DE10348509A1 - Determining image errors by computing test object Zernike coefficients involves detecting pupil edge according to position, size in computer system in addition to wavefront measurement by evaluating light from object recorded by CCD camera - Google Patents

Abstract

The method involves using a computer program in a computer system (6) connected to a first charge coupled device camera (3) used with a microlens array (7). The edge of the pupil is detected and determined according to position and size in a computer system in addition to a wavefront measurement by recording light direct from the test object in the first and/or second (10) CCD camera and evaluating the measurement values in the computer program in determining the coefficients. An independent claim is also included for an arrangement for implementing the inventive method.

Description

The The invention relates to methods and related arrangements for determination of image errors by calculating the Zernike coefficients of a sample with a computer program using a computer system, the connected to a first CCD camera, the first CCD camera a microlens array is provided, which is a wavefront sensor form.

In the measuring technique of optical components and systems in recent years is the Wavefront measurement technology become more and more important. Most frequently become the Hartmann sensor or its development in the form used by the Shack-Hartmann sensor. The pupil of a DUT in a transmitted light method or in an incident light method imaged a detector and the focal positions of rays Subpupils within the pupil relative to the ideal wave with a computer system. Such a detector will too referred to as wavefront sensor.

In the 1 is the basic operation of the Hartmann sensor and in 2 is that of the Shack Hartman sensor shown. 1 shows the Hartman sensor, where the wavefront 1 a pupil on a shadow mask 2 falls. The resulting intensity profile is measured with a CCD camera 3 detected and the centers of gravity of the intensity peaks are determined. The shadow mask 2 and the CCD camera 3 form the wavefront sensor 4 , Thereby are in the supervision 5 The CCD camera displays the intensity centers as filled points and the reference position of an ideal wavefront is shown as open points. In comparison with the reference positions for the ideal wavefront, the intensity focuses of the real wavefront are shifted. With the help of the computer system 6 These differences of the focal points are measured and local wavefront deformations are calculated and the wavefront is obtained by integration.

In the case of the Shack-Hartman sensor, which is in 2 is shown, is located at the location of the shadow mask 2 (in 1 ) a microlens array 7 , Again, the wavefront is calculated from the difference in the center of gravity positions.

In this wavefront calculation by the computer system 6 a circular pupil is assumed and Zernike polynomials are used for their mathematical description. The individual Zernike polynomial coefficients are assigned unique aberrations, see C. Hofmann "The Optical Image" Geest & Portig K.-G; Leipzig [1980] and D. Malacara "Optical Shop Testing" John Wiley and Sons; New York [1978].

The Zernike polynomials are normalized among other things to the pupil radius, so that the determination of the pupil margin the description of the pupil certainly. The Shack-Hartmann sensors according to the prior art determine the pupil edge over Threshold criteria. Today's wavefront sensors have arrays with about 64 × 64 Lenses, so that the Pupil margin can be determined with approx. 1 to 1.5 pixel accuracy can. This corresponds to a radius accuracy of 3 to 5%.

In the 3 an exemplary camera shot of a wavefront sensor is shown, which operates according to the Shack-Hartman method. The edge of a pupil 24 is determined by the individual no intensity points. The radial difference between the dashed circle and the dotted circle is 3.5%. The circles exemplify two possible solutions for the edge of a pupil 24 . 24 ' according to the criterion used for the calculation. The crossing point of the two perpendicular lines gives the center of the pupil, which is calculated from the determination of the edge.

With a lower error limit of one pixel and about 3% radius error receives one e.g. at spherical 3rd order aberration 12% and 5th order spherical aberration 17% uncertainty in the corresponding coefficient. In addition arise lower spherical ones Error (crosstalk).

To these errors are further shares are added from the provision the midpoint of the pupil result. At a decentering of 3% is e.g. spherical Aberration 3rd order issued as a coma.

In 4 The lowest orders of the spherical wave aberration sph04 (r) and sph06 (r) are plotted over the radius r. One recognizes the sensitivity of the edge to the in 5 shown derivations of the functions 4 , The strong increase of the model functions at the pupil edge causes a large influence of the pupil position on the error size.

Of the Invention is therefore the object of an improvement the accuracy and reproducibility of the determination of the Zernike coefficients improve.

This should in particular the determination of surface defects of optical assemblies and Aberrations of optical systems as well as aberrations of the human eye.

These Task is according to the invention in the Method thereby solved that the edge of the pupil depends on location and size in a computer system additionally for wavefront measurement is detected and determined by the first CCD camera or light registered by a second CCD camera which is directly from the examinee is displayed and these measured values from the computer program of the computer system in determining the Zernike coefficients to be evaluated with. This method manages the error at definition of edge of a pupil to lower under 3%.

at A first variant of the method is the determination of the edge the pupil by measurements of the second CCD camera in a Auflichtaufbau carried out.

at a second variant of the method is the determination of the edge the pupil by measurements of the second CCD camera in a transmitted light structure carried out.

at a third variant of the method is the determination of the edge the pupil by using a diffractive microlens array, which has an efficiency of less than 100% and thus partially transparent is done with the first CCD camera. Preferably, the Efficiency between 40% and 80%.

at a fourth variant of the method is the determination of the edge the pupil by using a refractive microlens array, which a degree of filling less than 100%, and that between the individual Lentils resulting gaps is transparent, performed with the first CCD camera. Preferably lies the degree of filling between 40% and 80%.

These Task is according to the invention with a first arrangement with a transmitted light structure consisting of a Light source with condenser, a pinhole in an object plane of a test object and a relay optic with a field stop, which is a pupil of the test piece on an array of the first CCD camera, continues to be the first CCD camera is connected to the computer system, taking in front of the array the first CCD camera, a microlens array is arranged and thus a wavefront sensor is present, solved in that between the relay optics and the wavefront sensor, a first beam splitter is arranged, the pupil of the specimen on both the wavefront sensor and a second CCD camera and the second CCD camera is connected to the computer system.

These Task is according to the invention with a second arrangement with a Auflichtaufbau consisting of the light source and the relay optics with the field stop, which is the pupil of the DUT on an array of a first CCD camera, still the first one CCD camera is connected to the computer system, taking in front of the array the first CCD camera, a microlens array is arranged and thus a wavefront sensor is present and - in the propagation direction of the Seen light - that Picture of the pinhole above a second beam splitter via the examinee on a reflector and from this again through the test object, over the second beam splitter is transferable to the relay optics, wherein the reflector before, within or after the focal point of the specimen is solved by the fact that between the relay optics and the wavefront sensor, the first beam splitter is arranged, the pupil of the specimen on both the wavefront sensor as well as the second CCD camera and the second CCD camera connected to the computer system.

Of the Reflector is preferably a plane mirror, the focal point of the test piece stands. The reflector further comprises alternatively a spherical mirror, which is before or after the focus test specimens, so that the rays returned to himself become. They pass through the test piece twice. The choice of Arrangement of the reflector is done according to which type of error prefers to be determined.

These Task is according to the invention with a third or a fourth arrangement with the wavefront sensor in Transmitted light or in Auflichtaufaufbau in which the microlens array over the array the CCD camera is arranged, solved by the microlens array is constructed of partially transparent diffractive optical elements, wherein the diffractive optical elements targeted an efficiency less than 100%, preferably less than 95%. In particular, lies the efficiency of the diffractive microlenses between 20% and 80%,

This object is achieved with a fifth or a sixth arrangement with the wavefront sensor in the transmitted light structure or in Auflichtaufbau in which the microlens array is disposed over the array of CCD camera, characterized in that between the individual lenses of the refractive microlens array transparent spaces are provided, the a fill factor of the microlens array ( 7 ) produce less than 100%, preferably less than 95%. In particular, the fill factor is between 40% and 90%.

The The invention will be described below with reference to figures. Show it:

1 : Hartmann detector

2 : Shack-Hartmann detector

3 : Camera shot of a Shack-Hartmann detector

4 : Representation of the lowest orders of the spherical wave aberration over the radius

5 : Derivatives of functions after 4

6 : Wavefront sensor in a transmitted-light structure according to the invention

7 : Wavefront sensor in an incident light construction according to the invention

8th : Wavefront sensor in incident light structure with modified microlens array according to the invention

9 : Wavefront sensor in the transmitted light structure with modified microlens array according to the invention

10 : Intensity section through the pupil center along a series of intensity peaks, measured with a diffractive microlens matrix

11 : Reduction of the packing density of the individual lenses on a microlens array

The 1 to 5 illustrate the state of the art, as described in the introduction.

In 6 an inventive transmitted light structure is shown, for the detection of the pupil position and the size of a second CCD camera 10 used. In front of the wavefront sensor 4 is a beam splitter 8th attached so that the relay optics 9 the pupil both on the wavefront sensor 4 as well as the second CCD camera 10 maps.

With a lighting 13 with condenser 14 the object plane is illuminated. There is a pinhole in the object plane 15 their point picture of the examinee 12 (shown here as a whole lens) and via the relay optics 9 on the first CCD camera 3 and the second CCD camera 10 is shown.

Within the relay optics a field stop is arranged, which is chosen so large that the point image is transmitted from this untrimmed. The relay optics match the pupil diameter of the specimen to the dimensions of the CCD cameras 3 and 10 at.

The centers of gravity of the intensity peaks are as described in the prior art with the first CCD camera 3 in the wavefront sensor 4 certainly. However, the pupil edge is replaced by the additional second CCD camera 10 certainly. The data of both CCD cameras 3 and 10 become the computer system 6 supplied and used in the calculation program for wavefront reconstruction and to determine the Zernike coefficients.

This arrangement can be realized with little effort. The wavefront sensor 4 and the second CCD camera 10 are to the state of the art anyway required a computer system 6 connected. The computer program performs a boundary search from the incoming data. Another advantage of this solution is that the intensity distribution corresponding to the number of pixels of the second CCD camera 10 versus the number of subpupils of the microlens array 7 of the wavefront sensor 4 can be determined more precisely. A commercial CCD camera has 1556 × 1024 or 768 × 512 pixels. Corresponding to this large number of pixels compared to the number of individual lenses of the microlens array 7 in the wavefront sensor 4 For example, with 64 × 64 individual lenses, the intensity distribution that results from the position of the pupil can be determined more precisely. Thus, the error of the pupil position is at least an order of magnitude better than when using the known Shack-Hartmann sensor.

7 shows, as an alternative to the previously described transmitted light arrangement, an incident light arrangement with reflection on a mirror 18 in the focal plane.

Due to the double light passage through the test object 12 the symmetrical aberrations double, the asymmetrical ones, such as coma, are eliminated. With the help of such an arrangement, the determination of spherical aberration continues to gain in accuracy.

The lighting 13 with the condenser 14 is via a second beam splitter 17 in the infinite beam path of the specimen 12 coupled. As a result of the reflection on the mirror 18 the beams pass through the test specimen twice. Apart from the axis of the beam, the light passes in two different optical paths, so that the symmetrical errors are amplified and the unbalanced ones are suppressed. Moreover, the designated meet Elements in 7 the same tasks as the designated items in 6 ,

Is located in place of the mirror 18 in 7 a sphere with a distance to the specimen that is shorter than the focal length f of the specimen 12 , then the light beam passes through the same optical path, highlighting other aberrations (see: D. Malacara "Optical Shop Testing", John Wiley and Sons, New York [1978], p.62).

8th shows a known structure of a wavefront sensor in Auflichtaufbau, but which is equipped with a modified according to the invention microlens array.

9 shows a known structure of a wavefront sensor in the transmitted light structure, but which is equipped with a modified according to the invention microlens array.

10 shows an intensity section through the pupil center along a series of intensity peaks measured with a diffractive microlens array according to the invention. The diffractive microlens array is used in the arrangements according to the 8th or 9 as a microlens array 7 used. With a CCD camera and in a measuring process, the wavefront and the edge of the pupil are determined.

According to the prior art, the highest possible efficiency of the diffractive optics is desired. However, the diffractive structures of the microlens matrix are designed here with an average efficiency of 70%. With the wavefront sensor 4 Then the center of gravity positions are determined from the (diffracted) focused intensity peaks and from the undeflected image the pupil edge.

By way of illustration is in 10 an intensity slice through the pupil along a series of intensity peaks (comparable to the carriage-wise thick bar in FIG 3 ). The finite efficiency results in an intensity background 21 on which the intensity peaks 20 lie. With the efficiency of the diffractive structure, the ratio of the intensity peaks 20 to the intensity background 21 be varied. Based on this background, the center and the exact edge of the pupil can 24 without an additional CCD camera or other technical means can be accurately determined. Thereafter, as known from the prior art (eg in: B. Schäfer, K. Mann: "Laser beam characterization with the Hartmann-Shack wavefront sensor" Laser Opto 32 (6) / 2000) known, the focal points of the intensity peaks for calculation Ideally, a threshold above the intensity background is used so that it has no effect on the result, ideally a CCD camera with a resolution of 10 to 16 bits is used so that the intensity background and focused intensity peaks can be separated sufficiently.

11 shows a refractive microlens array 7 in which the individual refractive microlenses 22 are designed so that they fill only a part of the subpupils. The degree of filling in the example shown is about 47%. The refractive microlens matrix with transparent spaces 23 between the individual microlenses 22 is used in the arrangements according to the 8th or 9 as a microlens array 7 used. Here, the light is from the spaces 23 between the microlenses 22 directly to the CCD camera 3 used to determine the radius and position of the pupil, thereby increasing the accuracy of the wavefront sensor.

So far has been in the wavefront sensors of the prior art this Light content as disturbing considered and was in the evaluation algorithm of the actual Intensity peaks separated as well as suppressed.

The not 100% degree of filling is in 11 through the gaps 23 visible, noticeable. The diameter of the microlenses can be reduced so much (thus reducing the degree of filling) that around each ideal intensity peak, a border with the radius of the microlens is formed (idealized description without diffraction). With the size of the border falls the dynamic range, ie it reduces the maximum deflection. Therefore, the border is on the one hand chosen so large that the position of the pupil 24 On the other hand, the scanning range of the intensity peaks is sufficiently wide.

By the reduction of the individual pupil sizes on the microlens array - without the Increase packing density - will be the Space between each lens directly to the CCD camera displayed. In this solution results in the intensity distribution, at every ideal focal point in the radius of the microlens pupil has a border. Due to the border and the resulting limited storage distance is this technique especially for testing systems with very low image errors.

11 shows top left next to the microlens array 7 the measured intensity gradients, which are measured for illustration only in two different levels.

The upper left intensity curve shows a comparison measurement of the intensity background 21 in the space between the individual refractive microlenses 22 ,

The left lower intensity curve shows the measurement through the cross section of a microlens array with a CCD camera according to the invention. The intensity peaks of higher intensity are the actual Wellenfrontmeßpunkte. The intensity peaks of lesser intensity correspond to the light passing through the spaces 23 between the refractive microlenses 22 passes through and represents an intensity background 21 ' , the exact intensity background 21 the comparison measurement corresponds. The intensity peaks of the intensity background 21 ' are used to determine the diameter of the pupil 24 evaluated by the computer system. Thus, the wavefront and the edge of the pupil are determined with a CCD camera and in a measuring process.

1

wavefront a pupil

2

shadow mask

3

(first) CCD camera

4

Wavefront sensor

5

view on the CCD camera

6

computer system

7

Microlens array

8th

(First) beamsplitter

9

relay optics

10

second CCD camera

11

field stop

12

examinee

13

lighting

14

condenser

15

pinhole

16

collection optics

17

second beamsplitter

18

mirror

19

diffractive single lens

20

Intensity peak

21

Intensity Underground

22

refractive microlens

23

gap

24

pupil

HH '

all levels of the test piece

f

focal length of the test piece

Claims (9)

Method for determining aberrations by calculating the Zernike coefficients of a specimen ( 12 ) with a computer program using a computer system ( 6 ) with a first CCD camera ( 3 ), the first CCD camera ( 3 ) a microlens array ( 7 ), characterized in that the edge of the pupil is in position and size in a computer system ( 6 ) is detected and determined in addition to the wavefront measurement by the first CCD camera ( 3 ) or from a second CCD camera ( 10 ) Light is registered, which is imaged directly from the test object and these measured values are evaluated by the computer program of the computer system in the determination of the Zernike coefficients with. A method according to claim 1, characterized in that the determination of the edge of the pupil by measurements of the second CCD camera ( 10 ) is performed in a Auflichtaufbau. A method according to claim 1, characterized in that the determination of the edge of the pupil by measurements of the second CCD camera ( 10 ) is performed in a transmitted light structure. A method according to claim 1, characterized in that the determination of the edge of the pupil by use of a diffractive microlens array ( 7 ), which has an efficiency of less than 100%, preferably between 40% and 80%, and is thus partially transparent, with the first CCD camera ( 3 ) is carried out A method according to claim 1, characterized in that the determination of the edge of the pupil by use of a refractive microlens array ( 7 ), which has a degree of filling of less than 100%, preferably between 40% and 80%, and which is transparent within the spaces created between the individual lenses, with the first CCD camera (FIG. 3 ) is carried out. Arrangement for carrying out the method according to Claim 2, having a transmitted-light structure consisting of a light source ( 13 ) with condenser ( 14 ), a pinhole ( 15 ) in an object plane of a test object ( 12 ) and a relay optics ( 9 ) with a field stop ( 11 ), which has a pupil of the test specimen ( 12 ) to an array of the first CCD camera ( 3 ), the first CCD camera ( 3 ) with the computer system ( 6 ), wherein in front of the array of the first CCD camera ( 3 ) a microlens array ( 7 ) and thus a wavefront sensor ( 4 ), characterized in that between the relay optics ( 9 ) and the wavefront sensor ( 4 ) a first beam splitter ( 8th ), which is the pupil ( 1 ) of the test piece ( 12 ) on both the wavefront sensor ( 4 ) as well as a second CCD camera ( 10 ) and the second CCD camera ( 10 ) with the computer system ( 6 ) connected is. Arrangement for carrying out the method according to Claim 3, having a reflected light structure consisting of the light source ( 13 ), as well as the relay optics ( 9 ) with the field stop ( 11 ) showing the pupil of the test specimen ( 12 ) to an array of the first CCD camera ( 3 ), the first CCD camera ( 3 ) With the computer system ( 6 ), wherein in front of the array of the first CCD camera ( 3 ) a microlens array ( 7 ) and thus a wavefront sensor ( 4 ) and - seen in the propagation direction of the light - the image of the pinhole ( 15 ) via a second beam splitter ( 17 ), about the test piece ( 12 ) on a reflector and from this again through the test specimen ( 12 ), via the second beam splitter ( 17 ) to the relay optics ( 9 ) is transferable, wherein the reflector within, before or after the focus of the specimen ( 12 ), characterized in that between the relay optics ( 9 ) and the wavefront sensor ( 4 ) the first beam splitter ( 8th ), which is the pupil of the specimen ( 12 ) on both the wavefront sensor ( 4 ) as well as the second CCD camera ( 10 ) and the second CCD camera ( 10 ) with the computer system ( 6 ) connected is. Arrangement for carrying out the method according to claim 4, with the wavefront sensor ( 4 ) in the transmitted light structure or in Auflichtaufaufbau in which the microlens array ( 7 ) over the CCD camera array ( 3 ), characterized in that the microlens array ( 7 ) is constructed of partially transparent diffractive optical elements, wherein the diffractive optical elements specifically have an efficiency below 100%, preferably below 95%, in particular between 20% and 80%. Arrangement for carrying out the method according to claim 5, with the wavefront sensor ( 4 ) in the transmitted light structure or in Auflichtaufaufbau in which the microlens array ( 7 ) over the CCD camera array ( 3 ), characterized in that between the individual lenses of the refractive microlens array ( 7 ) transparent interspaces are provided, which have a fill factor of the microlens array ( 7 ) produce less than 100%, preferably less than 95%, in particular between 40% and 90%.