product is imaged, there is a demand for extracting a background or measuring the surface shape of a measurement object, and several measurement methods for this purpose are known (for example, Non-Patent Document 1 and Patents). Reference 1 and 2).
Such measurement is usually called a light cutting method disclosed in Non-Patent Document 1, and the distance from the camera to the measurement object by observing the deformation with the camera while irradiating the surface of the measurement object with the optical slit, That is, the surface shape is often measured, or the surface shape of the measurement object is often measured using parallax information from a plurality of cameras called a stereo method.
Since these methods use the principle of triangulation, a distance is required between the light source and the camera or between the cameras. These parallaxes may result in parts that are not visible to each other (so-called “hidden”). Since shape data cannot be obtained for invisible parts, it can be avoided by analogizing from neighboring points or by changing the measurement position and re-measurement, but it may be impossible to eliminate hiding if the surface shape is complicated .
There is a TOF (Time of Flight) method as a measurement method that does not cause hiding. This method measures the optical feedback time by emitting light from a distance measuring device and reflecting it back on the measurement object by some method.
FIG. 6 is a block diagram illustrating a conventional measuring apparatus. This measuring apparatus includes a modulation circuit 1, a semiconductor laser 2, a collimator lens 3, a beam splitter 4, a mirror 5 that reflects light from the beam splitter 4, a photodetector 6, and a condenser lens that collects light that has passed through the beam splitter 4. 7, a motor 9 for driving the scanner 8 that scans the measurement object 13, a period / frequency measurement circuit 10, a distance calculation circuit 11, and an obstacle detection circuit 12.
The measurement apparatus of FIG. 6 disclosed in Patent Document 1 is a method of measuring a phase shift corresponding to a distance by irradiating a frequency-modulated laser light source and taking a beat with reflected light.
According to the technique of Patent Document 1, distance measurement can be performed with relatively high accuracy. However, since there is no element capable of two-dimensionally detecting a phase shift, there is a disadvantage that only one distance can be measured at a time, and it is not much for shape measurement. Not suitable. Another problem is that the circuit for detecting the phase of the beat corresponding to the distance information is complicated.

FIG. 7 is a block diagram for explaining another conventional measuring apparatus. FIG. 8 is a diagram showing light intensity modulated in a linear manner that affects distance accuracy. FIG. 9 is another diagram showing the light intensity modulated in a straight line that affects the distance accuracy.
In FIG. 7, the illumination light modulation signal S <b> 1 from the signal generation unit 15 enters the light projecting unit 16 and projects the illumination light S <b> 6 onto the subject 17. The reflected light S7 from the subject 17 enters the imaging unit 18, and the video signals S4a and S4b are processed by the signal processing unit 19 as outputs from the imaging unit 18, and the three-dimensional information signal S5 is output from the signal processing unit 19.
In this case, the imaging gain signal S2 from the signal generation unit 15 is applied to the imaging unit 18, and the control signal S3 is applied to the signal processing unit 19. This measuring apparatus irradiates light pulses increasing and decreasing in the time direction, and measures the distance by measuring the ratio of the reflected light intensity measured for each of these two pulses.
The measurement apparatus of FIG. 7 which is the disclosure of Patent Document 2 can have a distance resolution of several centimeters by performing light quantity measurement in a very narrow time zone of 1 ns or less. In addition, by using a two-dimensional light receiving element having a short-time shutter function, distance measurement can be performed for each pixel of the image at once.
FIG. 8 is a graph showing the light intensity modulated linearly, and FIG. 9 is another graph showing the light intensity modulated linearly. In order to know the position based on the timing of the reflected light pulse that has been reflected (corresponding to phase detection in a sine wave), the distance accuracy is linearly modulated light intensity as shown in FIG. 8 and FIG. It depends on the accuracy of the shape.
Since the triangular pulse includes a wide range of frequency components, the circuit tends to be expensive to produce a highly accurate triangular pulse. Similarly, since an optical shutter that opens and closes for a very short time is driven by a rectangular pulse including a high frequency, the circuit is also likely to be expensive.
Further, since the reflectance of the surface of the measurement object can be canceled because the intensity ratio of the two types of light pulses is used, it is assumed that the background light can be ignored in determining the distance of the reflected pulse returning from the direction of the measurement object.
Toru Yoshizawa, Optical 3D Measurement, New Technology Communications, 1993, pp. 28-37 JP-A-7-103714 JP 2000-121339 A

However, the conventional distance measurement method based on the TOF method has a problem that the response is high-speed and wide-bandwidth or has a complicated circuit in order to determine the phase of the reflected pulse returning from the direction of the measurement object.
Accordingly, an object of the present invention is to easily realize a modulation pattern at the time of imaging in consideration of the above-described situation, and to easily implement this modulation pattern, and from the reflected pulse light returning from the measurement object direction (phase) It is an object of the present invention to provide a simple and highly accurate shape measuring apparatus for making a decision.

In order to solve the above-described problem, the invention according to claim 1 obtains an irradiation unit that irradiates a measurement object with an irradiation pattern that is a light pattern, and an imaging pattern that is a light pattern reflected by the surface of the measurement object. And a shape measuring device that calculates the surface shape of the measurement object from the imaging pattern, and the measurement object exists within a predetermined distance from the irradiation means. The irradiation unit irradiates the measurement object with an irradiation pattern having a predetermined intensity modulated at an angular frequency ω, and the imaging unit returns to the imaging unit after being reflected on the surface of the measurement object and following the same path as the irradiation. The reflected light is modulated with the angular frequency ω and the phase φd, and then the time average of the amount of light is detected, and the distance L of the measurement object is obtained under a plurality of modulation conditions based on the detected time average. And obtaining, based on the.
In the invention according to claim 2, the light quantity distribution of the imaging pattern obtained for the four types of modulation conditions in which the phase φd is 0, π / 2, π, and 3π / 2 is represented by Ppc, Pns, If Pnc and Pps, the shape measuring apparatus according to claim 1 is characterized in that the phase function S is φ = atan ((Pns−Pps) / (Ppc−Pnc)).
In the invention according to claim 3, when the light quantity distribution of the imaging pattern obtained with respect to the three types of modulation conditions in which the phase φd is 0, π, and 3π / 2 is Ppc, Pnc, and Pps, The shape measuring apparatus according to claim 1, wherein the phase function S is φ = atan ((Pnc−Pps) / (Ppc−Pps)) + π / 4.

In the invention according to claim 4, the light quantity distribution of the imaging pattern obtained for the two types of modulation conditions in which the phase φd is 0 and π / 2 is Ppc, Pns, and D = 0. 2. The shape measuring apparatus according to claim 1, wherein the phase function S is φ = atan ((Pns−Pf) / (Ppc−Pf)), where Pf is a light amount distribution of the imaging pattern obtained with respect to the modulation condition. To do.
According to the fifth aspect of the present invention, the modulation frequency of irradiation and imaging is nω (n is a natural number), the outline of the shape is obtained from the phase φ1 of n = 1, and the phase φn obtained by sequentially increasing n is obtained. From the above, the shape measuring device according to claim 1 is characterized in that the distance L is obtained in detail while estimating an undetermined integer m by L = L2-Lo (φn−2mπ + nφo) / (2πn).
In the invention described in claim 6, a reflector is provided at a distance Lr in front of the irradiating means (L1 ≦ Lr ≦ L2), and the phase φr of the reflected light returning from the reflector is φo = φr−2π ( The shape measuring apparatus according to claim 1, wherein φo is obtained from L 2 −Lr) / Lo.
The invention according to claim 7 is characterized in that the irradiation means is a semiconductor light source, and the imaging means is a CCD element with a light shutter or a CMOS element.
In the invention described in claim 8, the irradiation means is a semiconductor light source such as a laser diode or a light emitting diode, and the imaging means is a CCD element or a CMOS element capable of modulating optical gain. Features the device.
The invention according to claim 9 is characterized by the shape measuring apparatus according to claim 1, wherein the light source of the irradiating means is an infrared light source, and the image pickup means is provided with an optical filter for removing visible light.
The invention according to claim 10 uses the light source of the irradiating means as an infrared light source, and provides a mirror for separating infrared light and visible light in the imaging means, and measures the shape from the separated infrared light, The shape measuring apparatus according to claim 9, wherein texture data to be attached to a shape from visible light is obtained.

  According to the present invention, a light pulse modulated in a sine wave shape is irradiated, and the reflected light is modulated with a sine wave of the same frequency and then averaged and phase detection is performed, so the circuit is simple for both irradiation and imaging. Since the phase accuracy is high, the cost of the apparatus can be reduced.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a schematic view showing a first embodiment of a shape measuring apparatus according to the present invention. In this shape measuring apparatus 1, the irradiation means includes a laser diode 3, and generates an irradiation pattern by modulating the output light of the laser diode 3.
The imaging means comprises a light receiving element array 6 in which a lens 5 and a CMOS photodiode are two-dimensionally arranged, and an optical shutter 4 disposed immediately before the lens 5. Since the optical center of the lens 5 and the light emission center of the laser diode 3 are optically aligned at the same position using the half mirror 7, when viewed from the measurement object 2, the irradiation means 3 and the imaging means seem to be at the same place. Looks like.
Therefore, unlike a normal light cutting method, there is no parallax between illumination and imaging. The light pulse irradiated from the laser is bent by the half mirror 7 and irradiated onto the measurement object 2. A part of the light scattered on the measurement object 2 follows the same path as the irradiation, passes through the half mirror 7, passes through the optical shutter 4, and forms an image on the light receiving element. If the focal length of the lens 5 is f, the distance between the lens 5 and the light receiving element is approximately f.
As described above, the shape measuring apparatus 1 includes the irradiation unit 3 that irradiates the measurement object 2 with the light pattern (irradiation pattern), and the irradiation unit 3 that obtains the light pattern (imaging pattern) reflected from the surface of the measurement object 2 and the optical unit. Imaging means 4, 5, 6 at the same position, and calculation means (not shown) for calculating the surface shape of the measurement object 2 from the imaging pattern.

The measurement object 2 exists in the range of distances L1 to L2 (L2> L1) from the irradiation unit 3, and the irradiation unit 3 and the imaging units 4, 5, and 6 are optically on the same optical axis. Therefore, the irradiation means 3 irradiates the measurement object 2 with an irradiation pattern of intensity P0 = A + Bcos (ωt) modulated at the angular frequency ω (ω = πc / Lo, Lo = L2-L1, c is the speed of light, t is the time, A and B are constants depending on measurement conditions).
The imaging means 4, 5, 6 are reflected by the surface of the measurement object 2, and return to the imaging means 4, 5, 6 by following the same path as the irradiation Pr = Ω (A + Bsin (ωt + φ)) (Ω is measurement) The time average P = <PrM> of the light quantity is detected after the modulation M = C + Dcos (ωt + φd) is performed with the angular frequency ω and the phase φd.
Here, C and D are constants depending on measurement conditions, and <> means a time average of the time interval T. Ta is sufficiently large with respect to the modulation period To (To = 2π / ω). Using the light amounts P1, P2,... Obtained under a plurality of modulation conditions with different E, D, and φd, and eliminating A, B, C, D, and Ω from those P, the phase delay φ of the modulation The distance L of the measuring object 2 is obtained from L = L2−Lo (φ−φo) / 2π (φo is a constant) through the phase function φ = S (P1, P2,.
The measurement object 2 is at a distance L (L1 ≦ L ≦ L2) from the irradiation position (L1 <L2), Lo = L2−L1, c is the speed of light, and the modulated light of wavelength λ = 2Lo, that is, from λ = 2πc / ω The measurement object 2 is irradiated with light modulated at an angular frequency ω = πc / Lo.
In this case, when reflected light that is reflected on the surface of the measurement object 2 and returns to the irradiation position by following the same path as the irradiation is detected, the phase φ of the observed modulation depends on the distance L from the irradiation position. There is a delay. Under an appropriate initial phase φo, for example, this phase φ can correspond to φ = 2π at L1 and φ = 0 at L2.

In the case of a plane wave, the time can be expressed as t, and the propagation of the plane wave in the z direction can be described as sin (ωt− (2π / λ) z + φo). The light reflected at the position L and returning to the irradiation position changes from z = 2L and λ = 2Lo to sin (ωt− (2π / Lo) L + φo) when taking into consideration the round-trip path. The phase φ at t = td is φ = ωtd− (2π / Lo) L + φo
Here, for example, if φo = −ωtd + (2π / Lo) L2, it is possible to correspond to φ = 2π in L1 and φ = 0 in L2.
Further, when considering the intensity of the light, if the modulation intensity of the irradiation light is A + B cos ωt (A and B are constants), the attenuation according to the reflectance of the surface of the measurement object 2 and the irradiation means 3 to the imaging means 4, The modulation intensity of the reflected light is Ω (A + Bcos (ωt + φ)), where Ω is a coefficient depending on the attenuation of light depending on the geometrical arrangement of 5 and 6, and Ω.
At this time, the phase φ of the reflected light is φ = 2π (L2−L) / Lo + φo.
Can be written. If the reflected light phase φ can be measured by some method,
L = L2−Lo (φ−φo) / (2π)
From this, the distance L is known, and the shape of the surface of the measuring object 2 can be measured. In order to determine the value of φo, for example, the obtained phase may be adjusted so as to correspond to 2π ·· 0 at the distance L1 ·· L2.

In the first embodiment, irradiation light of A + B cos ωt is generated by a laser or the like whose light quantity is modulated, and this is irradiated to a measurement object. The reflected light Ω (A + Bcos (ωt + φ)) having a phase delay of φ is modulated by (C + Dcos (ωt + φd)) by a high-speed modulator or the like in the detection means (C, D, and φd are constants). .
Thereafter, the light receiving element array 6 detects the time average of the light amount. The time average is a value obtained by dividing the integrated value of the light amount by Ta at a time Ta = nTo + Tm (n is a sufficiently large integer, 0 ≦ Tm <To), where To = (2π / ω) is a modulation period. To do.
When the time average of the time series signal f (t) with respect to t = 0 ·· Ta is represented by (∫f (t) dt) / Ta as <f (t)>, the irradiation light quantity P0 and the reflection on the surface of the measurement object 2 are reflected. The amount of reflected light Pr returned to the imaging means 4, 5, 6 is as follows.
P0 = A + Bcos (ωt)
Pr = Ω (A + Bcos (ωt + φ))
Of the modulation conditions C, D, and φd, an example of the light modulation M when φd and D are specified and the amount of received light P received by the light receiving element (light receiving element array) 6 of the imaging means after modulation is shown.
Light modulation when φd = π3 / 2 Mps = C + Dsinωt
Amount of received light Pps = <PrMps>
= Ω (AC + BD <sinωtcos (ωt + φ)>)
= Α-βsinφ
Light modulation when φd = π / 2 Mns = C-Dsinωt
Amount of received light Pns = <Pr Mns>
= Ω (AC-BD <sinωtcos (ωt + φ)>)
= Α + βsinφ

Light modulation when φd = 0 Mpc = C + Dcosωt
Amount of received light Ppc = <Pr Mpc>
= Ω (AC + BD <cosωtcos (ωt + φ)>)
= Α + βcosφ
Light modulation when φd = π Mnc = C-Dcosωt
Amount of received light Pnc = <Pr Mnc>
= Ω (AC-BD <cosωt cos (ωt + φ)>)
= Α-βcosφ
Light modulation when D = 0 Mf = C
Received light intensity Pf = <Pr Mf>
= Ω <AC + BC cos (ωt + φ)>
= Α
Here, α = ΩAC, β = ΩBD / 2, and the fact that the following relationship holds is used.
<Sin ^ 2ωt> = <cos ^ 2ωt> = 1/2
<Sinωt cosωt> = 0
<Sin (ωt)> = 0
<Cos (ωt)> = 0
<Sinφ> = sinφ
<Cosφ> = cosφ
As described above, the amount of received light that is time-averaged after the modulated irradiation light is modulated again at the same frequency can be expressed by three variables of unknown variables α and β and phase φ. If the phase φ is calculated by eliminating α and β from P with different conditions such as the received light quantity Pps, Pns, Ppc, Pnc, Pf, etc., the distance L can be obtained.

By the way, the relational expression relating to the above time average strictly holds when Ta is an integral multiple of To (when Tm = 0). An error occurs when Tm is not zero, but the error can be reduced when n is large. In the following, error evaluation is simply performed for an arbitrary periodic function f (t) of the period To.
Here, ∫af (t) dt is integrated in the range of t = 0 ·· Ta, ∫mf (t) dt is integrated in the range of t = 0 ·· Tm, and ∫of (t) dt means that integration is performed in the range of t = 0 ·· To, and that <f (t)> o = (∫of (t) dt) / To.
<F (t)> = (∫af (t) dt) / (nTo + Tm)
= (N∫of (t) dt + ∫mf (t) dt) / (nTo + Tm)
= (<F (t)> o + Rf / n) / (1 + Rt / n)
Where Rf = (∫mf (t) dt) / To
Rt = Tm / To
And
In summary, when <f (t)> o = 0, when n is increased, <f (t)> gradually approaches 0 in the order of O (1 / n). In the case of <f (t)> o = K, when n is increased, <f (t)> gradually approaches K in the order of K + O (1 / n).
That is, <sin ^ 2ωt> and <cos ^ 2ωt> are halved in the order of O (1 / n), and <sinωtcosωt>, <sin (ωt + φ)>, and <cos (ωt + φ)> are O (1 / Asymptotic to 0 in the order of n).
From this, it can be seen that if Ta is sufficiently long (n is sufficiently large), the measurement error can be reduced with an arbitrary accuracy. For example, if n = 1000, the error can be reduced to about 0.1%. If the phase error is about 1%, it can be said that the relational expression relating to the above time average is established with a sufficiently good approximation.
Since the modulation frequency f is given by f = c / λ for the wavelength λ = 2Lo, for example, f = 1 GHz at Lo = 0.15 [m], f = 100 MHz at Lo = 1.5 [m], When Lo = 3.0 [m], f = 50 MHz. When the phase resolution is 1% of 2π, resolutions of 1.5 [mm], 15 [mm], and 30 [mm] are obtained, respectively.
The feature of the shape measurement according to the present embodiment is that a circuit having a broadband response such as a short-time rectangular pulse is not necessary, and it is only necessary to perform sinusoidal modulation at a constant frequency of the angular frequency ω for both irradiation and imaging. However, it is necessary to strictly manage the frequency and phase. Since the light receiving element (light receiving element array) 6 of the imaging means detects the average light amount, and does not require high-speed response and high sensitivity, the consumer light receiving element array 6 can be used as it is.

In the shape measuring apparatus as the first modification of the first embodiment, in addition to the first embodiment, four types of modulation conditions of φd = 0, π / 2, π, and 3π / 2 are used. When the light quantity distribution of the obtained imaging pattern is Ppc, Pns, Pnc, Pps, the phase function S is φ = atan ((Pns−Pps) / (Ppc−Pnc)).
In this example, four types of sine waves shifted by π / 2, which are often used in a normal phase shift method, are used as the irradiation pattern. As described above, since there are three unknown variables, α, β, and φ, obtaining the phase φ from the four conditions seems to have an excessive number of imaging conditions. However, since the sin φ component is extracted from Pns-Pps and the cos φ component is extracted from Ppc-Pnc, this is a proven phase function in that the symmetrical distortion of the sine wave can be canceled and the phase can be detected stably.
The laser diode 3 emits modulated light having a light intensity P0. When the distance from the optical center of the lens to the measurement object 2 is L, the modulated light emitted from the laser diode 3 is reflected on the measurement object 2, and the amount of light is proportional to the reflectance of the surface of the measurement object 2.
The reflected light propagates by 2L while being inversely proportional to the square of the distance, passes through the optical shutter 4, and is imaged as reflected light Pr by the light receiving element (light receiving element array) 6. Ω and φ are different for each location on the surface of the measurement object 2.

The phase shift φ occurs because the reflection is delayed according to the distance of the measurement object 2, and the phase is most advanced when the light is reflected at the distance L = L1 (φ = 2π), and the light is reflected at the distance L = L2. In this case, the phase is most delayed (φ = 0).
Using the light quantity distributions Ppc, Pns, Pnc, and Pps of the imaging pattern obtained for four types of modulation conditions where φd = 0, π / 2, π, and 3π / 2, the phase function φ = atan ((Pns− Pps) / (Ppc−Pnc)), and finally the distance L = L2−Lo (φ−φo) / (2π). The phase φr is measured in advance by providing a reflector at the position L2, and φo is obtained from φo = φr−2π (L2−Lr) / Lo = φr.
Only the laser diode 3 and the optical shutter 4 have to operate at high speed in this apparatus, and it is advantageous that the number of high-speed modulation elements that tend to be expensive can be suppressed to two. Since the light receiving element array 6 is used, the distance of the surface of the measurement object 2 is simultaneously obtained for each direction viewed from each pixel through the optical center of the lens 5.
As the light receiving element array 6, a normal CCD or CMOS light receiving element array is sufficient, but linearity of a signal with respect to an incident light amount is necessary to correctly detect the phase. Since the linearity of about 1% is usually obtained in an industrial CCD, the phase error accompanying the light amount detection is about 1%, that is, the distance accuracy is about 1% with respect to L2-L1 corresponding to one modulation period. Is obtained.
Since an irradiation pattern modulated in a sine wave is averaged and detected under modulation conditions of four types of sine waves having different phases, a simple modulation pattern is obtained, so that imaging can be easily realized.

In the shape measuring apparatus which is the second modification of the first embodiment, in addition to the first embodiment, it is obtained for three types of modulation conditions with φd = 0, π, and 3π / 2. The phase function S is φ = atan ((Pnc−Pps) / (Ppc−Pps)) + π / 4, where the light quantity distribution of the imaging pattern is Ppc, Pnc, and Pps.
The second modification is obtained by reducing the irradiation pattern by one sheet and changing the phase function as compared with the first modification. Since it is sufficient to image three times, it is advantageous that the time and effort of imaging can be reduced compared to the first modification.
Theoretically, the same phase as in the first modified example should be obtained. However, if the modulation deviates from the sine wave, the phase error becomes larger than the four types of patterns in the first modified example. Since the detection effort (number of imaging conditions) and the phase error are in a trade-off relationship, which one is selected is selected depending on what the system is configured for. The configuration of the second modification is focused on reducing the imaging effort with respect to the first embodiment.
The specific configuration of the second modified example is the same as that in FIG. 1, and imaging is performed under three types of conditions at the time of imaging, the phase is obtained using the above phase function, and the distance is obtained from the phase. Since an irradiation pattern modulated in a sine wave is averaged and detected under modulation conditions using three types of sine waves having different phases, a simple modulation pattern is obtained, so that imaging can be easily realized.

In the shape measuring apparatus which is the third modification of the first embodiment, in addition to the first embodiment, the imaging pattern obtained for two types of modulation conditions with φd = 0 and π / 2 The phase function S is φ = atan ((Pns−Pf) / (Ppc−Pf), where Pf is the light amount distribution of the imaging pattern obtained for the modulation condition of Ppc, Pns, and D = 0. ).
The third modification has the same three types of imaging condition numbers as the second modification, but one of the conditions Pf is not modulated at the time of imaging, and light quantity detection is performed under these conditions. The modulation and phase management of the optical shutter 4 are not necessary.
In the second modification, three types of phase conditions must be managed, but in the third modification, only two types of phases need be managed, and a simpler circuit is sufficient than the second modification. .
When the irradiation and imaging modulation are ideal sine waves, the first, second, and third modifications can theoretically measure distance with the same accuracy, but the modulation actually deviates from the ideal sine wave. Affects the phase accuracy.
The order of the first, second, and third modified examples is the order of the highest accuracy, but the order of the imaging conditions and the number of conditions to be phase-managed are also the order of the first, second, and third modified examples. Therefore, it may be determined which of the first, second, and third modified examples is selected depending on whether accuracy is given priority or simplicity of the drive circuit is given priority.
The specific configuration of the third modified example is the same as that shown in FIG. 1, and images are taken under three types of conditions at the time of imaging, the phase is obtained using the above phase function, and the distance is obtained from the phase. Since the irradiation pattern modulated in a sine wave is averaged and detected under the modulation condition of two types of sine waves with different phases and the condition without modulation, imaging is simple because it is a simple modulation pattern. it can.

In the shape measuring apparatus which is the fourth modification of the first embodiment, in addition to the first embodiment, the modulation frequency of irradiation and imaging is nω (n is a natural number), and the phase φ1 of n = 1 is used. Find an outline of the shape. From the phase φn obtained by sequentially increasing n = 2, 3,..., L = L2−Lo (φn−2mπ + nφo) / (2πn) to obtain a detailed distance L while estimating an undetermined integer m. It is characterized by.
When the modulation frequency is 2ω, there are two periods of modulated waves between the distances L1 ·· L2, and the phase range corresponding to L1 ·· L2 is 0 ·· 4π. If the phase error of the imaging means is Δφ, the relative phase error is Δφ / (2π) in the case of a modulated wave of one cycle as in the first embodiment, while the relative phase error is Δφ / (4π) and 1/2 of one cycle.
Similarly, when the modulation frequency is nω, the relative phase error becomes Δφ / (2nπ) and 1 / n. Considering that the initial phase φo is n times because the phase range is n times, the phase becomes 2nπ with respect to the distance L1 ·· L2, and therefore the phase φ˜ { The relationship for obtaining the distance from n} is as follows.
L = L2−Lo (φ˜ {n} −nφo) / (2πn)
If the relative phase error is reduced to 1 / n, the relative distance error is also reduced to 1 / n in conjunction with this, and the position can be detected with high accuracy. For example, if the detection is performed under the condition of 2ω, the distance accuracy is improved twice.
The problem here is that φ to {n} that changes from 0 to 2nπ cannot be detected directly, and that that can be detected is φ {n} obtained by convolving φ to {n} into 0 ·· 2π. That is. When m {n} is an integer in the range of 0 ·· n−1, φ {n} is expressed by the following equation.
φ˜ {n} = φ {n} + 2m {n} π
Even if φ {n} is known, since the integer m {n} is not known, φ˜ {n} cannot be obtained uniquely. As a practical solution to solve this problem, there is a method in which the phase of a certain point on the measurement object 2 is given and the phase is specified over the entire imaging region on the assumption that the phase around the point is continuous. Are known.

In addition, when the value of n is 1, the phase φ and the distance L can be determined without ambiguity, so that the uncertain integer m {n} is estimated using this as an approximate value, and the distance accuracy is improved while increasing n. May be taken. Specific examples are shown below.
When φ {p} and m {p} are known when the modulation frequency is pω, the distance L {p} is as follows.
L {p} = L2−Lo (φ {p} + 2m {p} π−pφo) / (2pπ)
When φ {q} is known when qω, m {q} is set as an undetermined integer, and the distance L {q} is as follows.
L {q} = L2-Lo (φ {q} + 2m {q} π−qφo) / (2qπ)
Since L {p} can be regarded as an approximate value of L {q}, m {q} satisfying L {q} −L {p} = 0 is set as an estimated value of m {q}.
m {q} = (q / p) (φ {p} + 2m {p} π−pφo) / (2π) − (φ {q} −qφo) / (2π) −q (L {q} −L { p}) / Lo
Since n = 1 corresponds to the case of the first embodiment, φ {1} = φ, m {1} = 0, and L {1} = L. Based on this, φ {2}, m {2}, L {2}, φ {3}, m {3}, L {3},... Go.
The specific embodiment has the same configuration as that shown in FIG. 1, and images are taken while changing the modulation frequency at the time of imaging. An undetermined integer is obtained from the above assumption of continuity of the phase and the method of increasing the frequency sequentially, and the phase is calculated by the phase function. Find the distance from the phase.
By obtaining a more detailed distance from the phase obtained by sequentially increasing the modulation frequency of irradiation and imaging, the phase detection accuracy of the irradiation pattern is improved by n times, so that the distance resolution can be increased.
If Lo is 30 [m] and the modulation frequency is 5 [MHz] and the phase error is 1%, a distance error of 300 [mm] occurs when n = 1, but first, the outline is measured when n = 1, and n = 2 If it is increased to 4, 8, 16 (modulation frequency 10, 20, 40, 80 [MHz]), the distance error of about 19 [mm] can be achieved even in the same 30 [m] measurement range without causing phase ambiguity. Can improve.

In the shape measuring apparatus which is the fifth modification of the first embodiment, in addition to the first embodiment, the irradiation means 3 is a semiconductor light source such as a laser diode or a light emitting diode, and the imaging means 4, 5, Reference numeral 6 denotes a CCD element with a light shutter 4 or a CMOS element.
It is desirable that the irradiation unit 3 and the imaging units 4, 5, and 6 are made as easily as possible and can be made with low-cost parts. If the irradiation means 3 is a semiconductor light source such as a laser diode or a light emitting diode, and the imaging means 4, 5, 6 are CCD elements or CMOS elements with an optical shutter 4, this fifth modification can be easily configured. be able to.
As the optical shutter, a PLZT optical shutter, a car cell using the Kerr effect, or the like can be used. In particular, since a laser diode can easily generate linear modulation, there is an advantage that light loss in a polarization type optical shutter can be reduced.
Since the irradiation unit 3 is a semiconductor light source and the imaging unit is a CCD element or a CMOS element with an optical shutter 4, the device can be configured with easily available elements, so that the system cost can be reduced.

FIG. 2 is a schematic view showing a second embodiment of the shape measuring apparatus according to the present invention. In this shape measuring apparatus, in addition to the first embodiment, a reflecting plate is provided at a distance Lr in front of the irradiating means 3, and the phase φr of reflected light returning from the reflecting plate is φo = φr−2π (L2−Lr). ) / Lo to obtain φo.
Specifically, as shown in FIG. 2, a reflecting plate (reference mirror) 8 serving as a reference surface is provided at a distance of Lr from the optical center of the lens 5. Assuming that the detected phase is φr, φo = φr−2π (L2−Lr) / Lo is obtained from the relationship Lr = L2−Lo (φr−φo) / (2π).
A translucent reflecting plate (reference mirror) 8 with a known distance is provided, and the phase shift of irradiation with respect to imaging is measured from the return time of reflected light returning from the reflecting plate 8, thereby correcting the phase shift between irradiation and imaging. Therefore, the distance measurement error can be reduced.
FIG. 3 is a schematic view showing a third embodiment of the shape measuring apparatus according to the present invention. In this shape measuring apparatus, in addition to the first embodiment, the irradiation means 3 is a semiconductor light source such as a laser diode or a light emitting diode, and the imaging means 6 is a CCD element or a CMOS element capable of modulating the optical gain. Features.
During measurement, the reflector (reference mirror) 8 may be removed and the measurement object 2 may be placed in place of the reflector 8 for measurement. When there is in-plane variation in the phase response of the optical shutter 4 of the fifth modification of the first embodiment and the gain modulation circuit of the third embodiment, for each element of the light receiving element array 6 in the imaging means Alternatively, φo may be obtained.
Since recent logic LSIs operate with a clock exceeding several GHz, a modulation function at the time of imaging can be built in a CCD or CMOS substrate. Modulating the acquired light at the time of imaging is the same as modulating the gain of the light receiving element, and therefore a circuit for modulating the gain by modulating the voltage applied to each photodiode is provided.
As shown in FIG. 3, since the modulation and light reception can be performed by the gain modulation light receiving element array 6a of one chip without preparing an external optical shutter, the manufacturing cost can be reduced at the time of mass production.
That is, by using the semiconductor light source 3 as the irradiating means and the CCD element or the CMOS element capable of modulating the optical gain as the irradiating means, the apparatus can be configured with easily available elements, so that the system cost can be reduced.

FIG. 4 is a schematic view showing a fourth embodiment of the shape measuring apparatus according to the present invention. In this shape measuring apparatus, in addition to the first embodiment, the light source of the irradiating means is an infrared light source 3a, and the imaging means 4, 5, 6 are provided with an optical filter (infrared filter) 9 for removing visible light. It is characterized by.
As shown in FIG. 4, by using an infrared laser diode 3a as a light source and providing an infrared filter 9 that transmits infrared rays to the optical shutter 4, visible light incident on the lens 5 from the outside, which becomes noise, is removed, and light is emitted. Intensity measurement errors can be reduced.
By using the infrared light source 3a as the light source of the irradiating means and providing the image pickup means with the optical filter 9 that removes visible light, disturbance light can be removed, so that a distance error due to a light quantity error can be reduced.
FIG. 5 is a schematic view showing a fifth embodiment of the shape measuring apparatus according to the present invention. In addition to the fourth embodiment, the light source of the irradiating means is an infrared light source (infrared laser diode) 3a, and mirrors for separating infrared light and visible light are provided in the imaging means 4, 5 and 6, and the separated red The shape is measured from outside light, and texture data to be pasted on the shape from visible light is obtained.
As the optical filter of the fifth embodiment, a second half mirror 10 is newly provided, and the second half mirror 10 reflects visible light. The reflected visible light is imaged on the second light receiving element array 12 by the second lens 11 and picked up as a visible texture image.
The point that shape measurement is performed by the infrared laser diode 3a, the lens 5, the optical shutter 4, and the light receiving element array 6 is the same as in the fourth embodiment. The optical centers of the lens 5 and the second lens 11 are merely bent by a mirror and are at the same optical position.
For this reason, a texture image and a distance image can be associated with each pixel of the first and second light receiving element arrays (6 and 12). For example, since Pf of the third modification of the first embodiment detects the time average of the reflected light amount without modulation, it can also be used as texture data.
Using infrared and visible light separation mirrors, the shape is measured from the infrared light, and texture data to be attached to the shape from the visible light is obtained. Thereby, since shape measurement and texture image acquisition are performed with light of different wavelengths, shape measurement and texture image acquisition can be performed simultaneously.

It is the schematic which shows 1st Embodiment of the shape measuring apparatus by this invention. It is the schematic which shows 2nd Embodiment of the shape measuring apparatus by this invention. It is the schematic which shows 3rd Embodiment of the shape measuring apparatus by this invention. It is the schematic which shows 4th Embodiment of the shape measuring apparatus by this invention. It is the schematic which shows 5th Embodiment of the shape measuring apparatus by this invention. It is a block diagram explaining the conventional measuring apparatus. It is a block diagram explaining the other conventional measuring apparatus. It is a figure which shows the light intensity modulated in the linear form which influences distance accuracy. It is another figure which shows the light intensity modulated in the linear form which influences distance accuracy.

Explanation of symbols

1 Shape measuring device
2 Measurement object 3 Irradiation means (laser diode)
3a Irradiation means (infrared laser diode)
4 Imaging means (optical shutter)
5 Imaging means (lens)
6 Imaging means (light receiving element, light receiving element array)
6a Imaging means (gain modulation light receiving element array)
7 Half mirror 8 Reflector (reference mirror)
9 Infrared filter 10 Second half mirror 11 Second lens 12 Second imaging means (light receiving element, light receiving element array)

Claims (10)

  Irradiation means for irradiating a measurement object with an irradiation pattern that is a light pattern, and an imaging means that is optically in the same position as the irradiation means for obtaining an imaging pattern that is a light pattern reflected by the surface of the measurement object, In the shape measuring apparatus for calculating the surface shape of the measurement object from the imaging pattern, the measurement object exists within a predetermined distance from the irradiation unit, and the irradiation unit displays the irradiation pattern with a predetermined intensity modulated at an angular frequency ω. Irradiating the measurement object, the imaging means modulates the reflected light reflected on the surface of the measurement object and returning to the imaging means by following the same path as the irradiation in the reverse direction with the angular frequency ω and the phase φd. A shape measuring apparatus that detects a time average and obtains a distance L of the measurement object based on the detected light average based on a light amount obtained under a plurality of modulation conditions.   When the light quantity distribution of the imaging pattern obtained for the four types of modulation conditions with the phase φd being 0, π / 2, π, and 3π / 2 is Ppc, Pns, Pnc, and Pps, the phase function S is φ The shape measuring apparatus according to claim 1, wherein: = atan ((Pns−Pps) / (Ppc−Pnc)).   When the light quantity distribution of the imaging pattern obtained for the three types of modulation conditions with the phase φd being 0, π, 3π / 2 is Ppc, Pnc, Pps, the phase function S is φ = atan ((Pnc− 2. The shape measuring apparatus according to claim 1, wherein Pps) / (Ppc−Pps)) + π / 4.   The light quantity distributions of the imaging patterns obtained for the two types of modulation conditions where the phase φd is 0 and π / 2 are obtained from the imaging patterns obtained for the modulation conditions where Ppc, Pns and D = 0. The shape measuring apparatus according to claim 1, wherein the phase function S is φ = atan ((Pns−Pf) / (Ppc−Pf)) where the light quantity distribution is Pf.   The modulation frequency of irradiation and imaging is nω (n is a natural number), the outline of the shape is obtained from the phase φ1 of n = 1, and from the phase φn obtained by sequentially increasing n, L = L2−Lo (φn−2mπ + nφo ) / (2πn) to obtain the distance L in detail while estimating an undetermined integer m.   A reflector is provided at a distance Lr in front of the irradiating means (L1 ≦ Lr ≦ L2), and φo is obtained from φo = φr−2π (L2−Lr) / Lo as the phase φr of the reflected light returning from the reflector. The shape measuring apparatus according to claim 1.   2. The shape measuring apparatus according to claim 1, wherein the irradiating means is a semiconductor light source, and the imaging means is a CCD element with a light shutter or a CMOS element.   2. The shape measuring apparatus according to claim 1, wherein the irradiating means is a semiconductor light source such as a laser diode or a light emitting diode, and the imaging means is a CCD element or a CMOS element capable of modulating optical gain.   2. The shape measuring apparatus according to claim 1, wherein the light source of the irradiating unit is an infrared light source, and an optical filter for removing visible light is provided in the imaging unit.   An infrared light source is used as the light source of the irradiating means, and a mirror for separating infrared light and visible light is provided on the imaging means, the shape is measured from the separated infrared light, and texture data to be attached to the shape from the visible light is obtained. The shape measuring apparatus according to claim 9.

Images