CN103791853A - Microstructure measuring device and measuring method based on color strip information processing - Google Patents

Abstract

A microstructure measurement system and measurement method based on color fringe information processing. The system includes a Zeiss optical microscope, an input light source halogen lamp, a scanning mechanism connected to a computer at the sample scanning end of the Zeiss optical microscope, and a signal acquisition end of the Zeiss optical microscope. A CCD color camera and an image acquisition card connected with the CCD color camera are arranged in sequence, and the image acquisition card is also connected with a computer. Methods: Perform inverse Bayer transformation on a group of original images collected by a CCD color camera; obtain the light intensity values of R, G, and B at different scanning positions; calculate the phase information of the R, G, and B channels; determine the zero optical path difference position; obtain the height information corresponding to the pixel point; obtain the height information corresponding to different pixel points, and finally obtain the surface topography of the object. The invention uses a CCD color camera to collect white light interference fringe images, which can effectively reduce the influence of environmental noise and improve measurement accuracy.

Description

Microstructure measurement system and measurement method based on color fringe information processing

technical field

The invention relates to a microstructure measurement system. In particular, it relates to a microstructure measurement system and measurement method based on color fringe information processing using a CCD color camera to collect white light interference fringe images.

Background technique

In the field of micro-nano manufacturing, the surface morphology of micro-nano devices has a significant impact on the reliability and quality of the system. For example, the surface morphology of the two parallel plates of a MEMS capacitor will significantly affect the electromechanical properties of the capacitor, such as capacitance, voltage, and electric field distribution, which in turn will affect the performance and yield of MEMS capacitors. In addition, the surface morphology of micro-nano devices can reflect the process parameters in the processing process, which is an important basis for monitoring and diagnosing the processing process. Therefore, the precise measurement of microstructure surface topography is of great significance.

In the field of micro-nano testing, white light interferometry technology is an important non-contact measurement method. During the test process, this technology does not constitute damage to the surface of the measured object, and also has the advantages of large measurement range and high test accuracy. The algorithm for white light interference signal processing is mainly divided into two types: one is to obtain the shape of the measured object by analyzing the light intensity distribution of the white light interference signal, and the main algorithms include center of gravity method, polynomial fitting method, etc.; The phase information of the white light interference signal obtains the shape of the object, and the main algorithms include Fourier transform method, wavelet transform method, etc.

E.

等人利用3-CCD彩色相机获取白光干涉图像，但由于采集的干涉信号蓝色波段光谱能量较弱，所以只使用了R、G两个通道的信息分析得到被测物体的形貌；Suodong Ma等人基于加窗傅里叶变换的方法，通过对3-CCD彩色相机采集的R、G、B三个通道的图像进行分析，获得了被测物体的形貌；相较于通过棱镜分光实现彩色图像获取的3-CCD彩色相机，单CCD彩色相机通过拜尔（Bayer）滤波获取彩色图像的方法成本更为低廉，Zdeněk Buchta等人基于实验验证了使用单CCD彩色相机采集白光干涉图像，进而分析获得物体三维形貌的可行性。In the process of collecting white light interference fringes, CCD black-and-white cameras are usually used, and the use of CCD color cameras can expand the acquired image information from two-dimensional to three-dimensional, that is, the information of the three channels of R, G, and B.

e.

et al. used a 3-CCD color camera to obtain white light interference images, but because the collected interference signals have weak spectral energy in the blue band, they only used the information analysis of the R and G channels to obtain the shape of the measured object; Suodong Ma et al. based on the windowed Fourier transform method, by analyzing the images of the R, G, and B channels collected by the 3-CCD color camera, the morphology of the measured object was obtained; The 3-CCD color camera for color image acquisition, the method of acquiring color images with a single CCD color camera through Bayer (Bayer) filtering is cheaper, and Zdeněk Buchta et al. have verified based on experiments that a single CCD color camera is used to collect white light interference images, and then Analyze the feasibility of obtaining the three-dimensional shape of the object.

Contents of the invention

The technical problem to be solved by the present invention is to provide a microstructure measurement system and measurement method based on color fringe information processing that can effectively reduce the impact of environmental noise and improve measurement accuracy.

The technical solution adopted in the present invention is: a microstructure measurement system based on color fringe information processing, including a Zeiss optical microscope, the light source input end of the Zeiss optical microscope is provided with a halogen lamp, and the sample scanning end of the Zeiss optical microscope is provided with a The sample is scanned vertically, and the scanning mechanism connected to the computer, the signal acquisition end of the Zeiss optical microscope is sequentially provided with a CCD color camera and an image acquisition card connected to the CCD color camera, and the image acquisition card is also connected to the computer.

A white balance filter for reducing light transmittance in bands other than the blue band is arranged between the Zeiss optical microscope and the halogen lamp.

The scanning mechanism includes an objective nanopositioner corresponding to the sample scanning end of the Zeiss optical microscope and a piezoelectric controller connected to the objective nanopositioner, and the piezoelectric controller passes through The RS232 is connected to the computer, and the objective lens nanopositioner is connected to the Mirau type interference objective lens, and the Mirau type interference objective lens is set correspondingly to the samples arranged on the experimental table.

The measuring method for the microstructure measuring system based on color fringe information processing of the present invention comprises the following steps:

1) Perform inverse Bayer transformation on a group of original images collected by a CCD color camera to obtain real color image information;

2) Extract the R, G, and B channel light intensity information of any pixel of the color image transformed by step 1), and then obtain the light intensity values of R, G, and B at different scanning positions;

3) Select the Morlet wavelet as the mother wavelet of the wavelet transform, perform one-dimensional continuous wavelet transform on the light intensity values of the R, G, and B channels obtained in step 2), and use the transformation results to calculate the phase information of the R, G, and B channels;

初步确定零光程差的位置；4) For the phase information of the R, G, and B channels obtained in step 3), based on the constructed evaluation function

Preliminary determination of the position of zero optical path difference;

5) Select the phase information of the R, G, and B channels near zero optical path difference, and perform least squares fitting on the phase information of the R, G, and B channels to obtain the optimal estimated value of the zero optical path difference position, that is, accurately determine The position of zero optical path difference;

6) Using the optimal estimated value of the zero optical path difference position obtained in step 5), obtain the height information corresponding to the pixel in step 2);

7) Perform the same processing as steps 2) to 6) on the other pixels of the color image transformed in step 1), so as to obtain the height information corresponding to different pixels, and finally obtain the surface topography of the object.

Step 2) The light intensity values of R, G and B at any pixel point are obtained by the following formula:

${\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&gamma;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}\mathrm{<span\; class="notranslate">\; m</span>}}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; cos</span>}<span\; class="notranslate">\; [</span>\frac{\mathrm{<span\; class="notranslate">\; 4</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; )</span>}{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; cm</span>}}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

Among them, m represents R, G, B respectively, representing the R, G, B channels of the CCD color camera; I _0m represents the background light intensity; γ _m represents the visibility of stripes; λ _cm represents the central wavelength of the light source in the corresponding channel; The vertical scanning position of the nanopositioner of the objective lens, when the optical path difference between the two beams that interfere is zero, the position of the nanopositioner of the objective lens is expressed as z ₀ , referred to as the zero optical path difference position; g(z _m -z ₀ ) is the coherent envelope term.

The coherent envelope term is expressed as

$\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; [</span><span\; class="notranslate">\; -</span>{<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}}{{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; cm</span>}}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

Among them, l _cm represents the coherence length of the light source in the corresponding channel.

The R, G, B channel phase information described in step 3) is obtained through the following three steps:

In the first step, Morlet wavelet is selected as the mother wavelet of wavelet transform, namely

In the second step, one-dimensional continuous wavelet transform is performed on I _m (z) to obtain:

${\mathrm{<span\; class="notranslate">\; W</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; \&Integral;</span>}_{<span\; class="notranslate">\; -</span><span\; class="notranslate">\; \&infin;</span>}^{<span\; class="notranslate">\; +</span><span\; class="notranslate">\; \&infin;</span>}{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; \&psi;</span>}}_{\mathrm{<span\; class="notranslate">\; ab</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; dz</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span>$

Among them, ψ(z) is the mother wavelet of wavelet transform, a and b respectively represent the scale factor and translation factor of wavelet transform, ${\mathrm{\&psi;}}_{\mathrm{ab}}\left(z\right)=\frac{1}{\sqrt{a}}\mathrm{\&psi;}\left(\frac{z-b}{a}\right),$ * is the conjugate operator;

通过W_m(a,b)的辐角即The third step, R, G, B channel phase information

Through the argument of W _m (a,b) that is

Obtained, where a ₀ , b ₀ is the scale factor and translation factor corresponding to the maximum modulus value of W _m (a,b), Re[W _m (a ₀ ,b ₀ )] and Im[W _m (a ₀ ,b ₀ )] represent the real and imaginary parts of W _m (a ₀ ,b ₀ ), respectively.

The height information H corresponding to the pixel points in step 2) obtained by using the optimal estimated value of the zero optical path difference position in step 6) is obtained by the following formula:

H=H′+z ₀ (5)

Among them, H′=NΔ, Δ represents the scanning distance, N is the number of scanning steps corresponding to the position of the maximum sampling light intensity, and z ₀ is the optimal estimated value of the position of zero optical path difference.

The microstructure measurement system and measurement method based on color fringe information processing of the present invention uses a CCD color camera to collect white light interference fringe images, which can effectively reduce the impact of environmental noise compared with the traditional use of grayscale images to measure the shape of objects , improve measurement accuracy.

Description of drawings

Fig. 1 is the overall structure schematic diagram of microstructure measurement system of the present invention;

Figure 2a is a schematic diagram of the relative intensities of R, G, and B light intensities without using a white balance filter;

Figure 2b is a schematic diagram of the relative intensity of R, G, and B light intensities using a white balance filter;

Fig. 3 is the flowchart of the inventive method;

Figure 4a is an R channel light intensity distribution diagram of a certain pixel point;

Figure 4b is a light intensity distribution diagram of the G channel of a certain pixel point;

Figure 4c is a B-channel light intensity distribution diagram of a certain pixel point;

Fig. 5 is the R, G, B channel phase distribution diagram of a pixel near the zero optical path position;

Fig. 6 is a schematic diagram of the evaluation function.

in the picture

Detailed ways

The microstructure measurement system and measurement method based on the color fringe information processing of the present invention will be described in detail below in conjunction with the embodiments and the accompanying drawings.

As shown in Figure 1, the microstructure measurement system based on color fringe information processing includes a Zeiss optical microscope 2, the light source input end of the Zeiss optical microscope 2 is provided with a halogen lamp 1 with wide spectral characteristics, and the sample scanning of the Zeiss optical microscope 2 The end is provided with the scanning mechanism that is used for carrying out vertical scanning to sample 9, and is connected with computer 8, and the signal acquisition end of Zeiss optical microscope 2 is provided with the CCD color camera 6 that the model of Basler company is avA1600-65kc successively, and CCD color camera 6 The connected PCI-1428 image acquisition card 7 of NI Company realizes the acquisition of images, and the described image acquisition card 7 is also connected to a computer 8, and images can be observed in the computer. The computer includes the software part of the system, mainly including image acquisition using LabVIEW and data processing using MATLAB. A white balance filter 11 is arranged between the Zeiss optical microscope 2 and the halogen lamp 1 to reduce the light transmittance in bands other than the blue band.

As shown in Figure 2a and Figure 2b, adding a white balance filter at the illumination source can reduce the transmittance of light in other bands (except the blue band). When the white balance filter is not used in Figure 2a, the signal-to-noise ratio of the blue band is very low; after using the filter in Figure 2b, it reflects that by increasing the output light intensity of the light source, the light in the green and blue bands The intensity of the red band is enhanced, and the light intensity of the red band is attenuated, so that the signal-to-noise ratio of the green and blue band spectra is improved. utilization rate.

Described scanning mechanism comprises the objective lens nanopositioner 4 that is arranged correspondingly with Zeiss optical microscope 2 sample scanning ends and the piezoelectric controller 5 that links to each other with described objective lens nanopositioner 4, and described objective lens nanopositioner 4 selects The model that PI Company produces is the product of PIP-721.CL, and piezoelectric controller 5 selects the product that the model that PI Company produces is E-509.C1A, and described piezoelectric controller 5 is connected computer 8 by RS232, and described The objective lens nanopositioner 4 is connected on the Mirau type interference objective lens 3, and the Mirau type interference objective lens 3 adopts the Mirau type interference objective lens (magnification is 10X, and the numerical aperture N.A. is 0.30) produced by Nikon Corporation, and the described interference The objective lens 3 is set corresponding to the sample 9 set on the test bench 10 .

As shown in Figure 3, the microstructure measurement system and measurement method based on color fringe information processing of the present invention include the following steps:

1) Perform inverse Bayer transformation on a group of original images collected by a CCD color camera to obtain real color image information;

2) Extract the R (red), G (green) and B (blue) channel light intensity information of any pixel of the color image transformed by step 1), and then obtain the light intensity information of R, G, and B at different scanning positions. Strong value;

The light intensity values of R, G and B of any pixel point are obtained by the following formula:

${\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&gamma;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}\mathrm{<span\; class="notranslate">\; m</span>}}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; cos</span>}<span\; class="notranslate">\; [</span>\frac{\mathrm{<span\; class="notranslate">\; 4</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; )</span>}{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; cm</span>}}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

Among them, m are represented as R, G, and B respectively, representing the R, G, and B channels of the CCD color camera; I _0m represents the background light intensity, reflecting the DC component of light intensity; γ _m represents the visibility of stripes; λ _cm represents the light source in the corresponding The central wavelength of the channel; z represents the vertical scanning position of the objective lens nanopositioner, when the optical path difference between the two beams that interfere is zero, the position of the objective lens nanopositioner is expressed as z ₀ , and this position is referred to as the zero optical path difference position ; g(z _m -z ₀ ) is the coherent envelope item, which is determined by the spectral characteristics of the light source in the corresponding channel.

The light source halogen lamp used in the experiment has the characteristics of Gauss distribution in the spectra of R, G, and B channels. Therefore, the coherent envelope term is expressed as

$\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; [</span><span\; class="notranslate">\; -</span>{<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}}{{\mathrm{<span\; class="notranslate">\; l</span>}}_{\mathrm{<span\; class="notranslate">\; cm</span>}}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

Among them, l _cm represents the coherence length of the light source in the corresponding channel. In the R, G, and B channels, there will be interference fringes only near the zero optical path difference position z ₀ .

3) Select the Morlet wavelet as the mother wavelet of the wavelet transform, perform one-dimensional continuous wavelet transform on the light intensity values of the R, G, and B channels obtained in step 2), and use the transformation results to calculate the phase information of the R, G, and B channels;

The phase information of the R, G, and B channels is obtained through the following three steps:

由于信号I_m呈现出高斯分布的特征，而Morlet小波事实上是由高斯函数调制而成的，所以选择Morlet小波作为小波变换的母小波。In the first step, Morlet wavelet is selected as the mother wavelet of wavelet transform, namely

Since the signal _Im presents the characteristics of Gaussian distribution, and Morlet wavelet is actually modulated by Gaussian function, Morlet wavelet is chosen as the mother wavelet of wavelet transformation.

In the second step, one-dimensional continuous wavelet transform is performed on I _m (z) to obtain:

${\mathrm{<span\; class="notranslate">\; W</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; \&Integral;</span>}_{<span\; class="notranslate">\; -</span><span\; class="notranslate">\; \&infin;</span>}^{<span\; class="notranslate">\; +</span><span\; class="notranslate">\; \&infin;</span>}{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; \&psi;</span>}}_{\mathrm{<span\; class="notranslate">\; ab</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; dz</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span>$

Among them, ψ(z) is the mother wavelet of wavelet transform, a and b respectively represent the scale factor and translation factor of wavelet transform, ${\mathrm{\&psi;}}_{\mathrm{ab}}\left(z\right)=\frac{1}{\sqrt{a}}\mathrm{\&psi;}\left(\frac{z-b}{a}\right),$ * is the conjugate operator;

通过W_m(a,b)的辐角即The third step, R, G, B channel phase information

Through the argument of W _m (a,b) that is

Obtained, where a ₀ , b ₀ is the scale factor and translation factor corresponding to the maximum modulus value of W _m (a,b), Re[W _m (a ₀ ,b ₀ )] and Im[W _m (a ₀ ,b ₀ )] represent the real part and imaginary part of W _m (a ₀ ,b ₀ ), respectively, as shown in Figure 5. W _m (a,b) obtained by performing one-dimensional continuous wavelet transform on Im (z) reflects the degree of correlation _between ψ _ab and _Im . The change of the scale factor a reflects the frequency change of the signal _Im , and the change of the translation factor b reflects the change of the displacement of the signal _Im . Through the continuous change of the scale factor a and the translation factor b, the ψ at different frequencies and displacements can be obtained The degree of correlation between _ab and _Im . At a certain scanning position, when ψ _ab (z) has the highest correlation with I _m (z), that is, when the modulus of W _m (a, b) is the largest, the argument of W _m (a, b) is the Phase value of _Im (z) at the scan position.

初步确定零光程差的位置，此时评价函数在零光程差位置处有最小值；4) For the phase information of the R, G, and B channels obtained in step 3), based on the evaluation function constructed as shown in Figure 6

Preliminarily determine the position of zero optical path difference, at this time, the evaluation function has a minimum value at the position of zero optical path difference;

5) Select the phase information of the R, G, and B channels near zero optical path difference, and perform least squares fitting on the phase information of the R, G, and B channels to obtain the optimal estimated value of the zero optical path difference position, that is, accurately determine The position of zero optical path difference.

In the actual measurement process, measurement errors make the zero optical path difference positions of R, G, and B not at the same position. In order to obtain the optimal estimated value of the zero optical path difference position, the least square method is used to determine it.

其中，A_m=4π/λ_cm，B_m=-4πz₀/λ_cm。根据最小二乘法原理，零光程差位置z₀的最优估计值可在

有最小值时求得，此时需满足条件The phase information of R, G, and B channels can be expressed as

Among them, A _m =4π/λ _cm , B _m =-4πz ₀ /λ _cm . According to the principle of the least square method, the optimal estimated value of the zero optical path difference position z ₀ can be found in

It is obtained when there is a minimum value, and the condition needs to be met at this time

It can be obtained that the optimal estimated value of the zero optical path difference position z ₀ is

${\mathrm{<span\; class="notranslate">\; <figure-callout\; id="0"\; label="z"\; filenames="HDA0000459050710000012.png,HDA0000459050710000021.png"\; state="\{\{state\}\}">z</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; -</span>\frac{{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; R</span>}}{\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; R</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}{\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}{\mathrm{<span\; class="notranslate">\; B</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}}{{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="R"\; filenames="HDA0000459050710000042.png,HDA0000459050710000051.png"\; state="\{\{state\}\}">R</figure-callout></span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="G"\; filenames="HDA0000459050710000042.png,HDA0000459050710000051.png"\; state="\{\{state\}\}">G</figure-callout></span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; B</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 6</span>}<span\; class="notranslate">\; )</span>$

做分析。采样点

满足函数关系The method of determining A _m and B _m is as follows: firstly, the position of zero optical path difference is preliminarily determined through the evaluation function EF, and then the sampling point in the zero-order fringe

Do analysis. Sampling point

Satisfy the functional relationship

According to the principle of the least square method, A _m and B _m in formula (7) satisfy the following conditional relationship:

Available

The values of A _m and B _m can be obtained, and then the optimal estimated value of the zero optical path difference position z ₀ can be obtained. The optimal estimated value z ₀ is limited to the zero-order fringe, which is essentially the deviation between the position of the maximum value of the sampled light intensity and the position of zero optical path difference.

6) Using the optimal estimated value of the zero optical path difference position obtained in step 5), the height information corresponding to the pixel point in step 2) is obtained, and the height information H is obtained by the following formula:

H=H′+z ₀ (10)

Among them, H′=NΔ, Δ represents the scanning distance, N is the number of scanning steps corresponding to the position of the maximum sampling light intensity, and z ₀ is the optimal estimated value of the position of zero optical path difference.

7) Perform the same processing as steps 2) to 6) on the other pixels of the color image transformed in step 1), so as to obtain the height information corresponding to different pixels, and finally obtain the surface topography of the object.

The invention utilizes a single CCD color camera to collect images of white light interference color fringes. In addition, different from the method of windowed Fourier transform, since the theory of wavelet transform is becoming more and more mature and its application is becoming more and more extensive, the present invention uses the continuous wavelet transform method to analyze the image information of the three channels of R, G, and B, so as to obtain the measured The geometric size and surface shape of the object.

Claims (8)

1. the microstructure measuring system based on color fringe information processing, include Zeiss optical microscope (2), it is characterized in that, the light source input end of Zeiss optical microscope (2) is provided with Halogen lamp LED (1), the Sample Scan end of Zeiss optical microscope (2) is provided with for sample (9) is carried out to vertical scanning, and the scanning mechanism being connected with computing machine (8), the signals collecting end of Zeiss optical microscope (2) is disposed with CCD color camera (6) and the image pick-up card (7) being connected with CCD color camera (6), described image pick-up card (7) also connects computing machine (8).

2. the microstructure measuring system based on color fringe information processing according to claim 1, it is characterized in that, between described Zeiss optical microscope (2) and described Halogen lamp LED (1), be provided with the white balance optical filter (11) for reducing the light transmission of wave band beyond blue wave band.

3. the microstructure measuring system based on color fringe information processing according to claim 1, it is characterized in that, include and the object lens nanometer positioning device (4) and the piezo controller (5) being connected with described object lens nanometer positioning device (4) of the corresponding setting of Sample Scan end of described Zeiss optical microscope (2) at described scanning mechanism, described piezo controller (5) connects computing machine (8) by RS232, described object lens nanometer positioning device (4) is connected on Mirau type interference objective (3), described Mirau type interference objective (3) and the corresponding setting of sample (9) being arranged on experiment table (10).

4. for a measuring method for the microstructure measuring system based on color fringe information processing described in claim 1～3 any one, it is characterized in that, comprise the steps:

1) one group of original image of CCD color camera collection is carried out to Baeyer inverse transformation, obtain real color image information;

2) extract respectively R, the G and the B passage intensity signal that are converted the arbitrary pixel of coloured image obtaining by step 1), and then obtain the light intensity value of different scanning position R, G, B;

3) selecting the Morlet small echo female small echo as wavelet transformation, to step 2) R, the G obtaining, the light intensity value of B passage do One Dimension Continuous Wavelet Transform, utilizes transformation results to calculate the phase information of R, G, B passage;

4) R, G step 3) being obtained, the phase information of B passage, based on the evaluation function of structure the preliminary position of determining zero optical path difference;

5) choose the phase information of near R, G zero optical path difference, B passage, the phase information of described R, G, B passage is done to least square fitting and obtain zero optical path difference position optimal estimation value, accurately determine the position of zero optical path difference;

6) the zero optical path difference position optimal estimation value of utilizing step 5) to obtain, obtains step 2) elevation information that described pixel is corresponding;

7) other pixel that step 1) is converted to the coloured image obtaining is done and step 2 respectively) to the identical processing of step 6), put corresponding elevation information thereby obtain different pixels, finally obtain the surface topography of object.

5. measuring method according to claim 4, is characterized in that step 2) described R, the G of arbitrary pixel and the light intensity value of B be to be obtained by following formula:

$I_m\left(z\right)=I_{0m}+{\mathrm{\&gamma;}}_m{I}_{0m}g(z-z_0)\cos \left[\frac{4\mathrm{\&pi;}(z-z_0)}{{\mathrm{\&lambda;}}_{\mathrm{cm}}}\right]---\left(1\right)$

Wherein, m is expressed as R, G, B, represents R, G, the B passage of CCD color camera; I _0mrepresent background light intensity; γ _mrepresent fringe visibility; λ _cmrepresent the centre wavelength of light source in respective channel; Z represents the position of object lens nanometer positioning device vertical scanning, and in the time that the two light beam light path differences that interfere are zero, the positional representation at object lens nanometer positioning device place is z ₀, referred to as zero optical path difference position; G (z _m-z ₀) be relevant envelope item.

6. measuring method according to claim 5, is characterized in that, described relevant envelope item is expressed as

$g(z-z_0)=\exp \left[{-\left(\frac{z-z_0}{l_{\mathrm{cm}}}\right)}^2\right]---\left(2\right)$

Wherein, l _cmrepresent the coherent length of light source in respective channel.

7. measuring method according to claim 4, is characterized in that, R, G, B channel phases information described in step 3) are to obtain by following three steps:

The first step, selects the female small echo of Morlet small echo as wavelet transformation,

Second step, to I _m(z) carrying out One Dimension Continuous Wavelet Transform obtains:

$W_m(a,b)={\&Integral;}_{-\&infin;}^{+\&infin;}{I}_m\left(z\right){\mathrm{\&psi;}}_{\mathrm{ab}}^{*}\left(z\right)\mathrm{dz}---\left(3\right)$

Wherein, female small echo that ψ (z) is wavelet transformation, a, b represents respectively scale factor and the shift factor of wavelet transformation, ${\mathrm{\&psi;}}_{\mathrm{ab}}\left(z\right)=\frac{1}{\sqrt{a}}\mathrm{\&psi;}\left(\frac{z-b}{a}\right),$ * be conjugate operation symbol;

The 3rd step, R, G, B channel phases information

pass through W _mthe argument of (a, b)

Obtain, wherein, a ₀, b ₀for W _mcorresponding scale factor and shift factor when the mould value of (a, b) is maximum, Re[W _m(a ₀, b ₀)] and Im[W _m(a ₀, b ₀)] represent respectively W _m(a ₀, b ₀) real part and imaginary part.

8. measuring method according to claim 4, is characterized in that, utilizes zero optical path difference position optimal estimation value described in step 6), obtains step 2) elevation information H that described pixel is corresponding is to obtain by following formula:

H=H′+z ₀ (5)

Wherein, H '=N Δ, Δ represents sweep span, N is scanning step number corresponding to sampling light intensity maximum value position, z ₀for the optimal estimation value of zero optical path difference position.

Images