CN100342400C - System and method for detecting differences between complex images - Google Patents

System and method for detecting differences between complex images Download PDF

Info

Publication number
CN100342400C
CN100342400C CNB038248700A CN03824870A CN100342400C CN 100342400 C CN100342400 C CN 100342400C CN B038248700 A CNB038248700 A CN B038248700A CN 03824870 A CN03824870 A CN 03824870A CN 100342400 C CN100342400 C CN 100342400C
Authority
CN
China
Prior art keywords
complex pattern
complex
modified
pattern
aberration
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CNB038248700A
Other languages
Chinese (zh)
Other versions
CN1695165A (en
Inventor
E·弗尔克尔
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
nLine Corp
Original Assignee
nLine Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by nLine Corp filed Critical nLine Corp
Publication of CN1695165A publication Critical patent/CN1695165A/en
Application granted granted Critical
Publication of CN100342400C publication Critical patent/CN100342400C/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Images

Landscapes

  • Holo Graphy (AREA)

Abstract

A system and method for detecting differences between complex images are disclosed. The method includes acquiring a first complex image and a second complex image and determining if an aberration value difference exists between the first and second complex images. The aberration value difference is corrected by iteratively modifying the first complex image by an aberration function and comparing the modified first complex image with the second complex image in a high frequency range. The method further determines if the modified first complex image matches the second complex image by modifying the second complex image with a low frequency ratio to replace low frequency components of the second complex image with low frequency components of the first complex image. The high frequency components of the modified first complex image and the modified second complex images are then compared to determine if the first complex image matches the second complex image.

Description

Detect the system and method for differences between complex images
TECHNICAL FIELD OF THE INVENTION
The present invention relates in general to image processing field, more particularly, relates to the system and method that is used to detect the difference between the complex pattern (complex image).
Background of invention
In a Direct Digital formula holophotal system, can obtain hologram from highly similar object.Serial processing to hologram allows the relatively real image ripple of object.These image wave are compared the little and big detailed information that obviously contains more object with traditional non holographic image, because kept the phase information of image in hologram, and do not keep the phase information of image in traditional image.
In order to find the nuance of highly similar object, importantly holophotal system to be remained on a stable status.Yet, obtaining between the hologram related with highly similar object during this period of time, some little variations may appear in the system.Because object may not have accurately to be placed in the same manner different positions, for example object may be in different focal lengths in different positions, so also other variations may occur.Distinctive variation of system and the change relevant with the position can cause to occur pseudomorphism (artifact) (for example, Jia difference or virtual difference) when whether having difference between definite these objects.If distinctive variation of system and/or the variation relevant with the position do not occur, then measured difference can clearly reflect the actual difference between the object.
In such as the fault detection of semiconductor wafer some are used, can obtain a plurality of images from the diverse location that is considered on the identical object with holophotal system.Though being in the object of diverse location may be highly similar, the aberration value in each position, for example focal length can be different, and therefore the image that obtains from different positions may look like different.In traditional image-taking system, can obtain an image from position on the object.Then, system moves to that another comprises a position with image of roughly the same feature on the object.System obtains an image in the second place, and focal length value is poor between definite again first and second image.If the focal length value difference between two images, system just the focal length value related with second image adjusted to the related focal length value of first image near coupling, use focal length value to obtain second image again through adjusting.The above time of this process need twice obtains image from the second place, thereby has increased and obtain the relevant cost of a plurality of images from an object.
Hologram image is different with real image, because hologram image contains intensity and phase information, and real image only contains strength information.Phase information additional in the holographic imaging has increased a new multiple dimension, also increased the new Flame Image Process instrument that is above standard may and ability.For example, the wave front matching capacity is for the few of value of intensity image (for example real image), and is important for image wave (for example complex pattern), because they have considered the phase place in the non-existent image wave in the intensity image.
Complex pattern is similar to real image, comprises HFS and low frequency part.Usually, the actual difference of image will occur in the high fdrequency component of image.Yet any change or system distinctive change relevant with the position can cause false difference or virtual difference in the high fdrequency component of image and low frequency component.In some complex patterns, the low frequency part of two different images can be owing to the distinctive slight change of the system such as small air turbulence difference.The low frequency difference can cause the difference of the vacation between the image, and the system that makes can not determine the actual high frequency difference between the image exactly.
In standard picture is handled, can obtain the low frequency part of two images again with a low-pass filter to two image applications Fourier filtering devices.Then, can time domain be returned in image transformation with contrary Fourier filtering device, so that can more described two images.Yet, owing in image wave, comprise additional phase information, so this solution is not removed the low frequency part of complex pattern.
Brief summary of the invention
Can significantly reduce in accordance with the teachings of the present invention or shortcoming and problem that elimination is related with the difference between the image.In a specific embodiment, the method that is used to detect differences between complex images comprises by revising the difference that first complex pattern comes the aberration correction value with an aberration function, again modified images is compared with second complex pattern, thereby make between the complex pattern difference reduce to minimum at high-frequency range.
According to one embodiment of the present of invention, the method that is used for the difference between the detected image comprise obtain first complex pattern and second complex pattern and with an application of low pass filters in the ratio of first and second complex pattern to obtain a low frequency ratio.Second complex pattern by low frequency than correct, so that the low frequency component of second complex pattern low frequency component with first complex pattern is replaced.Then, modified complex pattern is compared with first complex pattern, to determine whether second complex pattern mates with first complex pattern.
According to an alternative embodiment of the invention, the system that is used for the difference between the detected image comprises the processing resource that is used for obtaining the numeroscope of first complex pattern and second complex pattern and is connected to this numeroscope.Handle resource with the ratio of an application of low pass filters, to obtain a low frequency ratio in first and second complex pattern.Second complex pattern by low frequency than correct, so that the low frequency component of second complex pattern low frequency component with first complex pattern is replaced.Then, modified complex pattern is compared with first complex pattern, to determine whether second complex pattern mates with first complex pattern.
According to another embodiment of the present invention, the method that is used to detect differences between complex images comprises to be obtained first complex pattern and second complex pattern with similar features and is that first complex pattern is selected a plurality of aberration value in the aberration scope of an expection.Be that each aberration value calculates an aberration function, and with each aberration function iterative modifications first complex pattern.Modified complex pattern is compared with second complex pattern, and by selecting to make the aberration value of the modified complex pattern and the second differences between complex images minimum determine an aberration correction value.
According to an alternative embodiment of the invention, the system that is used to detect differences between complex images comprises numeroscope that is used for obtaining first complex pattern with similar features and second complex pattern and the processing resource that is connected to this numeroscope.Handling resource is that first complex pattern is selected a plurality of aberration value in the aberration scope of an expection, and is aberration function of each aberration value calculating.With each aberration function iterative modifications first complex pattern, and modified complex pattern compared with second complex pattern.Handle resource by selecting to make the aberration value of the modified complex pattern and the second differences between complex images minimum determine an aberration correction value.
According to an alternative embodiment of the invention, the method that is used to detect differences between complex images comprises to be obtained first complex pattern and second complex pattern with similar features and determines whether there is aberration value difference between first and second complex patterns.By with aberration function iterative modifications first complex pattern with at high-frequency range the modified first complex image and second complex pattern are compared the difference of aberration correction value.This method also by low frequency than revising second complex pattern so that the low frequency component of second complex pattern is replaced determining with the low frequency component of first complex pattern whether modified first complex image mates with second complex pattern.Then, the high fdrequency component of modified first complex image is compared with the high fdrequency component of modified second complex pattern, to determine whether first complex pattern mates with second complex pattern.
The important technical advantage of some embodiments of the present invention comprises that minimizing obtains the Direct Digital formula system of the needed time quantum of a plurality of images from an object.In some applications, this system diverse location that can be used for from the object obtains image.The aberration value related with each image that is obtained can be different.This system uses the aberration value related with one second image to adjust first image, rather than obtains an image with the aberration value through adjusting again.
Another important technical advantage of some embodiment of the present invention comprises the Direct Digital formula holophotal system of eliminating pseudomorphism from an image that is obtained.In this system may small variation appear obtaining first image and obtain between time of second image.These small variations appear in the low frequency component of the image that is obtained usually.This system in the ratio of two images that obtained, multiply by one of them image this than to eliminate low frequency component from image relatively with an application of low pass filters again.Therefore, be the high fdrequency component of movement images, this makes this system can accurately determine whether there is real difference between two images.
In various embodiment of the present invention, may present all these technological merits, some of them advantage or not present these advantages.For a person skilled in the art, from the following drawings, instructions and claims, will be clear that other technologies advantage of the present invention easily.
The accompanying drawing summary
By with reference to can more fully understanding the present invention and advantage thereof in the explanation of doing below in conjunction with accompanying drawing, in these accompanying drawings, same Reference numeral is represented same feature, wherein:
The synoptic diagram of Fig. 1 explanation Direct Digital formula holophotal system in accordance with the teachings of the present invention;
The synoptic diagram of Fig. 2 explanation another Direct Digital formula holophotal system in accordance with the teachings of the present invention;
Two complex patterns that Fig. 3 explanation obtains by a Direct Digital formula holophotal system and definite in accordance with the teachings of the present invention and use the image that obtains behind the aberration correction value;
The complex pattern that Fig. 4 explanation obtains under the situation that the pseudomorphism that variation of holophotal system is not caused compensates by a Direct Digital formula holophotal system;
The complex pattern of Fig. 5 key diagram 4 is being eliminated the situation behind the pseudomorphism in accordance with the teachings of the present invention; And
Fig. 6 explanation is used to detect the process flow diagram of the method for differences between complex images in accordance with the teachings of the present invention.
Describe in detail
By understanding the preferred embodiments of the present invention and advantage thereof best referring to figs. 1 to 6, wherein same Reference numeral is used for representing similar and corresponding part.
Following the present invention relates in general to digital holographic imaging systems and application, digital holographic imaging systems and using as described in following: title is the U.S. Patent No. 6 of " Direct-to-Digital Holography andHolovision (Direct Digital formula holography and hologram television) ", 078,392, title is the U.S. Patent No. 6 of " Acquisition and Replay Systems forDirect-to-Digital Holography and Holovision (collection of Direct Digital formula holography and hologram television and playback system) ", 525,821, title is the U.S. Patent application sequence No.09/949 of " System and Method for Correlated Noise Removal in ComplexImaging Systems (being used to eliminate the system and method for correlation noise in the multiple imaging system) ", 266 and title be the U.S. Patent application sequence No.09/949 of " System and Methodfor Registering Complex Images (being used to register the system and method for complex pattern) ", 423, all these here are incorporated herein by reference.
Fig. 1 illustrates the synoptic diagram of Direct Digital formula holophotal system 10.System 10 comprises laser instrument 12, optical beam expander/spatial filter 14, lens 16, beam splitter 18, target 20, condenser lens 22 and catoptron 24.In an illustrated embodiment, laser instrument 12 is a light beam directive extender/wave filter 14, advances to beam splitter 18 through the light scioptics 16 of expansion/filtering.Beam splitter 18 can be any optical element that can make a part of light beam transmission and a part of beam reflection.In one embodiment, beam splitter 18 can be one 50/50 beam splitter, wherein makes about 50 (50%) percent reflection and about 50 (50%) percent transmission in the light beam.In other embodiments, beam splitter 18 can reflect and/or the light of any suitable number percent of transmission.Beam splitter 18 can be including, but not limited to cube splitter and plate beam splitter.
The light through expansion/filtering by beam splitter reflection constitutes the target beam 26 of advancing to target 20.In one embodiment, target 20 can be an electron device made from silicon, germanium or any compound that contains the element of III of family and/or the V of family.In another embodiment, target 20 can be a photomask or a graticule that is included in the pattern that forms on the substrate.In other embodiments, target 20 can be any one object that can produce complex pattern, assembly or element.Advance by beam splitter 18 and to condenser lens 22 then from the part of the light of target 20 reflection.Condenser lens 22 can be used for target 20 is focused on the focal plane of numeroscope (clearly not illustrating).Condenser lens 22 can also be on demand by the lens that use different focal with adjust corresponding space geometry structure (for example ratio of object distance and image distance) and amplify or dwindle.The light of line focus advances on the numeroscope then.In one embodiment, numeroscope can be a high resolving power charge-coupled device (CCD) video camera, the hologram that it can write down and reset and obtain from target 20.Numeroscope can also with one comprise computing machine (the clearly not illustrating) interface of handling resource.In one embodiment, handling resource can be or combination in microprocessor, microcontroller, digital signal processor (DSP) or any other digital circuits that are configured to process information.
The part of passing through beam splitter 18 transmissions from the light of lens 16 constitutes reference beam 28.Reference beam 28 reflects from benchmark catoptron 24 with a little angle.Advance to beam splitter 18 then from the reference beam of benchmark catoptron 24 reflections.Advance to condenser lens 22 then in the reflecting part by beam splitter 18 reflections of reference beam.Come the reference beam of GRIN Lens 22 to advance on the numeroscope then.Come the object beam of GRIN Lens 22 and benchmark and the object light wave 30 that reference beam is formed a plurality of whiles that form a hologram together.
System 10 can adopt " Mechelson, A. A. " geometry (for example, the geometric relationship of beam splitter 18, reference beam catoptron 24 and numeroscope is similar to the Michelson interferometer geometry).This geometry makes reference beam and object beam can merge with a very little angle at condenser lens 22 places.For example, benchmark catoptron 24 can tilt to cause with generation and to be used for hologram is carried out the space heterodyne of Fourier analysis or the low-angle of sideband striped.
Fig. 2 illustrates the synoptic diagram of another example embodiment of Direct Digital formula holophotal system 40.System 40 comprises laser instrument 12, variable attenuator 42, variable beam splitter 44, target-arm, reference arm, beam combiner 54 and numeroscope 56.Target-arm can comprise lens 52 in target beam extender 46, target beam splitter 48, target objective 50, target 20 and the target tube.Reference arm can comprise lens 64 in reference beam extender 58, benchmark beam splitter 60, benchmark object lens 62, benchmark catoptron 24 and the benchmark tube.In an illustrated embodiment, laser instrument 12 is with a light beam directive variable attenuator 42, and the light through decaying advances to variable beam splitter 44.The optical element of the part that variable beam splitter 44 can be a transmitted light beam and another part of folded light beam.In an illustrated embodiment, variable beam splitter 44 is divided into target beam 66 and reference beam 68 with light beam.
Still with reference to figure 2, target beam 66 is conducted through target beam extender 46 head for target beam splitters 48, and target beam splitter 48 reflects the part of target beam 66 towards target objective 50.So the target beam that is reflected interacts with target 20, and passes back by target objective 50.The part of the reflected target beam that target beam splitter 48 will receive from target objective 50 is transmitted to beam combiner 54 by lens 52 in the target tube.In reference arm, from the reference beam 68 of variable beam splitter 44 by reference beam extender 58, and by 60 reflections of benchmark beam splitter.Benchmark object lens 62 are passed through in the reflecting part of reference beam 68, and by 24 reflections of benchmark catoptron.The reference beam of reflection is passed back by benchmark object lens 62 then, and by 60 transmissions of benchmark beam splitter.Lens 64 are guided light beam into beam combiner 54 in the benchmark tube, and what beam combiner 54 will be from target-arm and reference arm is also photosynthetic, and the light beam through merging is mapped on the numeroscope 56.In one embodiment, the light beam through merging can be the numerical data of institute's record, transmission and/or conversion.
System 40 can adopt the Mach-Zehnder geometry.As seen the Mach-Zehnder geometry (why being called Mach-Zehnder is because it is similar with the geometry of Mach-Zehnder interferometer) of Fig. 2 is compared with (as shown in Figure 1) Mechelson, A. A. geometry, condenser lens (for example target objective among Fig. 2 50) can more close target 20, because the irradiation of scioptics allows target beam splitter 48 to be arranged on target objective 50 backs rather than between target objective 50 and target 20.This permission use large-numerical aperture, the object (with the hologram of record wisp) that the high-amplification-factor object lens are little.For big object, original Mechelson, A. A. geometry as shown in Figure 1 can be preferred, and this depends on concrete condition.
Can also see that from Fig. 2 beam combiner 54 is arranged near numeroscope 56.Beam combiner 54 can combine reference beam 66 and object beam 68 and shine on the numeroscope 56.The angle of beam combiner 54 can change, so that make reference beam and the accurate conllinear of object beam, perhaps projects at an angle on the numeroscope 56 mutually usually, so that produce the heterodyne carrier fringe.This allows the carrier fringe frequency from the zero nyquist limit that changes to numeroscope 56.Beam combiner 54 can be used to amplify geometry (for example, amplifying the object hologram so that the geometry that digital camera obtains) at least than the more close numeroscope 56 with the Mechelson, A. A. geometry.This makes the merging angle between object beam and the reference beam can be relatively large and can not make the luminous point of reference beam and object beam no longer overlapping at numeroscope 56 places.This allows the carrier frequency striped is carried out much meticulous control.In fact, the maximum angle that system restriction allowed that the angle between two light beams can be no more than the nyquist frequency that numeroscope allows from the zero spatial carrier frequency that changes to heterodyne hologram always (for example, this angle can be increased, each striped up to the spatial carrier frequency has only two pixels, surpasses just no longer correct record space carrier frequency of this angle numeroscope).Adopt the Mechelson, A. A. geometry, the maximum space carrier frequency of hologram is not accessible just, because for some geometries, required angle no longer overlaps on numeroscope even as big as making reference beam and object beam.
In operation, system 10 and 40 can be fit to real time record and replay holograms picture or they are stored be provided with the back and reset.The hologram of a series of stored digital can be made a holographic movie, perhaps can be with described hologram electricity consumption submode broadcasting, for far resetting so that hologram television (HoloVision) to be provided.Because hologram has been stored amplitude and phase place, and phase place is directly proportional with wavelength and optical length, so Direct Digital formula holophotal system 10 and 40 also can be as the extremely accurate survey instrument of the shape and size of checking precision component, assembly etc.Similarly, immediately digitally the ability of stored hologram a kind of method that is used for the digital hologram interferometry is provided.Same object some holograms after some physics change (stress, temperature, little processing etc.) can be subtracted each other (Direct Phase is subtracted each other) mutually to calculate the physical measurements values of change, wherein the change of phase place is directly proportional with wavelength.Similarly, an object can be compared with a similar object, measure the deviation of second object with respect to first object or master object by corresponding hologram is subtracted each other.In order two pixels in the x-y plane clearly to be measured in z-plane greater than the phase change of 2 π, hologram should be with the enterprising line item of more than one wavelength.
System 10 and 40 is used in combination the high-resolution digital register such as video camera, (for example mix holographic object wave and reference wave with very little angle, to cause at least two pixels of each striped and mix) for the angle of each at least two striped of space characteristics that need differentiate, make object imaging on record (video camera) plane, and space low frequency heterodyne (sideband) hologram carried out fourier transform analysis, thereby can recording holographic image (for example, each pixel being recorded the image of phase place and amplitude).In addition, can in the back focal plane of one or more lens that object focus is used, use an aperture diaphragm to prevent the aliasing of any frequency higher than the distinguishable frequency of imaging system.If this imaging system of all spatial frequencys of object is all distinguishable, so just do not need the aperture.
Behind these hologram images record, the phase place or the amplitude image that can be used as 3D are reset on the display of one 2 dimension, perhaps with the ripple of the complete original record of a phase twin crystal and white light or laser playback with the original image of resetting.The original image of resetting be with laser with its recording phase change medium, reset it with white light or another laser again.Image by resetting with the image of laser log and the merging of three different colours might produce a very color hologram.By constantly writing and reset a series of images, might form holographic movie.Because therefore these images can also broadcast with radio frequency (RF) ripple (for example microwave) or by the digital network of optical fiber or cable with suitable digital coding with the digital form record, and reset in distant.This has realized hologram television and film or " HoloVision " effectively.
System 10 and 40 also can realize with many alternative methods.For example, system 10 and 40 can adopt phase shift rather than obtain for the phase place of each pixel of hologram and the heterodyne of amplitude.In another embodiment, system 10 and 40 can be written to intensity pattern on the photosensitive crystal with many diverse ways.These methods comprise the scanning laser beam (rather than usage space photomodulator) that uses sharp focus, with SLM but the laser beam of not setovering write, also can use many feasible change for how much of writing scheme to write.In another embodiment, system 10 and 40 can utilize optical effect rather than phase transformation to produce the diffraction grating of replay holograms with photosensitive crystal.In yet another embodiment, system 10 and 40 can produce intensity pattern with the SLM of a pixel very fine, thereby has avoided intensity pattern being written to replay holograms on the gyrotropi crystal.
As mentioned above, system 10 and 40 can be used for relatively complex pattern that obtains from same object or target or the complex pattern that relatively obtains from different targets after target generation physics changes.In addition, system 10 and 40 can be used for that different positions obtains image on the target 20.Image at diverse location can have similar feature.For example, target 20 can be a semiconductor wafer that comprises a plurality of examples of single die.In this example, system 10 and 40 can be used for obtaining the complex pattern of a specific region of each chip.Though the image that is obtained can have similar feature, the grade dizzy aberration value (for example focal length of single order aberration item) related with each image may be different.That this difference can cause is virtual, non-existent difference between image.Therefore, the aberration value of an image in these images should be adjusted to guarantee that the complex pattern that is obtained can accurately be compared.
The invention provides a kind of aberration correction value and do not increase and obtain the complex pattern solution of required time.System 10 and 40 can be used for that two different positions obtain two complex patterns on the target 20.At first, can determine the aberration scope, so that make the difference of the aberration between two images have a value between the determined limit.Can select one or more aberration value in determined aberration scope, and be that each selected value is calculated aberration function.Secondly, can be by taking advantage of each aberration value to come iterative modifications first complex pattern system 10 and/or 40 first complex patterns that obtain, so that obtain a modified first complex image for each aberration value of calculating.The 3rd, each modified first complex image can be compared with second complex pattern.Obtain between the modified first complex image and second complex pattern optimal approximation that comparison shows that of in high-frequency range difference minimum is used for proofreading and correct the difference of two aberrations between the complex pattern.In one embodiment, this process selects aberration value to be improved near can being used in the initial aberration value of determining more subtly.In another embodiment, this process can with two optimal approximations of best aberration value and between described two optimal approximations better aberration value of interpolation improved.
Except the aberration value difference between the image that is obtained, can occur other testing apparatuss that can in complex pattern, cause pseudomorphism in the system 10 and 40 and change.These pseudomorphisms can be very little, and can appear in the low frequency component of complex pattern.On the contrary, the difference between two similar objects is very little dimensionally usually, therefore mainly comprises the high frequency spatial component.In the Flame Image Process of standard, can in Fourier space, eliminate any because the pseudomorphism that the variation in the image processing system causes.Yet, in Fourier space, can not eliminate the pseudomorphism that causes owing to the variation in the system 10 or 40, because each pixel in the Fourier space all is to compose convolution with (the answering) of variation.
Suppose pseudomorphism to appear at the low frequency space component and any true difference between the similar object appears at the high frequency spatial component, then poor (the calculating according to hologram) of the low frequency component that the variations in the system 10 and 40 can be by calculating complex pattern comes approximate.In order to compensate described variation, can be with the ratio of an application of low pass filters in complex pattern.Then, can be with the multiplication factor of gained result with one of them image of opposing, its variation to system compensates.
Mathematical description of the present invention
Direct Digital formula holophotal system such as system 10 and 40 can write down some complex patterns or the hologram in the numeroscope plane.X and y coordinate representation can be used in the register plane.According to the hologram that is write down, can be by the complex pattern ripple being used the complex pattern ripple on Fourier transform calculating or reconstruct register plane.Can contain dizzy aberration such as grade such as the focal length of single order aberration item as Fourier transform according to the described ripple of hologram reconstruction.For example, image wave ψ (x, can be write as by Fourier transform y) (FFT)
FFT{ψ(x，y)}＝FFT{ψ′(x，y)}exp[ix(q x，q y)] (1)
I wherein 2=-1, q x, q yBe the coordinate in the Fourier plane, and ψ ' (x y) is complex pattern ripple different aberration value.X is the aberration function for each selected aberration value, and FFT and IFFT represent the positive and contrary conversion of Fourier respectively.In order to determine little aberration poor between similar or identical image, should more at least two images.
In one embodiment, system 10 and/or 40 can obtain two complex pattern ψ j(x, y) and ψ J+1(x, y), wherein j is an integer.Can use Fourier transform to the two, and be write as the form that the actual aberration value that makes them can be separated:
FFT{ψ j(x，y)}＝FFT{ψ j′(x，y)}exp[ix j(q x，q y)] (2)
FFT{ψ j+1(x，y)}＝FFT{ψ j+1′(x，y)}exp[ix j+1(q x，q y)] (3)
According to these formula, the difference of the aberration between two images can be described as:
FFT{ψ j+1(x，y)}/FFT{ψ j(x，y)}＝ (4)
exp[ix j+1(q x，q y)]/exp[ix j(q x，q y)]＝
exp[i(x j+1(q x，q y)-x j(q x，q y))]＝
exp[iΔx j+1，j(q x，q y)]；
ψ wherein j(x, y) ≈ ψ J+1(x, y).If image ψ j(x y) does not comprise and image ψ J+1(x, y) similar feature, then aberration function x (q x, q y) can not directly available.Yet, for similar image, ψ j(x, y) ≈ ψ J+1(x, y), then following formula is rational being similar to.Therefore, if image ψ J+1(x, y) and ψ j(x, difference ψ y) J+1(q x, q y)-x j(q x, q y) known, then first complex pattern just can be modified as:
ψ j′＝IFFT{FFT(ψ j)*exp[iΔx j+1(q x，q y)]} (5)
This is approximate and ψ in aberration value J+1Identical the time ψ jTherefore, by with image ψ jBe modified as and image ψ J+1Aberration value coupling, set up the step of the difference of eliminating two aberrations between the similar image.
Yet it is known that following formula requires the aberration value between two complex patterns.In one embodiment, can calculate image ψ j(x, y) and ψ J+1(x, y) the difference Δ ψ of the aberration between J+1, jIf image ψ j(x, y) and ψ J+1(x, y) similar, if these two images are with the coupling aberration value then any detected difference can minimum so.This hypothesis can be important for the height periodic structure, because the coupling aberration value can occur in some periodic aberration value.Yet in most applications, real aberration value also provides the optimum matching between the image, thereby remains and can discern uniquely.
In described embodiment, can find image ψ j(x, y) and ψ J+1(x, y) difference of the aberration between is at a given scope [x Min, x Max] in.In one embodiment, described specific aberration can be single order aberration, for example a focal length.What provide less than deriving in this embodiment, for high with the effective aberration function of low scattered through angles is:
x(q x，q y)＝(2∏/λ)Δz?sqrt((1-q 2λ 2)-1) (6)
Perhaps use standard equation for focal length
x(q x，q y)＝(2π/λ)Δz?q 2λ 2 (7)
Wherein λ is a wavelength, the focal length value of Δ z in focusing range, selecting, and q = sqrt ( q x 2 + q y 2 ) 。In order to determine actual aberration poor between the complex pattern, can in a preset range, select at least two aberration value.Can calculate the aberration function related then with each aberration value.So can be with each substitution following formula of the aberration value that these calculated to calculate an image of revising with aberration function.
ψ′ 1 j＝IFFT{FFT(ψ j)exp[ix 1(q x，q y)]} (8)
Can calculate a modified images with the aberration function of each calculating, again with each modified images and image ψ J+1(x y) compares, for example by calculating the variance Δ of following formula 1 J, j+1:
Δ 1 j，j+1(x 1)＝var[mod(ψ′ 1j/ψ j+1)] (9)
In case determined to have minimum Δ 1 J, j+1X 1O perhaps by interpolation between the value of two minimums, now just can know the poor of aberration, and can eliminate continuously.Need two ripples relatively finally to be:
ψ′ 1 j＝IFFT{FFT(ψ j)*exp[ix 1o(q x，q y)]}； (10)
ψ j+1(x，y) (11)
Because the aberration value of two complex patterns is approximate identical, therefore these two images can be compared accurately in high-frequency range, to determine whether to exist any actual difference.By adjusting the aberration value related with first complex pattern, need not obtain second complex pattern again, obtain the required time quantum of a plurality of images thereby reduced from target 20.
As mentioned above, the pseudomorphism that is caused by the peculiar variation of system can make the actual difference distortion between the image.The peculiar variation of system that these may take place between two complex patterns typically has the spectrum that is confined to lower frequency, and the actual difference between target 20 and the hologram that obtains typically is in upper frequency.In general, two complex pattern ψ as basis hologram reconstruction separately j(x, y) and ψ J+1(x y) can be expressed as:
ψ j(x，y)＝a jexp(i j) (12)
ψ j+1(x，y)＝a j+1exp(i j+1) (13)
Image ψ j(x is y) with image ψ J+1(x, difference y) can be expressed as following expression formula:
ψ j+1(x，y)＝a j′exp(i j′)*a xexp(i x) (14)
Wherein, a xExp (i φ x) expression is changed by the falseness between two images causing of variations in system 10 and/or 40, and a jExp (i φ i)～a j' exp (i φ j') show two similaritys between the complex pattern.
Because false the change just is confined to lower frequency, therefore can use a low-pass filter to the ratio of first and second complex pattern, as follows:
ψ x(x，y)＝IFFT{LPF[FFT(ψ j(x，y)/ψ j+1(x，y))]} (15)
Wherein, ψ x(x y) be being similar to that the falseness between two images that caused by any variation in system 10 and/or 40 changes, and LPF describes this low-pass filter.In one embodiment, low-pass filter can be a Butterworth wave filter.In other embodiments, low-pass filter can be the low-pass filter of the suitable type of any transmission low frequency component related with the image that is obtained.With ψ j(x, y) and ψ J+1(x, y) substitution following formula can get following result:
ψ x(x，y)＝IFFT{FFT{[a jexp(i j)]/[a j+1exp(i j+1)]}*LPF} (16)
ψ x(x，y)＝IFFT{[(a j/a j+1)exp[i( j 1- j+1 1)( j h- j+1 h)]]*LPF}
Wherein, for low frequency component LPF=1, for high fdrequency component LPF=0, φ j 1And φ J+1 1Difference presentation video ψ j(x is y) with image ψ J+1(x, low frequency component y), and φ j hAnd φ J+1 hDifference presentation video ψ j(x is y) with image ψ J+1(x, high fdrequency component y).Therefore, low-pass filter has been eliminated the high fdrequency component related with the ratio of image, thereby obtains low frequency part.The low frequency filtering result of low-pass filter is as follows:
ψ x(x，y)＝(a j/a j+1)exp[i( j 1- j+1 1)] (17)
Wherein, ψ x(x, the falseness that exists in y) the expression system 10 and/or 40 changes.So this result can be multiply by image ψ J+1(x, y), to obtain following modified images:
ψ j+1′(x，y)＝ψ j+1(x，y)*ψ x(x，y) (18)
ψ j+1′(x，y)＝a j+1exp[i( j+1 h+ j+1 1)]*(a j/a j+1)exp[i( j t- j+1 t)]
ψ j+1′(x，y)＝a jexp[i( j+1 h+ j 1)]
Modified images comprises image ψ J+1(x, high fdrequency component y) and image ψ j(x, low frequency component y) make with modified images and image ψ j(x, the high fdrequency component of only remaining following each image when y) comparing:
ψ j(x，y)＝ψ j+1′(x，y) (19)
a jexp[i( j 1+ j h)]＝a jexp[i( j h+ j t)]
Δψ j，j+1＝exp[i( j h- j+1 h)]
Wherein, Δ ψ J, j+1(x, y) presentation video ψ j(x, y) and ψ J+1(x, y) any actual difference between.Therefore, can eliminate the falseness change that system 10 and/or 40 causes in the low frequency component of each image, thereby can determine two actual difference between the complex pattern by the high fdrequency component that compares each image.
A plurality of complex patterns that Fig. 3 illustrative system 10 and/or 40 obtains.Specifically, the complex pattern that image 32 obtains for one on target 20 position that is in first focal length value, and the complex pattern that image 34 obtains for another position that is in second focal length value on target 20.The complex pattern of image 36 for behind the focus difference that calculates and use between described two complex patterns, representing with image 32.The aperture value that is used for obtaining described image is about 0.5nA, and focus correction is effective to high scattering angle.
Two differences between complex images that Figure 4 and 5 illustrative system 10 and/or 40 obtains.Specifically, Fig. 4 explanation two complex patterns poor under the situation that the pseudomorphism that is caused by system change is compensated not.As shown, this image comprises a plurality of pseudomorphisms that make image fault.Fig. 5 explanation difference images as shown in Figure 4 situation after the ratio of two images that obtained having been used (the above detailed description in detail) low-pass filter.As shown, can significantly reduce the pseudomorphism that minor alteration is introduced in system 10 and/or 40.In an illustrated embodiment, bright hickie is a defective, and undetectable defective can accurately be represented now in image shown in Figure 4, because all associated low frequency components are all eliminated by application of low-pass filters.
Fig. 6 a and 6b explanation is a kind of to be used to detect the process flow diagram of the method for differences between complex images.Usually, Direct Digital formula holophotal system can be used for obtaining the complex pattern of object of expression or target, and determines whether the image that is obtained has any actual difference.In image acquisition procedures, the variation of the precision of the image that obtains may appear in the holophotal system influencing.For example, if holophotal system two diverse locations from the object obtain similar images, each image that obtains can comprise the aberration value of a uniqueness so.In that the image that is obtained is actual when comprising identical feature, aberration value difference can make holophotal system determine that the image that is obtained is different.Do not influence the speed of image acquisition procedures in order accurately to obtain image, can adjust first image that is obtained iteratively, making wins obtains image and has second aberration value of obtaining image.So, can determine whether two images that obtained have similar feature with modified first image.Other variations in the holophotal system can be by calculating each image the difference of low frequency component be similar to.For the low frequency component of removal of images, can use a low-pass filter to the ratio of image, and the result can be used for revising one of them image, with the variation of compensation holophotal system.So can be in any difference of high fdrequency component detected image.
In step 70, system 10 or 40 can obtain one first complex pattern from target 20.In one embodiment, target 20 can be an electron device made from silicon, germanium or any compound that contains the element of III of family and/or the IV of family.In another embodiment, target 20 can be a photomask or a graticule that is included in the pattern that forms on the substrate.In other embodiments, target 20 can be any can analysis to check object, element or the assemblies of shape and size with system 10 and 40.In step 72, can obtain one second complex pattern from target 20.Second complex pattern can obtain from same object, and to calculate the physical change in this object, perhaps second complex pattern can obtain from a similar object, to measure the deviation of relative first object of second object.
In step 73, system 10 and/or 40 need to determine whether aberration correction to make first and second images match.If the aberration value of second image is different with the aberration value of first image, just determine the aberration scope of an expection in step 74.This desired extent can based on determined in the past with a value that certain objects is related, perhaps according to the estimation of the type (for example semiconductor wafer, photomask etc.) of object.In order to converge to the poor of actual aberration between first and second image, can in this scope, select at least two aberration value in step 76.In one embodiment, can select two optimum values, use the poor of an estimated aberration of described two optimum value interpolations again.In another embodiment, can select a plurality of values.In step 78, can calculate an aberration function with each selected aberration value.In one embodiment, can calculate a single order aberration value such as focal length with aberration function.
In step 80, with aberration function iterative modifications first complex pattern that calculates.In one embodiment, can use Fourier transform to first complex pattern, and this result be multiply by aberration function.Can repeat this process to the aberration function of each calculating, the complex pattern of winning can repeatedly be revised.After having used each aberration function, can carry out inverse fourier transform, so that time domain is returned in the modified first complex image conversion to modified first complex image.In step 82, then the modified first complex image related with each aberration function can be compared with second complex pattern, to determine aberration correction.In one embodiment, can compare image by the variance of calculating modified first complex image with the mould of the ratio of second complex pattern.
In step 84, can analyze the difference between the high fdrequency component of the modified first complex image and second complex pattern.If this difference is the minimum variance between these two images, just will calculate the optimal approximation that the used aberration value of this specific aberration function is elected aberration correction as in step 86.Otherwise, use first plural number related to revise first complex pattern in step 83 with another aberration value, in step 82 new modified complex pattern is compared with second complex pattern again.In case determined minimum variance between these two images just the aberration value related with this minimum variance to be applied to first complex pattern, so that obtain modified first complex image in step 86 with iterative process.
After obtaining a suitable aberration correction value, just can calculate the ratio of modified first complex image and second complex pattern in step 88.In one embodiment, the aberration value between first and second complex pattern can be similar, does not therefore need to adjust first complex pattern.In this embodiment, modified first complex image can be approximately equal to first complex pattern.These two complex patterns that obtained can contain the pseudomorphism that is caused by the variation in system 10 and/or 40.These pseudomorphisms may reside in the low frequency component related with image, and any actual difference may reside in the high fdrequency component.In order from the comparison of the modified first complex image and second complex pattern, to eliminate low frequency component effectively, can be in step 90 with an application of low pass filters in described ratio.In one embodiment, can be to described than using Fourier transform so that the image ratio is transformed to frequency domain, and in frequency domain application of low-pass filters.Can low frequency be returned time domain than conversion with inverse fourier transform then.
In step 92, the low frequency ratio can be multiply by second complex pattern to obtain modified second complex pattern.By the low frequency ratio is used for second complex pattern, just the low frequency component of second complex pattern low frequency component with first image is replaced.Can modified second complex pattern be compared with modified first complex image in step 94 then.This relatively in, just relatively with each related high fdrequency components of first and second complex patterns, so that determine any actual difference between image.In step 96, system determines whether the high fdrequency component of these two images is approximate identical.If between modified first complex image and modified second complex pattern, do not have difference, just determine that by system 10 and/or 40 image is similar in step 98.If define difference, just this difference is noted in step 100.In one embodiment, target 20 can be a semiconductor wafer, and the difference between the image that is calculated can show ad-hoc location defectiveness on wafer.
Though the invention has been described with regard to concrete preferred embodiment of the present invention, can propose various changes and modification for a person skilled in the art, be intended that and the present invention includes such change and the modification that belongs in the appended claims scope.

Claims (47)

1. method that is used for detecting differences between complex images comprises:
Obtain first complex pattern and second complex pattern;
With application of low pass filters in the ratio of first and second complex pattern to obtain the low frequency ratio;
Recently revise second complex pattern by this low frequency, so that the low frequency component of second complex pattern low frequency component with first complex pattern is replaced; And
Modified complex pattern is compared with first complex pattern, to determine whether second complex pattern mates with first complex pattern.
2. the method for claim 1 also comprises the Fourier transform of the ratio that calculates first and second complex pattern, so that in the frequency domain application of low-pass filters.
3. the method for claim 2 also comprises the inverse fourier transform that calculates the low frequency ratio, so that revise second complex pattern in time domain.
4. the process of claim 1 wherein the high fdrequency component that relatively comprises first and second complex pattern of comparison of the modified complex pattern and first complex pattern.
5. the process of claim 1 wherein that described first and second complex patterns comprise hologram image.
6. the method for claim 1 also comprises the described low frequency component related with second complex pattern, and presentation video obtains intrasystem system change.
7. method that is used for detecting differences between complex images comprises:
Obtain first complex pattern and second complex pattern;
The Fourier transform of ratio that calculates first and second complex pattern is to obtain a frequency domain ratio;
Application of low pass filters is compared to obtain a low frequency ratio in this frequency domain;
Calculate the inverse fourier transform of low frequency ratio, so that the low frequency ratio is transformed to time domain;
Recently revise second complex pattern by low frequency, so that the low frequency component of second complex pattern low frequency component with first complex pattern is replaced through conversion; And
Modified complex pattern is compared with first complex pattern, to determine whether second complex pattern mates with first complex pattern.
8. the method for claim 7, the high fdrequency component that relatively comprises first and second complex pattern of comparison of the wherein modified complex pattern and first complex pattern.
9. the method for claim 7, wherein this first and second complex pattern comprises hologram image.
10. the method for claim 7 also comprises can be used for reducing the low frequency ratio through conversion that the falseness in first and second complex patterns that image-taking system produces changes.
11. a system that is used for detecting differences between complex images comprises:
Can be used to obtain the numeroscope of first complex pattern and second complex pattern; And
Be connected to the processing resource of described numeroscope, described processing resource can be used to:
With an application of low pass filters in the ratio of first and second complex pattern to obtain the low frequency ratio;
Recently revise second complex pattern by low frequency, so that the low frequency component of second complex pattern low frequency component with first complex pattern is replaced; And
Modified complex pattern is compared with first complex pattern, to determine whether second complex pattern mates with first complex pattern.
12. the system of claim 11 also comprises the processing resource of the Fourier transform of the ratio that can be used to calculate first and second complex pattern, so that in the frequency domain application of low-pass filters.
13. the system of claim 12 also comprises the processing resource of the inverse fourier transform that can be used to calculate the low frequency ratio, so that revise second complex pattern in time domain.
14. the system of claim 11, the high fdrequency component that relatively comprises first and second complex pattern of comparison of the wherein modified complex pattern and first complex pattern.
15. the system of claim 11, wherein this first and second complex pattern comprises hologram image.
16. the system of claim 11, wherein said numeroscope comprises ccd video camera.
17. the system of claim 11 also comprises and the beam combiner of described numeroscope optically-coupled, can be used to receive reference beam and object beam to produce first and second complex patterns.
18. a method that is used for detecting differences between complex images comprises:
Obtain first complex pattern and second complex pattern, described first and second complex patterns comprise similar feature;
In the aberration scope of expection is that first complex pattern is selected a plurality of aberration value;
For each selected aberration value is calculated an aberration function;
By each aberration function iterative modifications first complex pattern;
Modified complex pattern is compared with second complex pattern; And
By selecting to make the aberration value of the modified complex pattern and the second differences between complex images minimum determine an aberration correction value.
19. the method for claim 18 comprises also first complex pattern is carried out Fourier transform that the feasible frequency domain that is modified in to first complex pattern carries out.
20. the method for claim 19 also is included in and modified complex pattern is carried out inverse fourier transform before modified complex pattern compared with second complex pattern.
21. the method for claim 18 is determined relatively the comprising of wherein modified complex pattern and second complex pattern variance of modified complex pattern and the mould of the ratio of second complex pattern.
22. the method for claim 21 determines that wherein aberration correction value comprises the ratio of selecting to make the variance minimum between the modified complex pattern and second complex pattern.
23. the method for claim 18, wherein this first and second complex pattern comprises hologram image.
24. the method for claim 18, wherein the aberration scope of this expection comprises minimum aberration value and maximum aberration value.
25. the method for claim 18, wherein said aberration value comprises focal length value.
26. a method that is used for detecting differences between complex images comprises:
Obtain first complex pattern and second complex pattern, described first and second complex patterns comprise similar feature;
In the aberration scope of expection is that first complex pattern is selected a plurality of aberration value;
For each selected aberration value is calculated an aberration function;
First complex pattern is carried out Fourier transform to obtain the complex pattern through conversion;
With the complex pattern of each aberration function iterative modifications through conversion;
Modified complex pattern is carried out inverse fourier transform, so that the low frequency ratio is transformed to time domain;
To compare with the high fdrequency component of second complex pattern through the high fdrequency component of the complex pattern of conversion; And
Determine an aberration correction value by selecting to make through the complex pattern of conversion and the aberration value of the second differences between complex images minimum.
27. the method for claim 26 is wherein determined the variance of modified complex pattern and the mould of the ratio of second complex pattern through the complex pattern of conversion and relatively comprising of second complex pattern.
28. the method for claim 27 determines that wherein aberration correction value comprises that selection makes through the complex pattern of conversion and the ratio of the variance minimum between second complex pattern.
29. the method for claim 26, wherein said first and second complex patterns comprise hologram image.
30. a system that is used for detecting differences between complex images comprises:
Can be used to obtain the numeroscope of first complex pattern and second complex pattern, described first and second complex patterns comprise similar feature; And
Be connected to the processing resource of described numeroscope, described processing resource can be used to:
In the aberration scope of expection is that first complex pattern is selected a plurality of aberration value;
For each aberration value is calculated an aberration function;
By each aberration function iterative modifications first complex pattern;
Modified complex pattern is compared with second complex pattern; And
By selecting to make the aberration value of the modified complex pattern and the second differences between complex images minimum determine aberration correction value.
31. the system of claim 30 also comprises the processing resource that can be used to first complex pattern is carried out Fourier transform, the feasible frequency domain that is modified in to first complex pattern carries out.
32. the system of claim 31 also comprises can be used to modified complex pattern and the processing resource of modified complex pattern being carried out before second complex pattern is compared inverse fourier transform.
33. the system of claim 30 determines relatively the comprising of wherein modified complex pattern and second complex pattern variance of modified complex pattern and the mould of the ratio of second complex pattern.
34. the system of claim 33 determines that wherein aberration correction value comprises the ratio of selecting to make the variance minimum between the modified complex pattern and second complex pattern.
35. the system of claim 30, wherein said first and second complex patterns comprise hologram image.
36. the system of claim 30, wherein said numeroscope comprises ccd video camera.
37. the system of claim 30 also comprises and the beam combiner of described numeroscope optically-coupled, can be used to receive reference beam and object beam to produce first and second complex patterns.
38. a method that is used for detecting differences between complex images comprises:
Obtain first complex pattern and second complex pattern, described first and second complex patterns comprise similar feature;
Determine between first and second complex pattern, whether there is aberration value difference;
By aberration function iterative modifications first complex pattern with at high-frequency range the modified first complex image and second complex pattern are compared the difference of aberration correction value;
Recently revise second complex pattern so that the low frequency component of second complex pattern low frequency component with first complex pattern is replaced by low frequency; And
The high fdrequency component of modified first complex image is compared with the high fdrequency component of modified second complex pattern, to determine whether first complex pattern mates with second complex pattern.
39. the method for claim 38 also comprises:
In the aberration scope of expection is that first complex pattern is selected a plurality of aberration value; And
For each aberration value is calculated aberration function.
40. the method for claim 38 comprises also first complex pattern is carried out Fourier transform that the feasible frequency domain that is modified in to first complex pattern carries out.
41. the method for claim 40 also is included in and modified first complex image is carried out inverse fourier transform before modified first complex image compared with second complex pattern.
42. the method for claim 38, wherein modified first complex image and second complex pattern relatively comprises the variance of determining modified first complex image and the mould of the ratio of second complex pattern.
43. the method for claim 42 also comprises by selecting to make recently definite aberration correction value of the variance minimum between the modified first complex image and second complex pattern.
44. the method for claim 38 also comprises application of low pass filters in the ratio of modified first complex image with second complex pattern, to obtain the low frequency ratio.
45. the method for claim 44 also comprises the Fourier transform that calculates modified first complex image and the ratio of second complex pattern, so that in the frequency domain application of low-pass filters.
46. the method for claim 45 also comprises the inverse fourier transform that calculates the low frequency ratio, so that revise second complex pattern in time domain.
47. the method for claim 38, wherein said first and second complex patterns comprise hologram image.
CNB038248700A 2002-09-12 2003-09-12 System and method for detecting differences between complex images Expired - Fee Related CN100342400C (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US41015402P 2002-09-12 2002-09-12
US60/410,156 2002-09-12
US60/410,154 2002-09-12

Publications (2)

Publication Number Publication Date
CN1695165A CN1695165A (en) 2005-11-09
CN100342400C true CN100342400C (en) 2007-10-10

Family

ID=35353504

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB038248700A Expired - Fee Related CN100342400C (en) 2002-09-12 2003-09-12 System and method for detecting differences between complex images

Country Status (1)

Country Link
CN (1) CN100342400C (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4750047B2 (en) * 2006-03-31 2011-08-17 株式会社リコー Misalignment detection apparatus, misalignment detection method, misalignment detection program, and recording medium
US20120281197A1 (en) * 2010-01-27 2012-11-08 Asml Netherlands B.V. Holographic Mask Inspection System with Spatial Filter
DE102010024684B8 (en) * 2010-06-23 2019-08-22 Siemens Healthcare Gmbh Method for reconstructing image data of a moving examination object, control and processing unit, computer tomography system and computer program

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE19646702A1 (en) * 1995-11-22 1997-05-28 Europ Gas Turbines Ltd Method for detecting production defects in an article

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE19646702A1 (en) * 1995-11-22 1997-05-28 Europ Gas Turbines Ltd Method for detecting production defects in an article

Also Published As

Publication number Publication date
CN1695165A (en) 2005-11-09

Similar Documents

Publication Publication Date Title
CN1161670C (en) Direct-to-digital holographic, holographic interferometry, and holovision
US11169367B2 (en) Three-dimensional microscopic imaging method and system
US20040057089A1 (en) System and method for detecting differences between complex images
WO2012099220A1 (en) Three-dimensional shape measurement method and three-dimensional shape measurement device
US8237933B2 (en) Method for image calibration and apparatus for image acquiring
KR101990009B1 (en) An Improved Holographic Reconstruction Apparatus and Method
JP5891566B2 (en) Digital holography device and image generation method using digital holography
CN1415083A (en) Improvements to acquistion and replay system for direct-to-digital holography and holovision
CN1688944A (en) Off-axis illumination direct-to-digital holography
US10684589B2 (en) Apparatus and method for in-line holographic imaging
JP5053193B2 (en) Particle size measuring apparatus and particle size measuring method
US7456976B2 (en) Statistical method of generating a synthetic hologram from measured data
WO2004066036A2 (en) Reduction of reference hologram noise and fourier space smearing in direct-to-digital holography
CN100342400C (en) System and method for detecting differences between complex images
CN1688945A (en) Content-based fused off-axis illumination direct-to-digital holography
JP2538435B2 (en) Fringe phase distribution analysis method and fringe phase distribution analyzer
JP3507345B2 (en) Moire measuring method and moire measuring apparatus using the same
JP2861723B2 (en) Confocal microscope
JP2005537517A (en) Direct digital holography acquisition of content-based off-axis illuminated objects
KR20170079441A (en) Device and method for recording and reconstructing digital hologram of high step sample with vibrant environment
JP5794152B2 (en) Image reconstruction method in digital holography
US20230259070A1 (en) Systems and Methods for Reconstruction of Digital Holograms
JP2003294412A (en) Holography apparatus
JP2005151387A (en) Method and apparatus for processing image formed by erecting equal size lens array
TWI494716B (en) Digital hologram imaging system and numerical reconstruction method for hologram

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
C19 Lapse of patent right due to non-payment of the annual fee
CF01 Termination of patent right due to non-payment of annual fee