KR20080032680A - Optical system and method for measuring shape of three dimensions using the system - Google Patents

Abstract

An optical system and a three-dimensional shape measuring method using a picture obtained from the optical system are provided to inspect defects accurately when measuring the three-dimensional shape of a semiconductor or minute fine components by acquiring a plurality of light interference patterns and measuring the three-dimensional shape. An optical system comprises a phase transfer unit(300), a light interference unit(400), a light splitting unit(500) and an image acquiring unit(600). The phase transfer unit moves a standard surface. The light interference unit reflects light from an illumination unit(200) in the standard surface and generates synthesized interference light. The light splitting unit splits the synthesized interference light from the light interference unit into a plurality of different wavelengths and outputs it. The image acquiring unit receives the split lights of different wavelengths, and acquires a plurality of light interference patterns as an image.

Description

Optical system and method for measuring shape of three dimensions using the system}

1 is a configuration diagram of an optical system for measuring a conventional three-dimensional shape,

2 is a block diagram of an optical system according to an embodiment of the present invention;

3A to 3D are graphs showing the beat phase calculation process obtained in the stereoscopic image measurement method;

4A to 4B are graphs showing a phase order calculation process of beat phase obtained in a stereoscopic image measuring method.

* Explanation of symbols for main parts of the drawings

200: lighting unit 300: phase transfer unit

400: optical interference unit 500: light division unit

600: image acquisition unit

The present invention relates to a system and method for measuring conformation.

In general, it is well known to use interferometric methods to inspect the surface of an object and measure the undulation of the object for defects such as semiconductors, flat panel displays and micro precision components. This method consists of generating an interferometer pattern on the surface of the object and analyzing the derived interferometer image (or interferogram) to obtain the solid shape of the object. Interferometer images generally include a series of black and white fringes.

       An interferometer method that requires the use of a laser to generate an interferometer pattern is called a "classic interferometer method." In this classical method, the wavelength of the laser and the configuration of the measuring assembly generally determine the period of the resulting interferogram. Classic interferometer methods are commonly used in the visible spectrum to measure changes in altitude on the order of microns.

1 is a block diagram of a device for measuring a conventional three-dimensional shape,

As shown in FIG. 1, shape information of a fine thin film layer is measured using an acoustic optical modulation filter.

    1 is a structure in which a Michelson interferometer module having a built-in blocking surface and an acoustic optical modulation filter capable of scanning in the visible light spectral region range are combined. This structure has the advantage of independently measuring the thickness and surface shape information of the thin film by turning on / off the blocking surface.

     As shown in FIG. 1, the apparatus of FIG. 1 applies an acoustooptic modulator filter (hereinafter abbreviated AOTF) 40 to a white light interference system to a transparent micro thin film layer 83 applied over an opaque metal layer 82 pattern. The measurement unit 80 having the multi-layer structure may be configured to separately measure information about the thickness and shape information using the interference phenomenon of monochromatic light.

       AOTF 40 is a kind of engineering band-pass filter, which acts as a diffraction grating for monochromatic light (hereinafter referred to as white light) incident on an acoustic optical crystal plane, and selects only a specific wavelength satisfying a condition. It is an optical filter with very narrow bandwidth.

       The light source 10 to which white light is emitted is a tungsten-halogen lamp of about 70W, and one side of the single mode optical fiber 11 is connected in the emission direction to transmit the emitted white light to the other side of the optical fiber 11.

       A fixing member 12 is positioned on the other side of the optical fiber 11, and the other side of the optical fiber 11 is correspondingly connected to a pinhole in the center. The white light emitted through the pinhole is spread around the pinhole.

       At this time, it is positioned in the first convex lens 13 at a predetermined distance in the emission direction of the white light which is the front surface of the fixing member 12. The white light is aligned at a constant width while passing through the first convex lens 13. The white light transmitted through the first convex lens 13 is incident on the first beam splitter 20 positioned at a predetermined distance from the first convex lens 13. The first beam splitter 20 is in the form of a non-polarized cube that can separate incident white light at a ratio of 50:50, and separation is sequentially performed according to a measurement process, not simultaneously.

       Since the reflection angle of the first beam splitter 20 is about 45 degrees with respect to the incident direction of the white light, the reflected white light is reflected perpendicularly to the incident direction. The second convex lens 31 is positioned corresponding to the reflection angle of the first beam splitter 20. The second convex lens 31 has a position opposite to that of the first convex lens 13. That is, in contrast to aligning the white light transmitted by the first convex lens 13 to have a constant width, focusing is performed so that the width of the transmitted white light is collected at one point along the direction of travel. The second beam splitter 32 is where the white light passing through the second convex lens 31 is collected at one point, that is, in focus. At this time, a part of the white light reaching the second beam splitter 32 is reflected toward the reference plane 33, and the other part is transmitted as it is and irradiated to the measuring unit 80.

       At this time, the blocking plate 34 is positioned on the front surface of the reference surface 33 at a predetermined interval. Blocking play is parallel to each other at a position close to the reference plane 33, and selectively blocks the white light incident on the reference plane 33.

       As such, the system consisting of the second convex lens 31, the second beam splitter 32, and the reference plane 33 is the Michelson interference module 30, and the blocking plate 34 is included in the blocking plate 34. Depending on the selective white light blocking), it operates in two modes.

       In addition, the white light which is separated by the second beam splitter 32 and incident on the reference plane 33 and the measuring unit 80 is irradiated to the measuring unit 80 to cause a change in wavelength.

       This change is caused by having shape information and thickness information, and the respective information can be measured separately according to each mode according to the operation of the blocking plate 34.

       The white light that has been irradiated in this way is reflected again and passes through the second beam splitter 32, and then the traveling width is aligned again while passing through the second convex lens 31. The first beam splitter 20 passes through the first beam splitter 20 and is incident on the AOTF 40 facing the second convex lens 31 with the first beam splitter 20 therebetween.

       As mentioned earlier, the AOTF 40 selectively scans only a short wavelength of a specific band. The AOTF 40 used in this example has a non-collinear type, and the filtering range is about 400 nm to 650 nm. The resolution is about 1 nm to 5.1 nm.

       The AOTF 40 filters only white light having an acoustic optical characteristic, that is, white light (or short wavelength light) in a band containing thickness information or shape information to be separated from white light in another band. The AOTF 40 is composed of an acoustic optical absorber, a driving device, and an acoustic optical crystal plane to which white light is irradiated. The crystal surface is composed of crystal crystals, and when the white light is irradiated, the refractive index changes by the acoustic wave generated by the driving device. Happens.

       At this time, a moving three-dimensional diffraction grating is formed on the surface of the crystal plane, so that the irradiated white light hits the diffraction phenomenon. Then, the spectral image, that is, the spectral image, can be obtained by separating in parallel. In this case, when the white light separated by the AOTF 40 is divided into + 1st and -1st order, -1st order is selected, and +1 The car is extinguished by hitting and absorbing the blocking member 60 provided.

       The CCD sensor 70 has a specification that the number of pixels is about 752 x 582, and one pixel size is about 11.1 mu m x 11.2 mu m. The third convex lens 50 is provided in the traveling direction of the white light thus selected. The white light passing through the third convex lens 50 is imaged by focusing and focusing the CCD sensor 70.

       The white light formed on the CCD sensor 70 may be obtained as a spectral image, and the information may be extracted by scanning it. In addition, shape information on the surface of the micro thin film layer 83 in the measuring unit 80 may be finally obtained using the maximum point information obtained from each extracted information.

     However, since the white light of a certain wavelength band is selected and measured using an acoustic optical modulation filter, the filtering range and resolution of the acoustic optical modulation filter have a great influence on the performance of the system, and the real-time measurement is a problem of selectively scanning short wavelengths of a specific band. Weak against external vibration and

Accordingly, an object of the present invention is to provide a system and method for measuring a three-dimensional shape of an object by obtaining at least two or more optical interference patterns formed with a plurality of wavelengths to solve the above problems.

The optical system according to the present invention for achieving the above object is a phase transfer unit for moving the reference plane in the optical system for obtaining the interference fringe by reflecting the light emitted from the illumination unit including the light source in the reference plane and the measurement plane And an optical interference unit for generating synthetic interference light by reflecting light from the illumination unit on a measurement plane and a reference plane, a light splitting unit for dividing the composite interference light from the optical interference unit into a plurality of different wavelengths, and a light splitting unit. And an image acquisition unit configured to receive light having different wavelengths and obtain a plurality of optical interference patterns as an image.

According to the present invention, the optical interference unit includes a beam splitter for separating incident light from the illumination unit into the reference plane and the measurement plane of the measurement object, and a lens for synthesizing the reflected light from the reference plane and the measurement plane and outputting the synthetic interference light. do.

In addition, according to the present invention, the light splitter is configured to correspond to a lens that passes through the composite interference light from the optical interference unit, a plurality of beam splitters that reflect and transmit the composite interference light to the image acquisition unit, and a front end of the image acquisition unit. It is characterized in that it comprises a plurality of filters positioned to pass only light of different wavelengths.

In addition, according to the present invention, the image acquisition unit is characterized by consisting of at least one camera.

In addition, according to the present invention, the filter is disposed in front of the camera to filter the composite trunk light of different wavelengths corresponding to the number of cameras of the image acquisition unit.

A three-dimensional shape measuring method according to the present invention for achieving the above object is a method for measuring a three-dimensional shape with an image obtained by a camera having an array of pixels, a plurality of optical interference patterns formed of a plurality of different wavelengths A first step of obtaining, a second step of calculating the phases of the obtained plurality of optical interference patterns, a third step of calculating the beat phase using the optical interference pattern phases for the corresponding pixels, and a phase order of the beat phases And a fifth step of calculating a height value for each pixel by using the wavelength obtained by the optical interference pattern and the beat phase and the phase order.

Further, according to the present invention, the second step is characterized by calculating the phase of the optical interference pattern of each pixel using at least three intensity values for each pixel of the plurality of optical interference patterns obtained from the camera.

Hereinafter, with reference to the accompanying Figures 2 to 4 will be described in detail a preferred embodiment of the present invention.

2 is a block diagram of an optical system according to an embodiment of the present invention.

The optical system of FIG. 2 includes a phase transfer unit 300 for moving the illumination unit 200 and the reference plane, and an optical interference unit 400 for generating synthetic interference light by reflecting light from the illumination unit 200 at the measurement plane and the reference plane. ) And a light splitting unit 500 for dividing the synthesized interference light from the optical interference unit 400 into different wavelengths, and receiving light having different wavelengths and splitting the light into three images. It includes an image acquisition unit 600 for.

The lighting unit 200 functions to provide a light source, align the light sources, adjust the amount of light, and block unnecessary light included in the light source. Although not shown, the lighting unit 200 includes a light source, a lens for aligning the light source and adjusting the amount of light, a reflector and an aperture, an optical fiber used to guide the light source, and an optical filter for blocking light of wavelengths unnecessary for measurement. do. In this case, when the distance between the light source and the optical interference optical system is close, the optical fiber may not be used. In addition, according to the optical characteristics of the light source, the lens may not be necessary when the light source does not need to be aligned, and the aperture may not be necessary when adjusting the amount of light in the light source itself. In addition, an optical filter may not be necessary depending on the constituent wavelength of the light source. The lighting unit 200 generally uses a white light source having light of various wavelengths, and a halogen lamp or a white LED lamp is used.

The phase transfer unit 300 is a device for transferring the reference plane 402 to a known position parallel to the traveling direction of light in order to transfer the phase of the interference fringe formed in the optical system. The phase transfer unit 300 is a precision plate spring structure is used as the transfer guide surface, the piezoelectric actuator (PZT) is used as the precision transfer drive. In addition, there are cases where a distance sensor (gap, sensor, LVDT, strain gauge, etc.) for detecting the transferred position is used, which is used for the transfer to a more accurate position.

The optical interference unit 400 is configured to split the incident light from the illumination unit 200 into the reference plane 402 and the side front surface A of the measurement object, and the reflected light from the reference plane 402 and the measurement plane A. And a lens 403 which is synthesized and emitted as synthetic coherent light. That is, the beam splitter 401 of the optical interference unit 400 divides the light incident from the illumination unit 200 into two lights, one of which is emitted toward the reference plane 402, and the other is measured. The reference light reflected from the reference plane 402 and the measurement light reflected from the measurement plane A are combined to exit toward the measurement plane A of the object and emitted as a synthetic interference light.

The light splitter 500 guides the synthesized coherence light emitted from the optical interference unit 400 to the image acquisition unit 600 having three cameras, so that the interference pattern is clearly formed in the image acquisition unit 600. Light of three different wavelengths is incident on the three image acquisition units 600 separately. The light splitter 500 of FIG. 2 includes a beam splitter 502 and a beam splitter 502 reflecting the composite interference light toward the lens 501 and the first camera 604 that pass through the composite interference light from the optical interference unit 400. The beam splitter 503 reflects the composite interference light toward the two cameras 605 and transmits the composite interference light toward the third camera 606, and is located at the front end of each camera 604, 605, and 606, and has different wavelengths. And first and third filters 504, 505, and 506 that pass through only the light. The composite coherence light emitted from the optical interference unit 400 reflects 1/3 of the composite coherence light in the first beam splitter 502 and passes through the first filter 504 to the first camera 604 of the image acquisition unit 600. To the side. The beam splitter 502 then transmits the remaining two thirds of the composite coherent light. At this time, the second beam splitter 503 reflects 1/2 of the transmitted composite interference light (1/3 of the total synthesized light) and passes through the second filter 505 toward the second camera 605 of the image acquisition unit 600. Induce. The beam splitter 503 guides the remaining synthesized interference light (1/3 of the total synthesized light) toward the third camera 606 of the image acquisition unit 600 through the third filter 506.

The image acquisition unit 600 includes first to third cameras 604, 605, and 606, and is disposed corresponding to the first to third filters 504, 505, and 506 of the light splitter 500. The first to third cameras 604, 605, and 606 obtain an image of the optical interference pattern from the synthetic interference light filtered by different wavelengths of the first to third filters 504, 505, and 506. Here, the first to third cameras 604, 605, and 606 have an array of pixels, which are in the form of a CCD camera or a CMOS camera. Such a camera provides a resolution of, for example, 1300 X 1024 pixels. In this way, three optical interference patterns acquired by the image acquisition unit 600 are obtained as images.

In the optical system disclosed in the embodiment of FIG. 2, three filters and three cameras that emit light of different wavelengths correspond to each other in order to obtain three optical interference patterns. It is also possible to configure three optical interference patterns by moving and replacing three filters and to obtain three optical interference patterns from a camera by emitting light of different wavelengths from the lighting unit. Do. In addition, it is apparent that the three optical systems described above may be variously configured to obtain two optical interference patterns.

A stereoscopic image measurement method using three optical interference patterns obtained in the embodiment of FIG. 2 will be described in detail with reference to FIGS. 3 to 4.

Hereinafter, general theory required in the description of the stereoscopic image measuring method according to the present invention will be described by Equations 1 and 2 below.

The intensity I (x, y) for all the pixels x, y of the interferometic image can be expressed by the following equation (1).

Equation 1

I (x, y) = A (x, y) + B (x, y) cos [2πM (x, y) + Φ (x, y)]

Here, x and y are pixel coordinates in the camera, A is the average brightness irrelevant to the interference fringe, B is 1/2 of the change in brightness of the interference fringe, and M has an integer value as the phase order of the interference fringe. In addition, Φ is a wrapped phase of the interference fringe and has a value between + π and -π, and 2πM + Φ is the unwrapped phase of the interference fringe.

In Equation 1, the phase order M has an integer value and the interference fringe is not affected by the change in the phase order M because the cosine function is multiplied by 2π. In order to know the phase Φ of the interference fringe in Equation 1, three unknowns A, B, and Φ included in Equation 1 should be calculated. For this purpose, three or more simultaneous equations are obtained by giving the phase change amount ΔΦ _n to a known value, and three unknowns A, B, and Φ are calculated using the equation. This calculation process is mainly used a numerical method, such as the least squares method, which is within the range that can be thought of by those skilled in the art.

       If the height value of the measurement object in the equation (1) H, H is expressed as in the equation (2).

       Equation 2

       H (x, y) = [2πM (x, y) + Φ (x, y)] * λ / (4π)

Is the wavelength of the light source for forming the interference fringe.

       In order to obtain the height value H using Equation 2, the phase order M of the interference fringe must be known, but since the height order H is not known, the height value H of the measurement object is measured only when M = 0.

       This will be explained by taking an example of a practical case. Since the phase Φ (x, y) of the interference fringe has a value between + π and -π when the wavelength of the used light source is 600 nm, the height value H eventually becomes-. Only shapes between 150 nm and +150 nm can be measured.

Equations 1 and 2 described above are applied to the stereoscopic image measuring method of the present invention.

First of all, for convenience of explanation, an interference pattern of light having a first camera 604 as Ca and having a center wavelength of λa is obtained. An interference pattern of light having a second camera 605 as Cb and having a center wavelength of λb is obtained. Assume that 3 camera 606 is Cc and an interference fringe of light having a center wavelength of λc is obtained. In addition, let the first filter 504 be Fa, the second filter 505 be Fb, and the third filter 506 be Fc.

 These interference fringes form contours at intervals corresponding to one-half of each center wavelength in the height direction (h direction) of the workpiece.

       In the interference fringes formed on the cameras Ca, Cb, and Cc, the brightness value of each pixel is represented by Equation 3 below.

       Equation 3

       Ian (x, y) = Aa (x, y) + Ba (x, y) cos [2πMa (x, y) + Φa (x, y) + ΔΦan], n = 1, 2, 3, ..

Ibn (x, y) = Ab (x, y) + Bb (x, y) cos [2πMb (x, y) + Φb (x, y) + ΔΦbn], n = 1, 2, 3, ..

    Icn (x, y) = Ac (x, y) + Bc (x, y) cos [2πMc (x, y) + Φc (x, y) + ΔΦcn], n = 1, 2, 3, ..

Where a, b and c denote cameras Ca, Cb and Cc, subscript n denotes a known position for phase-shifting the interference fringe, x and y are pixel coordinates in the camera, and A is the interference fringe The average brightness is irrelevant, B is half of the change in brightness of the interference fringe, indicating the visibility of the interference fringe, and M has an integer value as the phase order of the interference fringe. Also, Φ: has a value between + π and -π as the phase of the interference fringe, ΔΦ _n is the phase transfer amount of the interference fringe, and 2πM + Φ is the unwrapped phase of the interference fringe. .

       In order to know the phase Φ of the interference fringe in Equation 3, three unknowns A, B, and Φ included in Equation 3 must be calculated.

To this end, at least three or more simultaneous equations must be obtained by shifting the phase to a known position (ΔΦ _n ), which is implemented in one embodiment by using the phase shifter 300 to make the reference plane 402 perpendicular to the reference light. The image pickup unit 600 acquires the interference fringes Ia2, Ib2, and Ic2 whose positions are shifted in the directions and thus change in phase. It is assumed that once again the position is moved to another position, and the interference pattern Ia3, Ib3, or Ic3 whose phase is changed in the image acquisition unit 600 is acquired.

       In this process, the first camera 604 (Ca) of the image acquisition unit 600 obtains three phase shifted images Ia1, Ia2, and Ia3 of the interference fringe having a center wavelength of λa, and the second camera 605: Cb. Shows three phase-shifted images (Ib1, Ib2, Ib3) of the interference fringe having a center wavelength of λb. In the third camera (606: Cc), three phase-shifted images (Ic1, Ic2) of the interference fringe having a central wavelength of λc are obtained. , Ic3).

       In order to calculate the three unknowns included in Equation 3, only the image of three interference fringes having a minimum phase shift is required.However, in order to reduce the influence of noise during measurement, four or more phase shifted interference fringes may be used. This may cause a problem of lengthening the measurement time and increasing the calculation amount, and it should be properly adjusted according to the measurement conditions.

       At least three interference fringes obtained from each camera are used to obtain phase Φa (x, y) of an interference fringe having a wavelength of λa in the camera Ca, and a phase of the interference fringe having a wavelength of λb in the camera Cb. (phase) φ b (x, y) is obtained, and in the camera Cc, phase Φ c (x, y) of an interference fringe having a wavelength of λ c is obtained.

       Next, a process of obtaining the beat phase Φ 1 due to the beats of λa and λb and the beat phase Φ 2 due to the beats of λb and λc will be described with reference to FIGS. 3A to 3D.

       For convenience of explanation, it is assumed that the relationship is λa <λb <λc.

       3A to 3D are graphs showing the beat phase calculation process obtained in the stereoscopic image measurement method.

       The graph of FIG. 3A shows that the interference fringe phases Φa (x, y) and Φb (x, y) have values between + π and -π depending on the height of the workpiece. In addition, the interference pattern phase? A (x, y) having a short wavelength has a short period (1/2 of? A), and the interference pattern phase? B (x, y) having a long wavelength has a long period (1 / time of? B). Show that you have 2).

       In the graph of FIG. 3B, a difference value between the interference fringe phases Φa and Φb is obtained.

       The difference between the interference fringe phase Φa and Φb, that is, the beat phase Φ1 (x, y) is calculated by the following equation (4).

       Equation 4

       Φ1 (x, y) = Φa (x, y) -Φb (x, y)

Here, if φ1 (x, y) <− π, (Φ1 (x, y) + 2π) is used to make this result value Φ1 (x, y).

If φ1 (x, y)> + π, then (Φ1 (x, y) -2π) is used as the result of Φ1 (x, y).

       Φ1 (x, y) thus obtained is represented by the graph of FIG. 3C, which is a phase of wavelength λ1 due to beating of wavelengths λa and λb.

       Here, the wavelength lambda 1 by the beat of the wavelength lambda a and lambda b is calculated by the following equation (5).

       Equation 5

       λ1 = (λa X λb) / | λa -λb |

Here, lambda a is the pass center wavelength of the filter Fa provided in front of the camera Ca, and lambda b is the pass center wavelength of the filter Fb provided in front of the camera Cb.

The process of obtaining the beat phase Φ 2 by the beat wavelength λ 1 of λ b and λ c is also performed in the same manner as the process of finding Φ 1, which is obtained by Equation 6 below.

       Equation 6

       Φ2 (x, y) = Φb (x, y) -Φc (x, y)

Here, if φ2 (x, y) <− π, (Φ2 (x, y) + 2π) is made and the resulting value is denoted by Φ2 (x, y).

If Φ2 (x, y)> + π, we use (Φ2 (x, y) -2π) to make this result Φ2 (x, y).

       The result of Φ 2 thus obtained is shown in the graph of FIG. 3D, which is a phase of wavelength λ 2 due to beating of wavelengths λ b and λ c.

       Here, the wavelength lambda 2 by the beat of the wavelength lambda b and lambda c is calculated by the following equation.

       Equation 7

       λ2 = (λb X λc) / | λb -λc |

Here, lambda b is the pass center wavelength of the filter Fb provided in front of the camera Cb, and lambda c is the pass center wavelength of the filter Fc provided in front of the camera Cc.

       Next, a method of calculating the phase order of the phase Φ 1 of the beat wavelength λ 1 (the beat wavelength of λ a and λ b) and the phase Φ 2 of the beat wavelength λ 2 (the beat wavelength of λ b and λ c) will be described with reference to FIG. 4.

4A to 4B are graphs showing a phase order calculation process of beat phase obtained in the stereoscopic image measuring method.

The graph of FIG. 4A shows the change depending on the height of the beat phase Φ 1 of the beat wavelength λ 1 and the beat phase Φ 2 of the beat wavelength λ 2 obtained above. The beat phase Φ 1 shows a value between + π and -π with a half of λ 1, and the beat phase Φ 2 shows a value between + π and -π with a half of λ 2.

       The graph of FIG. 4B shows the beat phase order N1 (x, y) which is determined in such a way that the beat phase Φ 1 is increased or decreased by one every time one cycle is repeated.

       The graph of FIG. 4C shows the beat phase order N2 (x, y), which is determined in such a way that the beat phase Φ 2 is increased or decreased by one each time one cycle is repeated.

       If the difference between the two beat phases Φ1 (x, y) and Φ2 (x, y) is Φ12 (x, y), the following equation 8 is calculated.

       Equation 8

       Φ12 (x, y) = Φ1 (x, y) -Φ2 (x, y)

Here, if phi 12 (x, y) <− π, (phi 12 (x, y) + 2π) is performed and the resultant value is phi 12 (x, y).

If Φ12 (x, y)> + π, then (Φ12 (x, y) -2π) is used as the result of Φ12 (x, y).

       It can be seen that Φ 12 (x, y) thus obtained is a phase of wavelength λ 12 due to beating of wavelengths λ 1 and λ 2.

       Here, the wavelength λ 12 by the beat of the wavelengths λ 1 and λ 2 is calculated by the following equation (9).

       Equation 9

       λ12 = (λ1 X λ2) / | λ1 -λ2 |

Here, λ1 is the beat wavelength of λa and λb calculated by Equation 5, and λ2 is the beat wavelength of λb and λb calculated by Equation 7.

       If at any point the height H (x, y) value is in the range of 1/2 of the beat wavelength λ12, then this is done using Φ12 (x, y), Φ1 (x, y) and Φ2 (x, y). It can be expressed by the following equation (10).

       Equation 10

       H (x, y) = Φ12 (x, y) X λ12 / (4π)

       = [2πN1 (x, y) + Φ1 (x, y)] X λ1 / (4π)

       = [2πN2 (x, y) + Φ2 (x, y)] X λ2 / (4π)

When N1 and N2 are obtained from Equation 10, they are arranged as Equation 11 and Equation 12 below.

       Equation 11

       N1 (x, y) = [(λ12 / λ1) X Φ12 (x, y)-Φ1 (x, y)] / (2π)

       Equation 12

       N2 (x, y) = [(λ12 / λ2) X Φ12 (x, y)-Φ2 (x, y)] / (2π)

       The values of phase orders N1 and N2 calculated here should be theoretically accurate integer values, but in practical cases, they are usually calculated as real numbers rather than integers due to the noise of the system. The integer value closest to this real value determines the values of N1 and N2.

       The height values H (x, y) of the measurement object can be obtained by the following equation 13 using the beat phases Φ1, Φ2 and the phase orders N1, N2.

       Equation 13

       H1 (x, y) = [2πN1 (x, y) + Φ1 (x, y)] * λ1 / (4π)

       Equation 14

       H2 (x, y) = [2πN2 (x, y) + Φ2 (x, y)] * λ2 / (4π)

       H1 (x, y) and H2 (x, y) obtained in equations (13) and (14) should have the same value in theory, but generally have values that are not the same due to the vibration of the measurement or the noise of the system. to be. Therefore, to reduce this effect, it is advantageous to average these two values and determine the final height value. However, when a reduction in the time required for calculation is required, only the N1 value may be calculated, and only the H1 (x, y) value may be calculated.

       For example, in the case of λa = 600nm, λb = 620nm, λc = 640nm, the beat wavelength λ1 of λa and λb is calculated to be 18,600 nm by Equation 5, and the beat wavelength λ2 of λb and λc is 7 is calculated to be 19,840 nm, and the beat wavelength λ 12 of λ 1 and λ 2 is calculated to be 297,600 nm by equation (9).

       Therefore, in the exemplary embodiment of the present invention, it can be seen that the height measurement area is expanded to 148,800 nm, which is 1/2 of λ 12. It can be seen that it is greatly expanded compared to about 300 nm in the height measurement area possible with the prior art.

Therefore, the apparatus of the present invention obtains a plurality of optical interference patterns to measure the shape of a stereoscopic image, thereby providing an effect of accurately inspecting defects when measuring a three-dimensional shape of a more precise semiconductor or a micro-precision component.

Claims (7)

An optical system for obtaining an interference fringe by reflecting light emitted from an illumination unit including a light source on a reference plane and a measurement plane, A phase transfer unit for moving the reference plane; An optical interference unit for reflecting light from the illumination unit on a measurement plane and a reference plane to generate a composite interference light; A light splitting unit dividing and outputting the synthetic interference light from the optical interference unit into a plurality of different wavelengths; And And an image acquisition unit configured to receive the light having different wavelengths and obtain a plurality of optical interference patterns as an image. The method of claim 1, The optical interference portion A beam splitter separating the incident light from the illumination unit into a reference plane and a measurement plane of the measurement object; And a lens for synthesizing the reflected light from the reference plane and the measurement plane and emitting the synthesized interfering light. The method of claim 2, The light splitting part A lens that passes through the composite coherent light from the optical coherence unit and emits light; A plurality of beamsplitter for reflecting and transmitting the synthesized interference light to the image acquisition unit; And And a plurality of filters positioned corresponding to the front end of the image acquisition unit and passing only light having different wavelengths.        The method of claim 3, wherein        Video acquisition unit        An optical system comprising at least one camera. The method of claim 4, wherein The filter is And an optical system arranged at a front end of the camera so as to filter the composite trunk light having different wavelengths corresponding to the number of cameras of the image acquisition unit. A method for measuring three-dimensional shape with an image obtained by a camera having an array of pixels, A first step of obtaining a plurality of optical interference patterns formed at a plurality of different wavelengths; Calculating a phase of the obtained plurality of optical interference patterns; A third step of calculating a beat phase using the optical interference pattern phase for the corresponding pixel; A fourth step of calculating a phase order of the beat phase; And And calculating a height value for each pixel by using the wavelength obtained by the optical interference pattern, the beat phase, and the phase order. The method of claim 6, The second step is And calculating a phase of the optical interference pattern of each pixel by using at least three intensity values for each pixel of the plurality of optical interference patterns obtained from the camera.

Images