DE10321887A1 - Optical sensor for high speed scanning of an object, e.g. for use in optical metrology, has a light source and a focussing objective that generates an astigmatic image on a sensor camera chip - Google Patents

Abstract

Optical sensor has a light source (1) with at least one wavelength, a focussing objective (4) that is configured to be astigmatically imaging and a camera chip (9) placed in the beam path of the optical system. The optical system ensures that a focus figure, point or line from the object surface is imaged astigmatically on the camera chip. Also in the beam path is a lateral shear generating optical beam splitter. Independent claims are also included for the following:- (a) a method for imaging a narrow focal line in object space and; (b) a method for imaging a focal point in object space.

Description

That in the scriptures US 4,976,504 and the conoscopic method described in WO 99/42908 permits the measurement of the micro-profile of technical surfaces without mechanically moving parts in the sensor head. One or more birefringent crystals and several polarization-optical components are used in the sensor head. Such sensors with the names ConoProbe and ConoLine are also described in the brochures of Optimet Optical Metrology Ltd., www.optimet.com.

With one according to Scripture US 5,953,137 executed sensor, it is quite expensive to obtain a linear relationship between the defocusing of the object and the phase. The evaluation of the interference figures, which as a curved interference pattern, that is, as an annular stripe also in the arrangements US 4,976,504 and WO 99/42908 occur, can usually not be carried out extremely quickly.

A high accuracy is the ConoLine, the line-by-line sensor, only given in the center of the focus line. In addition, the depth measurement accuracy the ConoLine is significantly less than the ConoProbe, which is point by point measuring sensor. Through interactions between the object surface and the light can do this principle because of the use of light with different polarization directions but also influences the depth measurement accuracy due to polarization-optical effects come.

The The goal is a robust, yet accurate measurement, less expensive Sensor, in particular for determining the micro-profile of a surface, in particular along a line, the measurement being made using evaluation of a single picture. The sensor based of the inventive method, the profile should be in a wide scale range can detect. The sensor must also be designed without any mechanically moving parts. The Working distance should be adapted to the measuring task and quickly and not too small with moving material, so can also be a few millimeters.

A So the job is to get the profile and the micro profile fast and with great Accuracy in an extended depth measurement range on an object to measure absolutely along a line. The method is said to be more interference-optical Be nature. in principle Short-pulse light sources are also to be used to illuminate the object surface can be. This involves, for example, pulse lengths of the light sources in the area from 0.1 microsecond to 1000 microseconds.

A Another task is to make the layer thickness transparent To determine layers.

The known and undesirable 2n jumps that is known to be involved in the evaluation of interferograms Streaks can result should be in the inventive solution in principle do not occur. The achievable accuracy should be about correspond to that with point-by-point optical sensors same lens parameters. The achievable depth measurement accuracy for example for a correspondingly short focal length lens over the entire line length be 5 μm. The numerical depth resolution of the sensor can be at 1 μm or even below it.

The Object, for example a workpiece with a weld seam The target profile and microcracks to be checked should be used when measuring can be moved relatively quickly, for example with a Speed of more than 10 mm / s. The accuracy of the guidance of the transport system, especially the long period straightness, but should not be there are extreme requirements, especially not if the micro profile transverse to the direction of movement is of interest.

to The focusing lens should adapt to the respective measuring task exchanged easily and without adjustment for one with a different focal length can be whereby the depth measurement range and depth sensitivity or The area of layer thickness measurement by the user is optimal can be adapted to the measurement task.

Description of the invention

The Sensor effect is based on the interference of two coherent light beams Lateral shear. in principle the sensor can be designed as a point or line sensor, wherein here in particular the linear profile sensor is described, however, the point sensor is always included as a special case. The point sensor is particularly suitable for measuring the layer thickness.

The basic idea of the invention now consists in a two-beam interferometer with an average optical path difference of zero with an astigmatic image of a focus line projected onto the object or a focus point onto a camera chip by means of an in the overall optical system of the Sensor-set lateral shear between two coherent sub-bundles to generate a shear stripe pattern with an at least approximately constant spatial frequency in at least one line of the camera chip of a camera. The lateral shear can be fixed or made variable depending on the measuring task by optoelectronic means such as LCDs or LCOS or DMDs and can thus be optimally adapted to the respective measuring task. The coherent bundles with lateral shear are created by an optical beam splitter that generates lateral shear with a layer beam splitter or a diffraction grating in the beam path. The bundles of rays, which spread out in a plane parallel to the focal line, depict the focal line sharply on the camera chip.

Becomes Not a focal line, but a focal point, becomes a reference plane defines that the plane of incidence of the laser beam on one in the sensor arrangement corresponds to the existing divider layer. In this case, make up the bundles of rays that spread in a plane perpendicular to the reference plane, the Focus point sharp on the camera chip. The bundles of rays in the reference plane, on the other hand, depicts the focus point out of focus. In the figures shown always correspond to the reference plane Plane. This is the light figure on the camera chip in the reference plane of greater extent.

It becomes the spatial frequency of the shear stripe pattern or the phase response over a reference width 2 B determined in at least one line of the camera chip. This line is preferably in the defined reference plane. This determination of the spatial frequency can be carried out comparatively simply and quickly, for example, by performing the discrete Fast Fourier Transformation (FFT). The difference between the spatial frequency detectable by an object point and the spatial frequency of the same object point in the focus position is approximately proportional to the defocusing, at least in a limited depth range. Defocusing means the distance of an object point from the exact focus position or focus position of the projected laser line. This distance is the quantity to be measured here. The focus position of the laser line can, but does not have to, coincide with the focal plane of a focusing lens. The determination of the distance of object points in a line from the focus position of the laser line should be determined at the same time in order to determine the line profile in this line. Each object point of the surface to be measured is assigned a line on the camera chip, which line can be a row or column of the camera chip.

generally known can do the Fourier transform of an interference fringe pattern but also by means of an optically addressable spatial light modulator (OASLM) with which the center of gravity or centers of gravity bundle deflected by diffraction a readout laser in the Fourier plane of a lens be determined by a camera. This was done in 1998 at the Institute for Technical University optics Stuttgart by K.-P. Proll under the supervision of Prof. H. J. Titiani examined.

at of this invention addresses depending on the depth position the spatial frequency variable shear stripe pattern of an object point the OASLM. The advantage here is the use of fast OASLMs, who work in the KHz range. So a comparatively fast one execution the Fourier transformation with optical means.

The Illumination of the object surface takes place with a light source with at least one light wavelength, for example a cw laser or a pulsed laser, and with at least one wavefront-forming lens and the focusing lens already mentioned.

Becomes a high-frequency pulsed laser is used, for example with Pulse repetition rates in the kilohertz range or megahertz range it expedient that to synchronize pulsed lasers with a high-frequency camera. So can when using suitable powerful and high-frequency pulsed Laser sources and high frequency cameras with refresh rates in the Magnitude sections of 1 million frames per second or sampled in the range of move a few meters per second. Under these conditions z. Z. however - because of the still too low dynamics of the OASLMs - only the digital implementation of the Fast Fourier Transform (FFT) is considered feasible.

The wavefront-forming lens is preferably with a cylindrical lens to form at least one focus line, preferably at least approximately formed in the focal plane of the focusing lens. The light source can also consist of a row of individual laser diodes, for example one row with 256 single laser diodes. But it is also possible that the focusing objective a microlens arrangement, preferably one Microlens line, is upstream. So it is possible that according to the number the microlenses individual focus points preferably a line of separate focus points. The illumination of this microlens line can be done using a bundle cross-section former based on prism anamorphic.

So it will preferably be at least one narrow focus line or preferably a line of separated focus points in the object space. There is still talk of a focus line, including the case of a line of separated focus points.

The Light from the focus line striking the object is reflected and / or scattering using bundles of rays, the plane of which is parallel to the focus line lies on the camera chip of an area camera - or one Line scan camera with only one focus point - at least approximately sharply depicted. The light from rays from one to the focus line vertical plane is not focused on the camera chip, but clearly before or after the camera chip.

in the Camera space becomes a lateral shear perpendicular to the direction of the narrow Focus line firstly by beam splitting by means of diffraction or secondly, by splitting the amplitude on a beam splitter layer generated. Moreover becomes a wavefront formation of light on the way from the object done to the camera so that an aspherical, preferably a cylindrical wavefront is formed and so there is an astigmatic image of the focus line on the camera. The cylindrical wavefront is created by means of astigmatic imaging System generated, which system is a refractive or a diffractive Cylinder lens can be, the cylinder axis parallel to Focus line can be aligned. The astigmatic imaging system can also be combined with a prism anamorphic weak, spherical Be formed lens. The astigmatic imaging system can but also only in conjunction with a prism anamorphic a spherical trained focusing lens. Furthermore can a camera lens can be arranged in front of the camera, which is used for sharp Illustration of the rays of the tuft of light that is used in the Spread the focus line parallel planes.

The The location of this wavefront formation is the lateral shear generating optical beam splitter preferably upstream on the light path, i.e. in the beam path. From the pixels of the chip the areal Camera that is on a line perpendicular to the narrow focus line in the Object space are aligned, the spatial frequency of the light distribution certainly.

On Lateral-shear optical beam splitter can be used as a a light diffractive grating is formed with a subordinate lens be, with the light diffractive grating in the first focal plane of this lens, viewed in the direction of light propagation. The lens is preferably that assigned to the camera. there the light-diffractive grating can also be in a plane at this focal plane optically conjugate plane, but according to the location of the wavefront formation. Another possibility, an optical beam splitter producing lateral shear training consists in the application of a triangular interferometer, which has a cyclic beam path, which the space placed in front of the camera or the room in front of the focusing lens is. By means of this triangular interferometer, two coherent sub-beams, the a lateral shear 2Δq to each other, formed. This triangle interferometer can each with a complex, multi-level imaging arrangement conjugated space that optically to these two spaces mentioned is conjugated.

Out the spatial Frequency of light distribution in the plane of the camera chip becomes after deducting a carrier frequency from the radius of the cylinder wave in the detection plane arithmetically results in a difference frequency formed so that by means of pixels of each line of the camera chip, at least in one line, the difference frequency determines becomes.

This Differential frequency can each be the z-deviation from the narrow Focus line in the object space of each object point of the associated line on the camera at least approximately proportional his. in principle can do the Fourier transform of the lateral shear distribution - as already shown - too performed with optical means become.

in the inventive optical method becomes a focus figure, a Focus point or a focus line line on which to be examined object surface educated. The focus line is considered in more detail. This becomes more on the object surface or less sharply imaged so that it is within the depth measurement range preferably a linear light figure is formed on the object surface becomes. The object surface is mapped and detected. The object imaging, i.e. the tufts of rays coming back from the object the plane perpendicular to the focus line are fundamentally strong in the detection defocused, d. H. these bundles of rays form in the plane of the Detection not a linear light figure. On the way of the light from the object surface at the level of the detection, i.e. in the beam path, two coherent partial beams are formed, that in planes perpendicular to a focal line a lateral shear 2Δq that is not equal is zero. The object imaging rays in the planes parallel to the focus line are rather good for detection focused, so form the object rather sharply into the plane of the Detection.

The basic idea of the inventive Ver driving, reduced to the bare minimum, is the following: The light from the focus figure depicted on the object surface, preferably a focus line, whereby in principle a focus spot can also be generated, experiences a division of the amplitude. A lateral shear 2Δq, which is not equal to zero, is also impressed on the two coherent partial beams that arise in the plane perpendicular to the focal line. The object-imaging bundles of rays in the plane perpendicular to the focus line are defocused during the detection and the object-imaging bundles of rays in the plane parallel to the focus line are focused during the detection. The latter therefore depict the object rather sharply. In this way, shear strips are formed and detected in the plane of the detection. From the phase response perpendicular to the shear strips in a defined area 2 B of a camera chip, the placement of each object point from its reference position is determined by the phase difference from the reference phase response.

It becomes at least a single focus point or a single narrow one Focus line formed in the object space. Training multiple focus point or focus lines is also possible. The light from the focus figure (s), preferably focus lines the object is reflected and / or scattered in a directed manner.

The optical sensor for high-speed profile determination has a light source with at least a single light wavelength. This light source is at least one imaging level and also a Subordinate focusing lens, the imaging level and that Focusing lens on the one hand serve to form at least one focus line on the object surface. In this way, at least one linear light figure is formed on the object surface. The focusing lens serves about it also to map the object surface with the focus line shown on it. A camera with a camera chip is arranged downstream of the focusing lens. According to the invention, the focusing lens has an astigmatic imaging design or the focusing lens is anastigmatic imaging and on the light path from the object surface to the camera, the focusing lens is at least one astigmatic imaging system downstream. In any case, there is an astigmatic imaging system which astigmatically maps the focus line from the object surface onto the camera chip. The object imaging rays, i.e. the rays coming back from the object in the plane perpendicular to the focus line are fundamentally strong in the detection defocused, d. H. these bundles of rays form in the plane of the Detection then no linear light figure.

in the Beam path from the object surface to the plane of the camera chip an optical means generating lateral shear is arranged. So become two coherent in the beam path Partial beams formed a lateral shear in the planes perpendicular to a focal line Have 2Δq.

Preferably is the astigmatic imaging system that produces the lateral shear optical system Advance funds.

Preferably is the astigmatic imaging system as prism anamorphic or designed as a lattice anamorphic or as a prism lattice anamorphic and also upstream of the lateral shear generating beam splitter.

It can also be that the astigmatic imaging system with is formed at least one cylinder lens. This enables one volume-minimized arrangement.

Preferably is the lateral-shear optical beam splitter as an asymmetrical triangle interferometer. The triangle interferometer is very robust and works on the optical path difference zero. This is for the stable function of the sensor is of great importance. The most The resultant lateral shear 2Δq is a function of asymmetry and zero for a symmetrical triangular interferometer. Besides, is for a triangular interferometer Lateral shear independent on the wavelength of the Light. This property can also be used in the signal evaluation of great Be an advantage because light of different wavelengths cw light or pulsed light with pulses of different wavelengths, so interference fringe patterns different spatial frequency can form. The pulsed light can generated by one or more lasers, each pulse a different wavelength can have. When using several pulsed lasers, this can Light of different wavelengths be sent synchronized in time. So is a clear one Evaluation possible.

The generated light wavelengths can be recognized by means of frequency analysis using these interference fringe patterns of different spatial frequencies. This is advantageous if several linear light figures with light of different wavelengths are to be formed at different locations in the object space, laterally or in depth. Light sources that generate frequency combs can be used. By means of this spectral coding, both the z position and the lateral location of an object point can be determined in the evaluation based on the spatial frequency ν associated with the wavelength from a surface light distribution in the detector plane. The asymmetrical triangular interferometer is preferred A slit diaphragm is assigned in the beam path, where the tufts of rays intersect from planes that are perpendicular to focus lines. The intensity distribution I (x, ν) in the shear stripe field over x in the plane of the detection, the x direction being perpendicular to the stripes, can be determined with the mean intensity I ₀ , the stripe visibility ν, and an initial phase φ ₀ but usually should be close to zero, as follows: I (x, ν) = I 0 [1 + ν · cos (2πνx + φ 0 )].

wobei ρ_L die detektierte Streifenzahl auf dem Kamera-Chip der ausgewerteten Breite 2b beträgt. Für den Randstrahlwinkel α gilt dann für hinreichend kleine Winkel näherungsweise

Then for the spatial frequency ν:

where ρ _{L is} the detected number of strips on the camera chip of the evaluated width 2 B is. For the marginal jet angle α then approximately applies for sufficiently small angles

In this case, r represents the current distance of the idealized source point of the interfering waves in a section perpendicular to the focus line. In the event that an object point is in the focus of the focusing lens, r = r _{0 should} apply. A possible value for r ₀ can be the focal length of the camera lens f _K 'if the optical layout of the sensor is designed accordingly. If the object point under consideration is defocused by an amount Δz, the marginal beam angle α changes. If the idealized source point of the interfering partial beam bundles, which can be described sufficiently well by cylinder waves, lies in the main plane of the camera lens, Δz = Δr applies, and it follows for the sufficiently small, object point-dependent edge beam angle α

bestimmt werden. Bei einer Breite des ausgewerteten Streifenfeldes von beispielsweise 2b = 4mm, wobei dieser Breite etwa 640 Pixel auf dem Kamera-Chip zugeordnet sind, einer fest eingestellten Lateral-Shear von 2Δq = 0,5 mm, bei einer Brennweite des Fokussierobjektivs f_K' = 25 mm und einer Ablage des betrachteten Objektpunktes vom Fokus von 5 μm ergibt sich für die Lichtwellenlänge λ = 0,5 μm über dem Detektionsfeld eine Änderung der Streifenzahl um 0,032 bei einer Gesamtstreifenzahl von 160. Diese Änderung der Streifenzahl entspricht einer Phasenänderung im Feld von 11,5° über der Breite von 4mm. Eine derartige Phasenänderung kann mit den bekannten phasenauswertenden Algorithmen noch gut aufgelöst werden, wenn man bedenkt, dass im phasenauswertenden Algorithmus bis zu 640 Pixel für einen einzigen Phasenwert ausgewertet werden können. Dies trifft auch zu, wenn der Phasengang über dem Feld bei besonders kleinen Brennweiten f_K' nicht streng linear ist. Dafür sind die Algorithmen in geeigneter Weise anzupassen. Wie das Dimensionierungsbeispiel zeigt, kann mit einem derartig dimensionierten Sensor eine Auflösung von besser als 5 μm erreicht werden. Es ergibt sich somit im Vergleich zu einem abstandsmessenden Interferometer mit der Lichtwellenlänge von ebenfalls 0,5 μm ein Dehnungsfaktor der effektiven Wellenlänge des Sensors von 625.From these relationships, both the dependence of the number of strips ρ _L , which stands for the location-dependent phase response over the detection field, and the spatial frequency, depending on the offset Δz of the respective object point, on the focal plane of the camera lens

be determined. With a width of the evaluated strip field of, for example, 2b = 4mm, this width being assigned about 640 pixels on the camera chip, a fixed lateral shear of 2Δq = 0.5mm, with a focal length of the focusing lens f _K '= 25 mm and a deposit of the object point under consideration from the focus of 5 μm, for the light wavelength λ = 0.5 μm above the detection field there is a change in the number of strips by 0.032 with a total number of strips of 160. This change in the number of strips corresponds to a phase change in the field of 11. 5 ° across the width of 4mm. Such a phase change can still be resolved well with the known phase-evaluating algorithms, if one considers that up to 640 pixels can be evaluated for a single phase value in the phase-evaluating algorithm. This also applies if the phase response over the field is not strictly linear at particularly small focal lengths f _K '. The algorithms have to be adapted accordingly. As the dimensioning example shows, a resolution of better than 5 μm can be achieved with such a dimensioned sensor. In comparison to a distance-measuring interferometer with the light wavelength of also 0.5 μm, this results in a strain factor of 625 for the effective wavelength of the sensor.

The depth measurement range for this task is the size of at least 100 μm and is therefore usually sufficiently large for the measurement of a micro profile. It should also be noted that the different depth position only produces a comparatively small modulation of the spatial frequency. As a rule, the deviation from the mean spatial frequencies lies in a range of at most approximately +/- 2%. This results from the limited depth of field. Outside this depth of field, the contrast of the interference phenomenon is zero. For example, a stripe pattern whose associated light wavelength differs by more than 5% from a first wavelength can be "identified" by a suitable frequency-selective numerical evaluation method. With this approach, several numerically distinguishable focus lines can be "positioned" in the depth of the object space. This offers the possibility of creating a depth measurement range that is multiply enlarged compared to a single wavelength. A very high measurement accuracy can thus be coupled with a large depth measurement range using this multi-wavelength approach or chromatic approach. The numerical evaluation methods can be carried out, for example, on the basis of bandpass filtering, these methods not being discussed further here. In any case, however, it is possible to use modern phase evaluation methods, including wavelets, to cover the overall phase response across the width 2 B or to determine the deviation of the same from an average overall phase response with high resolution. The overall phase response can be of the order of 1000π. Ultimately, the total phase response in each line of the camera chip must be determined that is perpendicular to the Shear strip lies. In the case of a cw multi-wavelength light source, there are usually several patterns of different spatial frequencies at the same time in order to be able to cover the depth range without gaps. When evaluating, however, it must be noted that the spatial frequency in the stripe pattern is only approximately a constant when evaluating larger fields.

It is possible, too, at three laser wavelengths, for example, the 475 nm, 532 nm and 635 nm or 633 nm can be a refractive, dispersive or dispersive-refractive optical wedge in the beam path to arrange. The optical wedge, which is preferred for the medium wavelength having a deflection of zero is preferred in the beam path arranged in front of the camera lens and generates a wavelength-dependent deflection the beam of light. This results in three in the detection level, i.e. on the camera chip laterally completely separate areas for the three stripe patterns in the three wavelengths. So is an easy one Evaluation of the three stripe patterns possible, since only monofrequency Signals occur. The three measuring ranges in the Overlap depth slightly.

Farther the following inventive approach is proposed: the optical one Sensor for high-speed profile determination has a light source with at least one light wavelength. This is at least a mapping level and beyond a focusing lens downstream, the imaging level and the focusing lens on the one hand for training at least one Focus line on the object surface serve. This creates a linear light figure on the surface of the object. The Focusing lens serves over it to map the object surface with the focus line. The focusing lens are a camera lens and a camera with a camera chip. What is innovative here is that the focusing lens is on the light path subordinate at least one astigmatic imaging system and can be arranged upstream of the camera lens and this camera lens preferably a phase grating is assigned, which is at least approximately in the focal plane of the camera lens, which is the focusing lens is facing. The phase grating in the focal plane of the camera lens also produces a lateral shear together with the camera lens generating optical beam splitter represents, but this is known to be after the camera lens occurring lateral shear dependent on the wavelength λ of the light used. The effect is thus in the plane of the camera chip for all wavelengths same spatial frequency. This can be an advantage, but it can be no separation of light of different wavelengths can be achieved.

Farther the following inventive approach is proposed for the sensor: The optical sensor preferably for high-speed profile determination has a light source with at least one light wavelength. This is followed by at least one imaging level and also a focusing lens, the imaging level and the focusing lens on the one hand Form at least one focus line on the object surface. This creates a linear light figure on the surface of the object. The focusing lens serves above to map the object surface with the focus line. the Focusing lenses are a camera lens and a camera with one Subordinate camera chip. According to the focusing lens at least one astigmatic imaging system on the light path downstream and an optical beam splitter producing lateral shear downstream of the camera lens and upstream of the camera chip his. this leads to to a space-saving arrangement, if preferably the imaging System with at least one cylinder lens is formed. Preferably the cylinder axis of this cylinder lens is parallel to Focus line arranged. The astigmatic imaging system is included preferably formed with at least one prism anamorphic. The lateral-shear-generating optical beam splitter is preferably used designed as a triangular interferometer to produce a wavelength-independent lateral shear to be able to. However, for the optical beam splitter generating the lateral shear is a phase grating used, for example around a very compact sensor arrangement To obtain the phase grating is formed with equidistant furrows. The cylinder lens preferably has the same focal length as the camera lens is on and the focus of the cylinder lens is at least approximately in the main point of the camera lens that faces the cylinder lens is. This allows the camera lens to expand in the main section be trimmed to minimize volume. The cylinder lens is preferably designed as a phase grating to increase construction volume save up. A double slit diaphragm is preferably in the focal plane arranged of the cylinder lens. This double slit diaphragm locks undesirable scattering and diffracted light.

The sensor is preferably designed in the following manner. A laser is used as the light source and the focusing lens and camera lens are designed as an afocal imaging stage. The preferably arranged phase grating for generating two coherent sub-beams is arranged in the common focal plane of the focusing lens and the camera lens. The astigmatic imaging system is preferably designed as a diffractive cylinder lens and the diffractive cylinder A slit diaphragm is preferably arranged downstream of the lens.

Preferably can the diffractive and preferably also formed as a phase grating Cylinder lens should also be slightly off-centered, taking the direction the decentration is formed perpendicular to the focus line. So arise by splitting the amplitude preferably two coherent sub-beams with different ones Curvature. Preferably the coherent ones partial bundle depicted sharply in the plane of the detection by means of sagittal rays. Preferably a coherent one partial bundle imaged sharply into the plane of the detection by means of meridional rays. In this case, there are no equidistant shear interference fringes in the level of detection. However, one of the coherent bundles is from completely covered others. So the light energy is used optimally.

For the sensor can several lasers with different wavelengths each Light sources are used. It can be used for spectral separation the light a color sensitive camera with at least two color channels or As already mentioned an optical wedge for spectral splitting may be arranged, for example the inspector from ZEUTEC can be used. Here is the focusing lens preferably chromatic, but the camera lens preferably achromatic. This chromatic design of the Focusing lens enables a second or further focus lines at a different depth of the object space.

At the Process with the specified means such as light source, focusing lens, The camera chip of a camera is therefore a narrow line of focus in the object space educated. The light that strikes the object is turned off Focus line by reflection and / or scattering by means of bundles of rays, the Layers are parallel to the focus line, shown sharply. The bundles of rays whose Planes lying perpendicular to the focus line are defocused during detection and before the sharp image from the beam of the object becomes two coherent Partial beams which generates a lateral shear to each other and in the plane perpendicular to the focus line. So can upon detection in the plane of the sharp image of the bundle of shear strips be included.

It can thus according to the inventive method two coherent Partial beams with a lateral shear using firstly beam splitting by diffraction or secondly, amplitude division is generated on a divider layer. Furthermore, a wavefront formation of the light on the Is carried away from the object to the camera. So become cylindrical Wavefronts formed and so there is an astigmatic image the focus line on the camera chip of the camera. The location of this wavefront formation is the optical beam splitter that produces lateral shear upstream on the light path and from the pixels of the areal Camera that is on a line perpendicular to the narrow focus line in the Object space are aligned, the spatial frequency of the light distribution or the phase response over the lateral extent of the camera chip. It can be out this frequency after subtracting a carrier frequency from the radius the cylinder shaft results, a difference frequency is formed, so that by means of pixels of each line of the camera chip, at least however, in a line, a difference frequency is determined, whereby this difference frequency to calculate the z-deviation of each object point from the narrow focus line in the object space is used.

Farther can preferably several focus lines are arranged one above the other in the object space, whereby the micro profile is synchronized over a larger depth range can be recorded. Several focus lines are preferably different Color arranged side by side in the object space. But it can also several focus lines in the object space of different colors one weird area be arranged.

The Arrangement order of shear-generating and astigmatism-generating optical means is when using a triangular interferometer arbitrary for the lateral shear generation. When using an aspherical Lens, usually a cylindrical lens to produce astigmatism, and a phase grating beam splitter is this aspherical Lens always precedes the phase grating. When using a grid or prism anamorphic to produce astigmatism is its Position to the optical beam splitter that generates lateral shear any. For a multi-line arrangement, the line location diffractive over the wavelength or generated refractive is probably a triangle interferometer as a lateral-shear optical beam splitter Particularly suitable because every wavelength is always in the detector plane is converted into its own spatial frequency.

A large depth measuring range with reduced depth sensitivity can be realized for a sensor with a focusing lens that is particularly long in focus. With a correspondingly fast camera, comparatively rough 3D profiles can also be captured by hand. The amount of movement in the direction of movement of the sensor, here in the x direction, can be determined by an external optical position determination system, for example in the manner of a mini GPS, which at least approximately detects the position of the sensor head. But it can the movement of the sensor can also be detected by correlation evaluation of structural features of the measurement object itself. The optical position determination system can also be based on triangulation.

For the measurement the layer thickness is supplied by each reflex from an interface reflected bundle, which is in two coherent partial bundle split and preferably in the described to astigmatic is mapped. From the resulting interferogram in the detection plane the layer thickness is taken into account of the known refractive indices of the layer. With this optical Procedures will provide information about the layer thickness of a layer is obtained by the detected interferogram is evaluated. The detection with a is of particular advantage fast CCD line scan camera.

Description of the figures

In the 1 to 8th The sensors shown are called shearoline sensors because the focus figure on the object represents a line and the shear effect is applied.

This in 1 from a laser diode 1 , which is designed as a point light source, outgoing light bundles reach a refractive cylindrical lens 2 so that the rays are collimated in the main section. Possibly occurring astigmatism of the laser diode 1 only plays a subordinate role here, or is corrected by professional action. The rays perpendicular to the main cut after the cylindrical lens 2 continue unchanged. There is a cylindrical shaft with the cylinder axis parallel to the main plane. This wave enters the divider cube 3 on. The reflected light reaches the spherical lens 4 , whose focal point coincides with the point light source, and a cylindrical wavefront is created again, but now with the cylinder axis perpendicular to the main section. It forms on the object 5 a focus line in and around the focus position. The spherical lens 4 is refractive and corrected for the laser wavelength at least almost diffraction-limited, since the focus line can have an extension of a few millimeters. The light backscattered from the object surface by this line of focus passes through the lens 4 and the divider cube 3 and gets on a refractive cylindrical lens 6 , the second cylindrical lens in the sensor. The one on the cylindrical lens 6 converging cylinder shaft passes through a phase grating 7 with equidistant micro furrows. Diffraction creates two coherent sub-beams, in this case two cylindrical waves. It is very important that the phase grating 7 in the Fourier plane of the downstream camera lens 8th located. The cylindrical waves converge in the main section depending on the focus position, for example immediately in front of the camera lens in Om1 'and Om2', where a sufficiently wide slit diaphragm is arranged to block stray light and other diffracted light. However, the slit width is selected such that both coherent light beams can pass through the slit diaphragm even when the object is defocused, since the focal lines that form when the object is defocused move in depth. The developing focus lines here represent the real images Om1 'and Om2' of the penetration points of the focus lines on the object here in the main section. In principle, a light-scattering cylindrical lens can also be used, since it basically does not matter whether the images Om1 'and Om2' of the focus spot O are real or virtual. That is, the 2nd cylindrical lens lens can be light-collecting or light-scattering. In the case of a collecting cylindrical lens, the requirements for the subsequent lens are generally lower, since the aperture used is smaller.

The two coherent images of the focus lines, the focus lines forming on the object, are made by the camera lens 8th on the camera chip 9 imaged by a camera. The sagittal rays Os1 'and Os2' of the object point O are mapped there. That is on the camera chip 9 resulting shear strip image shows 2 , The change in spatial frequency on the camera chip 9 has been exaggerated here. The change in spatial frequency is typically only in the range of one to two percent of the mean frequency. Focal positions, which could cause a larger spatial frequency change, provide interference with the strip contrast around zero, since the object point is then already clearly outside the depth measurement range.

The spherical lens 8th is refractive here. It comes on the camera chip 9 for interference of the partial beams with lateral shear, so that lateral shear strips can be detected in the overlap area of the two coherent cylinder waves. The sagittal rays, i.e. the rays in tufts perpendicular to the main section, form the focus line on the object 5 at least approximately sharp on the camera chip 9 from. Depending on the current position of each point of the object 5 in the focal line, a spatial frequency ν depends on the deviation Δr from the focal plane. So the profile can correspond to the numerical aperture of the two lenses 4 and 8th laterally in the direction of the focus line. The 2 shows a typical lateral shear stripe distribution on the camera chip 9 , The evaluation to determine the spatial frequency can be carried out by an FFT. But it is also possible Lich the phase change over a defined area 2 B to be determined on the camera chip by means of known spatial phase evaluation. Because the diffractive cylindrical lens 6 If an additional wavefront curvature has been introduced that leads to a carrier frequency, this carrier frequency must be calculated out. Then the spatial frequency ν in each line of the camera chip is at least approximately proportional to the distance Δz of each object point from the focal plane of the focusing lens. The determination of the phase response via 2 B can be done using spatial phase cells. In this way, algorithms from the dynamic phase shift process can be applied. For example, the well-known 5-phase sliding algorithm according to J. Schwider can be used.

The lens change from lens 2 always takes place so that the focus of each lens is in position F2. A corresponding mechanical tube length must always be taken into account when designing lenses with different focal lengths.

It can use a CMOS area scan camera with a logarithmic characteristic, if, for example only the fundamental wave of an at least approximately known frequency is evaluated. A great dynamic can be achieved in this way.

The arrangement in the 1 enables a large depth measurement range and leads to a constant imaging scale and the spatial frequency difference is strictly proportional to the offset Δz of an object point from the focal plane.

The order after 3 that largely depend on the design 1 based on a simple arrangement with only one focusing lens 4 a slightly smaller depth measurement range. The image scale is not quite constant over the depth, since there is no telecentric beam path here. The prism group represents a prism anamorphic 10 which creates an aspherical wavefront from the spherical bundle. By enlarging the bundle after the prism anamorphic 10 the tufts of rays are mapped from planes parallel to the focal line "extrafocal". The image of the point O is imaged using meridional rays in Om1 'and Om2' and using sagittal rays according to Os1 'and Os2'. This is on the camera chip 9 resulting shear strip image shows 4 , Is the prism anamorphic? 10 bundle-reducing, by passing the first prism surface vertically, the bundles of rays are imaged "intrafocally" from planes parallel to the focal line.

In 5 is a light source 21 arranged, which is designed as a multi-wavelength laser light source. This light source can therefore have a frequency comb characteristic. The light source 21 is an astigmatic imaging, wavefront-forming optical system 22 downstream. This system 22 created together with the chromatic focusing lens 4 each focus lines at different depths of the object space depending on the wavelength of the light source 21 emitted light. Ultimately, it is reserved for professional action, the special properties of the light source 21 such as B. to compensate for different positions of the source centers and / or different radiation characteristics in the direction of light propagation. The system 22 It is best to map the source centers of all colors to infinity, whereby the direction of propagation should be the same for all wavelengths, at least in the main section. Then the goal is to generate linear focus figures that are "stacked" in depth or height in a standing plane. The arrangement of the focus lines in the standing plane brings advantages for the metrological application because it enables a precise cut The frequency distances are made so large that the respective wavelength is recognized by means of suitable evaluation technology. Pulsed lasers with a different wavelength can also be used so that temporal multiplexing can be carried out. The frequency modulation is based on the depth position always smaller than the spatial frequency spacing that results from the light wavelength, which is good for the evaluation if on the camera chip 9 only a maximum of two to three interference fringe patterns of different spatial frequencies are formed at the same time. The light of the other wavelengths is already on the single slit diaphragm 16 largely blocked. There is also a single slit diaphragm 16 in the focal plane of the cylinder lens 6 , The effect of this single slit diaphragm 16 is comparable to that of a confocal pinhole, even if the blocking effect only occurs in one direction.

7 shows an arrangement with a comparatively large depth measurement range. The image scale is constant and that on the camera chip 9 The spatial frequency difference to be detected is proportional to Δz. It becomes a diffractive cylindrical lens 6 used. The lateral shear-generating beam splitter consists of a phase grating 7 and the lens 8th , where the phase grating 7 in the focal plane of the lens 8th located. That is on the camera chip 9 resulting shear strip image shows 8th , Here, too, the spatial frequency variation is extremely exaggerated. Extraneous light, which can occur, for example, with translucent objects, appears on the double slit diaphragm 17 largely blocked. There is this double slit aperture 17 in the focal plane of the cylinder lens 6 ,

Claims (32)

Optical sensor, with a light source ( 1 ) with at least one light wavelength, a focusing lens ( 4 ), a mapping level ( 2 ) to form at least one focus line on the object surface ( 5 ) and a camera with a camera chip ( 9 ), characterized in that the focusing lens ( 4 ) is astigmatically imaging or that the focusing lens ( 4 ) is anastigmatic imaging and on the light path to the camera chip ( 9 ) the camera the focusing lens ( 4 ) at least one astigmatic imaging system is arranged downstream, so that there is an astigmatic imaging system that detects the focus figure, a focus point or a focus line line from the object surface ( 5 ) astigmatic on the camera chip ( 9 ) images and on the light path to the camera chip ( 9 ) a lateral shear-generating optical beam splitter is arranged in the beam path. Optical sensor according to claim 1, characterized in that the astigmatic imaging system the lateral shear generating optical beam splitter is upstream. Optical sensor according to at least one of claims 1 to 2, characterized in that the astigmatic imaging system as a prism anamorphic ( 10 ) or is designed as a lattice anamorphic and is arranged upstream of the optical beam splitter that generates lateral shear. Optical sensor according to at least one of claims 1 to 3, characterized in that the astigmatic imaging system with at least one cylinder lens ( 6 ) is trained. Optical sensor according to at least one of claims 1 to 4, characterized in that the lateral shear-generating optical beam splitter as an asymmetrical triangle interferometer ( 11 ) is trained. Optical sensor according to claim 5, characterized in that the asymmetrical triangular interferometer ( 11 ) a slit in the beam path 16 is assigned, at least approximately where the bundles of rays intersect from planes that are perpendicular to focal lines. Optical sensor, with a light source ( 1 ) with at least one light wavelength, a focusing lens ( 4 ), an imaging level for forming at least one focus line, a camera lens ( 8th ) and a camera chip ( 9 ) of a camera, characterized in that the focusing lens ( 4 ) at least one astigmatic imaging system is arranged on the light path and the camera lens ( 8th ) is arranged upstream and this camera lens is assigned a phase grating, which is at least approximately in the focal plane of the camera lens ( 8th ) which is the focus lens ( 4 ) is facing. Optical sensor, with a light source with at least one light wavelength, a focusing lens ( 4 ), an imaging level for forming at least one focus line, a camera lens ( 8th ) and a camera chip ( 9 ) of a camera, characterized in that the focusing lens ( 4 ) at least one astigmatic imaging system is arranged on the light path and an optical beam splitter that generates lateral shear is arranged downstream of the camera lens ( 8th ) and the camera chip ( 9 ) is upstream. Optical sensor according to at least one of claims 7 or 8, characterized in that the astigmatic imaging system with at least one cylinder lens ( 6 ) is trained. Optical sensor according to claim 9, characterized in that the cylinder axis of the cylinder lens ( 6 ) is arranged parallel to the focus line. Optical sensor according to claim 8, characterized in that the astigmatic imaging system with at least one prism anamorphic ( 10 ) is trained. Optical sensor according to claim 8, characterized in that the lateral shear-generating optical beam splitter as a triangular interferometer ( 11 ) is trained. Optical sensor according to at least one of claims 7, 9 and 10, characterized in that the phase grating ( 7 ) is designed with equidistant furrows. Optical sensor according to at least one of Claims 7 to 11 and 13, characterized in that the cylinder objective ( 6 ) the same focal length as the camera lens ( 2 ) and the focus of the cylinder lens ( 6 ) is located at least approximately in the main point of the camera lens facing the cylinder lens and the cylinder lens ( 6 ) is designed as a phase grating and a double slit diaphragm ( 17 ) at least approximately in the focal plane of the cylindrical lens ( 6 ) is arranged. Optical sensor according to at least one of claims 1 to 11 and 13 to 14, characterized net that the focusing lens ( 4 ) and the camera lens ( 8th ) are designed as an afocal arrangement and the phase grating ( 7 ) to generate two coherent sub-beams in the common focal plane of the focusing lens ( 4 ) and camera lens ( 8th ) is arranged and the astigmatic imaging system as a diffractive cylindrical lens ( 6 ) and the diffractive cylindrical lens ( 6 ) a double slit diaphragm ( 17 ) is subordinate. Optical sensor according to at least one of claims 1 to 15, characterized in that at least one laser ( 1 ) is used as a light source. Optical sensor according to at least one of claims 1 to 15, characterized in that several lasers with different wavelength be used as a light source and a color sensitive camera arranged with at least two color channels is. Optical sensor according to at least one of Claims 1 to 15, characterized in that at least one pulsed laser ( 1 ) is used as a light source. Optical sensor according to at least one of claims 1 to 15, characterized in that a plurality of pulsed lasers ( 1 ) are used as light sources that emit light of different wavelengths synchronized in time. Optical sensor according to at least one of claims 17 to 19, characterized in that the focusing lens ( 4 ) chromatic, but the camera lens ( 8th ) is achromatic. Optical sensor according to at least one of claims 16 to 20, characterized in that a refractive, dispersive or dispersive-refractive effect optical wedge is arranged in the beam path. Optical sensor according to at least one of claims 1 to 21, characterized in that the focusing lens ( 4 ) a microlens arrangement is arranged upstream. Optical sensor according to claim 22, characterized in that the microlens array as a microlens line is trained. Optical method with the specified means of the generic term for the sensor, in which a narrow focal line is formed in the object space, characterized in that the cw light striking the object or pulsed light from the focal line by reflection and / or scattering by means of bundles of rays, the Layers are parallel to the focal line, are sharply imaged during detection and the bundles of rays, the planes of which are perpendicular to the focal line, are defocused during detection and before the sharp imaging from the bundle of rays from the object ( 5 ) two coherent partial beams are generated, which have a lateral shear to each other and in the plane perpendicular to the focal line, and shear strips are formed by interference of the partial beams in the plane of the detection and from the phase response perpendicular to the shear strips in a defined area the storage of an object point to the focal plane is determined. Optical method according to claim 24, characterized in that two coherent partial beams with a lateral shear are generated by first beam splitting by diffraction or second amplitude splitting on a splitter layer and a wavefront shaping of the light on the way from the object to the camera is carried out and the location of this wavefront shaping is arranged upstream of the optical beam splitter producing lateral shear on the light path and by means of the pixels of the camera chip ( 9 ), which are aligned on a line perpendicular to the narrow focus line in the object space, the spatial frequency of the shear strips is determined in each case. Optical method according to claim 25, characterized in that several focus lines are arranged one above the other in the object space. Optical method according to claim 25, characterized in that several focus lines of different colors in the object space are arranged side by side. Optical method according to claim 25, characterized in that several focus lines in the object space are different Color are arranged on a plane inclined to the sensor. Optical method according to at least one of claims 24 to 28, characterized in that the amount of movement in the direction of movement the sensor through an external optical positioning system, which detects the position of the sensor head at least approximately becomes. Optical method according to at least one of claims 24 to 28, characterized in that the movement of the sensor a correlation evaluation of structural features of the measurement object is recorded. Optical method with the specified means of the generic term for the optical sensor, at which at least one focal point is formed in the object space, characterized in that the cw light or pulsed light striking the object from the focal point is depicted sharply by reflection and / or scattering and the bundles of rays, the planes of which lie parallel to the reference plane during detection are defocused and before the sharp image from the beam from the object ( 5 ) two coherent partial beams are generated, which have a lateral shear to each other and in the plane parallel to the reference plane and shear strips are formed by interference of the partial beams in the plane of the detection and from the phase response perpendicular to the shear strips in a defined area the storage of an object point to the focal plane is determined. Optical method according to at least one of claims 24 or 31, characterized in that information about the layer thickness of a layer can be obtained by evaluating the detected interferogram becomes.