A differential parallel confocal-based 3D measurement error correction method and system

Abstract

The application provides a 3D measurement error correction method and system based on differential parallel confocal, and belongs to the field of optical microscopic imaging. The method comprises the following steps: acquiring confocal images of a sample to be measured under first and second waveband illumination, and wide field images of the sample to be measured under third and fourth waveband illumination, to obtain first waveband confocal images, second waveband confocal images, third waveband wide field images and fourth waveband wide field images; determining a first waveband correction image according to the first waveband confocal images and the third waveband wide field images; determining a second waveband correction image according to the second waveband confocal images and the fourth waveband wide field images; subtracting the first waveband correction image from the second waveband correction image to obtain a differential signal image; and converting the differential signal image into a height image according to an axial response curve of the differential signal, so as to measure the surface height of the sample to be measured. The method effectively eliminates the influence of uneven illumination and uneven sample reflectivity on the differential signal, and improves the accuracy of multi-waveband differential parallel confocal 3D measurement results.

Description

Technical Field

[0001] This invention relates to the field of optical microscopy imaging, and in particular to a method and system for correcting 3D measurement errors caused by uneven illumination and variations in sample surface reflectivity on differential signals based on differential parallel confocal imaging. Background Technology

[0002] Confocal 3D imaging methods based on the conjugate imaging principle employ point illumination and point detection to eliminate stray light and defocus signals outside the detection point, thereby improving the signal-to-noise ratio and axial tomography capability of confocal images. Traditional confocal 3D imaging methods require axial scanning to detect the axial light intensity response curve, and the axial height of the measured object is located by analyzing the peak value of the acquired axial light intensity response curve, achieving microscopic three-dimensional topography measurement. To overcome the drawback of low sensitivity near the peak value in traditional confocal 3D imaging axial response curve peak localization methods, differential confocal 3D imaging methods have emerged. There are numerous specific implementation techniques for differential confocal 3D imaging methods, but they all share the common feature of acquiring two different confocal images (e.g., before and after focus), subtracting them to obtain a differential signal, and locating the height based on a pre-calibrated linear curve of the differential signal changing with axial height. For example, in 2017, Liu Zhiqun et al. proposed a parallel object-square differential confocal measurement method, achieving axial nanometer-precision measurement while improving confocal measurement efficiency by several orders of magnitude. Building upon this, in 2020, Zhu Xingxing et al. proposed a method to correct for uneven illumination and uneven sample reflectance. This method, in addition to traditional differential confocal three-dimensional measurement, introduces four extra images, resulting in redundancy and inevitably increasing systematic error. Furthermore, this method approximates the reflectance of the upper surface of the sample with the reflectance of the lower surface, which introduces a certain degree of error.

[0003] Based on the above problems, there is an urgent need for a new error correction method to improve the accuracy of 3D measurement. Summary of the Invention

[0004] The purpose of this invention is to provide a 3D measurement error correction method and system based on differential parallel confocal microscopy, which can improve the accuracy of 3D measurement.

[0005] To achieve the above objectives, the present invention provides the following solution:

[0006] A 3D measurement error correction method based on differential parallel confocal focusing includes:

[0007] Confocal images of the test samples under the first and second band illumination, and wide-field images of the test samples under the third and fourth band illumination were acquired respectively, resulting in the first band confocal image, the second band confocal image, the third band wide-field image, and the fourth band wide-field image; wherein, the first band ≠ the second band, and the third and fourth bands are arbitrary bands;

[0008] Based on the first band confocal plot and the third band wide-field plot, determine the first band correction plot;

[0009] Based on the second band confocal plot and the fourth band wide-field plot, determine the second band correction plot;

[0010] Subtract the first band correction image from the second band correction image to obtain the differential signal image;

[0011] Based on a predetermined differential signal axial response curve, the differential signal map is converted into a height map to measure the surface height of the sample under test; the differential signal axial response curve is the relationship curve between the differential signal and the sample surface height.

[0012] Optionally, a planar array color confocal 3D measurement device is used to acquire confocal images of the sample under the first and second band illumination, and wide-field images of the sample under the third and fourth band illumination, respectively. The planar array color confocal 3D measurement device includes a light source, a digital micromirror array, and a camera. The digital micromirror array includes multiple micromirror units. The light source is used to emit illumination light of the first, second, third, and fourth bands onto the sample under test. The digital micromirror array is used to form a two-dimensional planar array multifocal spot or wide-field illumination light on the sample under test. The camera is used to acquire the first band confocal image, the second band confocal image, the third band wide-field image, and the fourth band wide-field image.

[0013] Optionally, the method of using a planar array color confocal three-dimensional measurement device to acquire confocal images of the sample under the first and second band illumination, and wide-field images of the sample under the third and fourth band illumination, specifically includes:

[0014] Illumination light of the first wavelength is emitted onto the sample under test by a light source;

[0015] The digital micromirror array is controlled to form a two-dimensional illumination pinhole pattern with T micromirror units as one period, so as to form (C / T)*(R / T) two-dimensional area array multifocal light spots with different focusing states on the sample under test. The reflected light of the sample under test is collected by a camera to obtain a first-band confocal image; where C is the number of columns of micromirror units in the digital micromirror array, and R is the number of rows of micromirror units in the digital micromirror array.

[0016] The second-wavelength illumination light is emitted onto the sample under test by a light source;

[0017] The digital micromirror array is controlled to form a two-dimensional illumination pinhole pattern with T micromirror units as one cycle, so as to form (C / T)*(R / T) two-dimensional area array multifocal light spots with different focusing states on the sample under test, and the reflected light of the sample under test is collected by the camera to obtain the second band confocal image.

[0018] The third-band illumination light is emitted from the light source onto the sample under test;

[0019] The micromirror units of the digital micromirror array are controlled to form a wide-field illumination light on the sample under test, and the reflected light of the sample under test is collected by a camera to obtain a third-band wide-field image.

[0020] The fourth-band illumination light is emitted from the light source onto the sample under test;

[0021] The micromirror units of the digital micromirror array are controlled to form a wide-field illumination light on the sample under test, and the reflected light from the sample under test is collected by a camera to obtain a fourth-band wide-field image.

[0022] Optionally, the camera includes a first camera and a second camera; the area array color confocal three-dimensional measurement device further includes a first filter, a second filter, and a dichroic mirror; the first filter is disposed in the incident light path of the first camera, and the second filter is disposed in the incident light path of the second camera; the dichroic mirror is disposed in the incident light paths of the first filter and the second filter;

[0023] The method employs a planar array color confocal three-dimensional measurement device to acquire confocal images of the test sample under first and second band illumination, and wide-field images of the test sample under third and fourth band illumination, specifically including:

[0024] A composite light is emitted onto the sample under test by a light source; the composite light includes illumination light of a first band, illumination light of a second band, illumination light of a third band, and illumination light of a fourth band.

[0025] The digital micromirror array is controlled to form a two-dimensional illumination pinhole pattern with T micromirror units as one cycle, so as to form (C / T)*(R / T) two-dimensional area array multifocal light spots with different focusing states on the sample under test; where C is the number of columns of micromirror units in the digital micromirror array, and R is the number of rows of micromirror units in the digital micromirror array.

[0026] The dichroic mirror splits the reflected light from the sample into two paths: the first path of reflected light is incident on the first filter, and the second path of reflected light is incident on the second filter.

[0027] Adjust the filtering range of the first filter so that the first filter can only pass light in the first band. The first camera collects the first reflected light after passing through the first filter to obtain a confocal image of the first band.

[0028] Adjust the filtering range of the second filter so that it can only pass light in the second band. Then, use the second camera to collect the second reflected light after passing through the second filter to obtain a confocal image of the second band.

[0029] The micromirror array, which controls the digital micromirror array, forms a wide-field illumination light on the sample under test;

[0030] The dichroic mirror splits the reflected light from the sample into two paths: the first path of reflected light is incident on the first filter, and the second path of reflected light is incident on the second filter.

[0031] Adjust the filtering range of the first filter so that the first filter can only pass light in the third band. The first camera collects the first reflected light after passing through the first filter to obtain a wide field image of the third band.

[0032] Adjust the filtering range of the second filter so that it can only pass light in the fourth band. Then, use the second camera to collect the second reflected light after passing through the second filter to obtain a wide-field image of the fourth band.

[0033] Optionally, a first-band correction map is obtained based on the first-band confocal map and the third-band wide-field map, specifically including:

[0034] Divide the first band confocal image by the third band wide-field image to obtain the first band correction image.

[0035] Optionally, the first band correction map is determined based on the first band confocal map and the third band wide-field map, specifically including:

[0036] Divide the first band confocal image by the third band wide-field image to obtain the initial first band correction image;

[0037] The initial first band correction image is normalized to obtain the first band correction image.

[0038] Optionally, the initial first band correction map can be normalized using the following formula:

[0039]

[0040] in, This is the correction image for the first band. This is the initial correction image for the first band.

[0041] Optionally, converting the differential signal map into a height map based on a predetermined differential signal axial response curve specifically includes:

[0042] The differential signal graph is filtered to obtain a corrected differential signal graph;

[0043] Based on the predetermined differential signal axial response curve, the corrected differential signal map is converted into a height map.

[0044] Optionally, the method for determining the differential signal axial response curve includes:

[0045] Confocal images of samples at different surface heights under first-band illumination and second-band illumination were collected respectively, and the first-band confocal image and the second-band confocal image corresponding to each sample were obtained.

[0046] For any sample, the differential signal of the sample is obtained by subtracting the gray value of the corresponding pixel in the second band confocal image from the gray value of each pixel in the first band confocal image of the sample.

[0047] The surface height of each sample and the differential signal of each sample are fitted to obtain the relationship curve between the differential signal and the surface height of the sample.

[0048] To achieve the above objectives, the present invention also provides the following solution:

[0049] A 3D measurement error correction system based on differential parallel confocal focusing includes:

[0050] The image acquisition unit is used to acquire confocal images of the test sample under the first band illumination and the second band illumination, and wide-field images of the test sample under the third band illumination and the fourth band illumination, respectively, to obtain the first band confocal image, the second band confocal image, the third band wide-field image, and the fourth band wide-field image; wherein, the first band ≠ the second band, and the third and fourth bands are arbitrary bands;

[0051] The first correction unit is connected to the image acquisition unit and is used to obtain the first band correction map based on the first band confocal map and the third band wide field map.

[0052] The second correction unit is connected to the image acquisition unit and is used to obtain the second band correction map based on the second band confocal map and the fourth band wide field map.

[0053] A differential signal determination unit is connected to the first correction unit and the second correction unit respectively, and is used to subtract the first band correction diagram from the second band correction diagram to obtain a differential signal diagram;

[0054] A height measurement unit, connected to the differential signal determination unit, is used to convert the differential signal map into a height map according to a pre-determined differential signal axial response curve, so as to measure the surface height of the sample to be tested; the differential signal axial response curve is the relationship curve between the differential signal and the sample surface height.

[0055] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects: First, confocal images of the sample under test under first-band illumination and second-band illumination, and wide-field images of the sample under test under third-band illumination and fourth-band illumination are acquired respectively, to obtain the first-band confocal image, the second-band confocal image, the third-band wide-field image, and the fourth-band wide-field image. Then, a first-band correction image is determined based on the first-band confocal image and the third-band wide-field image, and a second-band correction image is determined based on the second-band confocal image and the fourth-band wide-field image. The first-band correction image is then subtracted from the second-band correction image to obtain the differential signal image. Finally, based on a predetermined differential signal axial response curve, the differential signal image is converted into a height image, so that the amplitude change of the corrected differential signal (differential signal image) mainly reflects the change in the surface height of the sample under test, effectively eliminating the influence of uneven illumination and uneven sample reflectivity on the measurement accuracy of the differential signal, and improving the accuracy of multi-band differential parallel confocal 3D measurement results. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 The flowchart shows the 3D measurement error correction method based on differential parallel confocal focusing according to the present invention.

[0058] Figure 2 This is a schematic diagram of the first type of area array color confocal three-dimensional measurement device;

[0059] Figure 3 This is a second schematic diagram of a planar array color confocal 3D measurement device;

[0060] Figure 4 This is a schematic diagram of the module of the 3D measurement error correction system based on differential parallel confocal focusing of the present invention.

[0061] Symbol explanation:

[0062] Polychromatic illumination source-101, total internal reflection prism-102, collimating lens group-111, linear polarizer 121, polarizing beam splitter-122, quarter wave plate-123, dispersive tube diameter-124, microscope objective-125, stage-126, first tube lens-131, first camera-132, first filter-133, second tube lens-134, second camera-135, second filter-136, dichroic mirror-137, sample to be tested-141;

[0063] Image acquisition unit-21, first correction unit-22, second correction unit-23, differential signal determination unit-24, height measurement unit-25. Detailed Implementation

[0064] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0065] The purpose of this invention is to provide a 3D measurement error correction method and system based on differential parallel confocal imaging. This method corrects the acquired confocal images of different bands before constructing the differential signal: by dividing the confocal images of different bands by the wide-field image of that band, a corrected image for the corresponding band is obtained. This eliminates the variation in differential signal amplitude caused by uneven illumination or different reflectivities of the sample surface material, ensuring that the amplitude change of the corrected differential signal mainly reflects the change in sample surface height, thereby improving the accuracy of multi-band differential parallel confocal 3D measurement results.

[0066] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0067] Example 1

[0068] like Figure 1 As shown, the 3D measurement error correction method based on differential parallel confocal microscopy provided in this embodiment includes:

[0069] S1: Collect the first band λ respectively i Under illumination and in the second band λ m Confocal plot of the sample under illumination, third band λ j Under illumination and in the fourth band λ n The wide-field image of the sample under illumination is used to obtain the first-band confocal image. Second-band confocal image Third band wide field image and the fourth band wide field diagram Where (i,j,m,n)∈[1,2,3,…,N],i≠m,N≥2,N is the total number of bands.

[0070] In this embodiment, a color confocal 3D measurement device is used to acquire confocal images of the sample under the first and second band illumination, and wide-field images of the sample under the third and fourth band illumination.

[0071] The area array color confocal three-dimensional measurement device includes a light source, a DMD (Digital Micromirror Device), and a camera. The DMD includes multiple micromirror units. The light source emits illumination light in four bands: a first band, a second band, a third band, and a fourth band, onto the sample. The DMD forms a two-dimensional area array multifocal spot or wide-field illumination light on the sample. The camera acquires the first band confocal image, the second band confocal image, the third band wide-field image, and the fourth band wide-field image. For the specific structure of the area array color confocal three-dimensional measurement device, please refer to Chinese Patent Publication No. CN115200508A, published on October 18, 2022, which describes an area array ultra-precision color confocal microscope three-dimensional measurement device.

[0072] like Figure 2 and Figure 3 As shown, the area array color confocal 3D measurement device includes an area array multifocal software and control module, a polychromatic illumination source module, a collimating lens group 111, a DMD, a longitudinal dispersion optical microscopy imaging module, and a single-camera imaging module (or a dual-camera imaging module). The area array multifocal polychromatic illumination source module includes, in sequence, a polychromatic illumination source 101 and a total internal reflection prism 102 in the optical path. The longitudinal dispersion optical microscopy imaging module includes, in sequence, a linear polarizer 121, a polarizing beam splitter prism 122, a quarter-wave plate 123, a dispersive tube 124, a microscope objective 125, and a stage 126 in the optical path. The sample 141 to be measured is placed on the stage 126. The single-camera imaging module (or dual-camera imaging module) includes a tube and a camera.

[0073] As a specific implementation method, the first-band confocal image under first-band illumination. After acquisition, the illumination band was changed to the second band, and the second band confocal image was obtained under the second band illumination. After acquisition, switch the illumination wavelength to the third band. (Third band wide-field plot) After acquisition, switch the illumination band to the fourth band and acquire a wide-field image of the fourth band. In this implementation, such as Figure 2As shown, the camera is a single camera without filters; the switching of wavelengths is controlled by the light source. Specifically, step S1 includes:

[0074] S101: Emitting the first band λ to the sample under test via a light source. i The illumination light.

[0075] S102: Control the digital micromirror array to form a two-dimensional illumination pinhole pattern with T micromirror units as one cycle, so as to form (C / T)*(R / T) two-dimensional area array multifocal light spots with different focusing states on the sample under test, and collect the reflected light of the sample under test through the camera to obtain the first band confocal image.

[0076] S103: Switch the light source to the second band and emit second band illumination light onto the sample under test through the light source.

[0077] S104: Control the digital micromirror array to form a two-dimensional illumination pinhole pattern with T micromirror units as one cycle, so as to form (C / T)*(R / T) two-dimensional area array multifocal light spots with different focusing states on the sample under test, and collect the reflected light of the sample under test through the camera to obtain the second band confocal image.

[0078] Where C is the number of columns of micromirror units in the digital micromirror array, and R is the number of rows of micromirror units in the digital micromirror array. Specifically, after the light spot is reflected by the sample under test, it is captured and imaged by black and white cameras 1, 2, ..., N that take pictures of different bands. The pixel grayscale values ​​of the corresponding DMD (C / T)*(R / T) "on" state are retained, while the grayscale values ​​of the images corresponding to the other "off" states of the DMD are set to zero, forming a band of λ. k Multispectral confocal images (k = 1, 2, ... N).

[0079] The two-dimensional illumination pinhole pattern formed by the DMD does not have a strict positional relationship with the position of the sample under test. The sample under test can be placed at any position within the camera's field of view (field of view size 2048*2048 pixels). The imaging pattern of each pixel (x,y) in the entire field of view is determined by the two-dimensional illumination pinhole pattern formed by the DMD.

[0080] S104: Switch the wavelength of the light source to the third wavelength and emit third-wavelength illumination light onto the sample under test through the light source.

[0081] S105: The micromirror units of the digital micromirror array are controlled to form a wide-field illumination light on the sample under test, and the reflected light from the sample is acquired by a camera to obtain a third-band wide-field image. Specifically, the digital micromirror array is controlled so that all C*R micromirror units are fully open to form a wide-field illumination light. The reflected light from the sample under test is captured and imaged by a monochrome camera that takes pictures of different bands, forming a band with wavelength λ. jWide-field image of (k = 1, 2, ..., N)

[0082] S106: Switch the light source to the fourth band and emit fourth band illumination light onto the sample under test through the light source.

[0083] S107: The micromirror units of the digital micromirror array are controlled to form a wide-field illumination light on the sample under test, and the reflected light of the sample under test is collected by the camera to obtain a fourth-band wide-field image.

[0084] As another specific implementation method, Implementation Method Two: First Band Confocal Map Confocal image with second band Simultaneously acquire the third-band wide-field image. Compared with the fourth band wide field map Simultaneous acquisition. In this implementation, such as Figure 3 As shown, the camera is a dual-camera system, including a first camera 132 and a second camera 135. The area array color confocal 3D measurement device also includes a first filter 133, a second filter 136, and a dichroic mirror 137. The first filter 133 is disposed in the incident light path of the first camera 132, the second filter 136 is disposed in the incident light path of the second camera 135, and the dichroic mirror 137 is disposed in the incident light paths of the first filter 133 and the second filter 136. Further, the area array color confocal 3D measurement device also includes a first tube mirror 131 and a second tube mirror 134. The first tube mirror 131 is disposed between the first filter 133 and the first camera 132, and the second tube mirror 134 is disposed between the second filter 136 and the second camera 135. Step S2 specifically includes:

[0085] S111: Emit composite light to the sample under test through a light source. The composite light includes illumination light of the first band, illumination light of the second band, illumination light of the third band, and illumination light of the fourth band.

[0086] S112: Control the digital micromirror array to form a two-dimensional illumination pinhole pattern with T micromirror units as one cycle, so as to form (C / T)*(R / T) two-dimensional area array multifocal light spots with different focusing states on the sample under test. Wherein, C is the number of columns of micromirror units in the digital micromirror array, and R is the number of rows of micromirror units in the digital micromirror array.

[0087] S113: The reflected light from the sample to be tested is split into two paths by the dichroic mirror. The first path of reflected light is incident on the first filter, and the second path of reflected light is incident on the second filter.

[0088] S114: Adjust the filtering range of the first filter so that the first filter can only pass light in the first band. The first camera collects the first reflected light after passing through the first filter to obtain a first band confocal image.

[0089] There are three scenarios: (1) When both the third and fourth bands are the same as the first band, the second band light in the first reflected light is filtered out by the first filter. (2) When the third band is the same as the first band and the fourth band is different from the first band, the second and fourth bands light in the first reflected light are filtered out by the first filter. (3) When the fourth band is the same as the first band and the third band is different from the first band, the second and third bands light are filtered out by the first filter.

[0090] S115: Adjust the filtering range of the second filter so that the second filter can only pass light in the second band. The second camera collects the second reflected light after passing through the second filter to obtain a second band confocal image.

[0091] There are three scenarios: (1) When both the third and fourth bands are the same as the second band, the light of the first band in the first reflected light is filtered out by the second filter. (2) When the third band is the same as the second band and the fourth band is different from the second band, the light of the first and fourth bands in the first reflected light is filtered out by the second filter. (3) When the fourth band is the same as the second band and the third band is different from the second band, the light of the first and third bands is filtered out by the second filter.

[0092] S116: The micromirror array of the digital micromirror array is controlled to form a wide field illumination light on the sample under test.

[0093] S117: The reflected light from the sample to be tested is split into two paths by the dichroic mirror. The first path of reflected light is incident on the first filter, and the second path of reflected light is incident on the second filter.

[0094] S118: Adjust the filtering range of the first filter so that the first filter can only pass light in the third band. The first camera collects the first reflected light after passing through the first filter to obtain a wide-field image of the third band.

[0095] There are five possible scenarios: (1) When the first, third, and fourth bands are all the same, the second band of light in the first reflected light is filtered out by the first filter. (2) When the third band is the same as the first band and the third band is different from the fourth band, the second and fourth bands of light in the first reflected light are filtered out by the first filter. (3) When the second, third, and fourth bands are all the same, the first band of light in the first reflected light is filtered out by the first filter. (4) When the third band is the same as the second band and the third band is different from the fourth band, the first and fourth bands of light in the first reflected light are filtered out by the first filter. (5) When the third and fourth bands are the same and the third band is different from both the first and second bands, the first and second bands of light in the first reflected light are filtered out by the first filter.

[0096] S119: Adjust the filtering range of the second filter so that the second filter can only pass light in the fourth band. The second camera collects the second reflected light after passing through the second filter to obtain a wide field image of the fourth band.

[0097] There are five possible scenarios: (1) When the first, third, and fourth bands are all the same, the second band light in the second reflected light is filtered out by the second filter. (2) When the fourth band is the same as the first band and different from the third band, the second band and the third band light in the second reflected light are filtered out by the second filter. (3) When the second, third, and fourth bands are all the same, the first band light in the second reflected light is filtered out by the second filter. (4) When the fourth band is the same as the second band and different from the third band, the first and third band light in the second reflected light are filtered out by the second filter. (5) When the fourth band is the same as the third band and different from both the first and second bands, the first and second band light in the second reflected light are filtered out by the second filter.

[0098] S2: Determine the first band correction map based on the first band confocal map and the third band wide-field map. Specifically, the first band confocal map... Divide by the aforementioned third-band wide-field diagram Obtain the initial first band correction map Correction plot of the initial first band After normalization, the first band correction image is obtained.

[0099] In this embodiment, the initial first band correction image is normalized using the following formula:

[0100]

[0101] in, This is the correction image for the first band. This is the initial correction image for the first band.

[0102] S3: Determine the second band correction map based on the second band confocal map and the fourth band wide-field map. In this embodiment, step S3 is processed in the same way as step S2, and the second band confocal map is... Divide by the aforementioned fourth band wide field diagram Obtain the initial second-band correction map Correction plot of the initial second band After normalization, the second band correction image is obtained.

[0103] S4: Correct the first band image Corrected image with the second band Subtracting them yields the differential signal diagram.

[0104] S5: Based on the pre-determined differential signal axial response curve The differential signal map is converted into a height map Z(x,y) to measure the surface height of the sample. The differential signal axial response curve is the relationship curve between the differential signal and the sample surface height. Here, Z represents the sample surface height.

[0105] Specifically, the first and second band corrected images contain a large amount of noise, and this noise is salt-and-pepper noise, which affects altitude measurement. Therefore, the differential signal image is first... Filtering is performed to obtain the corrected differential signal diagram. Then, based on the predetermined differential signal axial response curve, the corrected differential signal diagram is... Convert to a height map. Specifically, the filtering method can be median filtering of an m×n template (m≥3, n≥3).

[0106] In this embodiment, the differential signal axial response curve The methods for determining this include:

[0107] Data were collected on different surface heights (h1, h2, ..., h) under the first and second band illumination. n The confocal plots of the samples are obtained, and the first-band confocal plot, the second-band confocal plot, and the second-band wide-field plot are obtained for each sample.

[0108] For any sample, the differential signal of the sample is obtained by subtracting the gray value of the corresponding pixel in the second band confocal image from the gray value of each pixel (x,y) in the first band confocal image of the sample.

[0109] The surface height of each sample and the differential signal of each sample are fitted to obtain the relationship curve between the differential signal and the surface height of the sample.

[0110] Specifically, the first band confocal image of the sample with surface height h1 is obtained. Second band confocal image Using the first band confocal plot Subtract the second band confocal image from the gray value at (x,y). The differential signal of the corresponding sample is obtained by measuring the gray values ​​at the same location. Constructing the relationship between differential signal and sample surface height By replacing samples with different surface heights, the relationship between the differential signal and the surface height was constructed using the same method. Where h p Approaching the upper limit of the system's operating range, the number of samples p ≥ 2; fitting differential signal strength The curve showing the relationship between the height of the sample surface and the axial response curve of the differential signal is the curve of the differential signal.

[0111] This invention provides a correction method for height measurement errors caused by uneven illumination and uneven surface reflection of samples in differential parallel confocal 3D measurement. A planar array color confocal 3D measurement device is used to correct the acquired confocal images of different bands before constructing the differential signal. The correction method involves dividing the confocal images of different bands by the wide-field image of that band. This correction method effectively eliminates the influence of uneven illumination and uneven sample reflectivity on the differential signal constructed in differential parallel confocal 3D measurement, and eliminates the variation in differential signal amplitude caused by uneven illumination or different reflectivity of the sample surface material. This ensures that the amplitude change of the corrected differential signal mainly reflects the change in the object's surface height, thus achieving accurate results even if the measurement device has uneven illumination or the surface reflectivity of the sample is uneven.

[0112] The error correction method of this invention requires less data than the method proposed by Zhu Xingxing et al. It only requires two images for correction, instead of four, making it easier to implement and more efficient.

[0113] Example 2

[0114] In order to implement the method corresponding to Embodiment 1 above and achieve the corresponding functions and technical effects, a 3D measurement error correction system based on differential parallel confocal focusing is provided below.

[0115] like Figure 4As shown, the 3D measurement error correction system based on differential parallel confocal focusing provided in this embodiment includes: an image acquisition unit 21, a first correction unit 22, a second correction unit 23, a differential signal determination unit 24, and a height measurement unit 25.

[0116] The image acquisition unit 21 is used to acquire confocal images of the sample under the first band illumination and the second band illumination, and wide-field images of the sample under the third band illumination and the fourth band illumination, respectively, to obtain the first band confocal image, the second band confocal image, the third band wide-field image and the fourth band wide-field image.

[0117] The first correction unit 22 is connected to the image acquisition unit 21. The first correction unit 22 is used to obtain the first band correction map based on the first band confocal map and the third band wide field map.

[0118] The second correction unit 23 is connected to the image acquisition unit 21. The second correction unit 23 is used to obtain the second band correction map based on the second band confocal map and the fourth band wide field map.

[0119] The differential signal determination unit 24 is connected to the first correction unit 22 and the second correction unit 23 respectively. The differential signal determination unit 24 is used to subtract the first band correction map from the second band correction map to obtain the differential signal map.

[0120] The height measurement unit 25 is connected to the differential signal determination unit 24. The height measurement unit 25 is used to convert the differential signal diagram into a height diagram according to the predetermined differential signal axial response curve, so as to measure the surface height of the sample to be tested. The differential signal axial response curve is the relationship curve between the differential signal and the sample surface height.

[0121] Compared to existing technologies, the 3D measurement error correction system based on differential parallel confocal focusing provided in this embodiment has the same beneficial effects as the 3D measurement error correction method based on differential parallel confocal focusing provided in Embodiment 1, and will not be repeated here.

[0122] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0123] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A 3D measurement error correction method based on differential parallel confocal focusing, characterized in that, The 3D measurement error correction method based on differential parallel confocal microscopy includes: Confocal images of the test samples under the first and second band illumination, and wide-field images of the test samples under the third and fourth band illumination were acquired respectively, resulting in the first band confocal image, the second band confocal image, the third band wide-field image, and the fourth band wide-field image; wherein, the first band ≠ the second band, and the third and fourth bands are arbitrary bands; Divide the first band confocal image by the third band wide-field image to obtain the initial first band correction image; The initial first band correction image is normalized to obtain the first band correction image; Divide the second band confocal plot by the fourth band wide-field plot to obtain the initial second band correction plot; The initial second-band correction image is normalized to obtain the second-band correction image; Subtract the first band correction image from the second band correction image to obtain the differential signal image; The differential signal image is filtered to obtain a corrected differential signal image; based on a predetermined differential signal axial response curve, the corrected differential signal image is converted into a height image to measure the surface height of the sample under test; the differential signal axial response curve is the relationship curve between the differential signal and the sample surface height; The method for determining the axial response curve of the differential signal includes: Confocal images of samples at different surface heights under first-band illumination and second-band illumination were collected respectively, and the first-band confocal image and the second-band confocal image corresponding to each sample were obtained. For any sample, the differential signal of the sample is obtained by subtracting the gray value of the corresponding pixel in the second band confocal image from the gray value of each pixel in the first band confocal image of the sample. The surface height of each sample and the differential signal of each sample are fitted to obtain the relationship curve between the differential signal and the surface height of the sample.

2. The 3D measurement error correction method based on differential parallel confocal focusing according to claim 1, characterized in that, A confocal three-dimensional measurement device with a color array was used to acquire confocal images of the sample under the first and second band illumination, and wide-field images of the sample under the third and fourth band illumination. The area array color confocal three-dimensional measurement device includes a light source, a digital micromirror array, and a camera; the digital micromirror array includes multiple micromirror units; the light source is used to emit illumination light of the first band, the second band, the third band, and the fourth band onto the sample to be tested; the digital micromirror array is used to form a two-dimensional area array multifocal spot or wide-field illumination light on the sample to be tested; the camera is used to acquire the first band confocal image, the second band confocal image, the third band wide-field image, and the fourth band wide-field image.

3. The 3D measurement error correction method based on differential parallel confocal microscopy according to claim 2, characterized in that, The method employs a planar array color confocal three-dimensional measurement device to acquire confocal images of the test sample under first and second band illumination, and wide-field images of the test sample under third and fourth band illumination, specifically including: Illumination light of the first wavelength is emitted onto the sample under test by a light source; The digital micromirror array is controlled to form a two-dimensional illumination pinhole pattern with T micromirror units per cycle, so as to form a (C / T) pattern on the sample under test. (R / T) two-dimensional array multifocal light spots with different focusing states are used to collect the reflected light of the sample under test through a camera to obtain the first band confocal image; where C is the number of columns of micromirror units in the digital micromirror array and R is the number of rows of micromirror units in the digital micromirror array. The second-wavelength illumination light is emitted onto the sample under test by a light source; The digital micromirror array is controlled to form a two-dimensional illumination pinhole pattern with T micromirror units per cycle, so as to form a (C / T) pattern on the sample under test. (R / T) two-dimensional array multifocal light spots with different focusing states are used to collect the reflected light of the sample under test through a camera to obtain the second band confocal image; The third-band illumination light is emitted from the light source onto the sample under test; The micromirror units of the digital micromirror array are controlled to form a wide-field illumination light on the sample under test, and the reflected light of the sample under test is collected by a camera to obtain a third-band wide-field image. The fourth-band illumination light is emitted from the light source onto the sample under test; The micromirror units of the digital micromirror array are controlled to form a wide-field illumination light on the sample under test, and the reflected light from the sample under test is collected by a camera to obtain a fourth-band wide-field image.

4. The 3D measurement error correction method based on differential parallel confocal focusing according to claim 2, characterized in that, The camera includes a first camera and a second camera; the area array color confocal three-dimensional measurement device further includes a first filter, a second filter and a dichroic mirror; the first filter is disposed in the incident light path of the first camera, the second filter is disposed in the incident light path of the second camera; the dichroic mirror is disposed in the incident light path of the first filter and the second filter; The method employs a planar array color confocal three-dimensional measurement device to acquire confocal images of the test sample under first and second band illumination, and wide-field images of the test sample under third and fourth band illumination, specifically including: A composite light is emitted onto the sample under test by a light source; the composite light includes illumination light of a first band, illumination light of a second band, illumination light of a third band, and illumination light of a fourth band. The digital micromirror array is controlled to form a two-dimensional illumination pinhole pattern with T micromirror units per cycle, so as to form a (C / T) pattern on the sample under test. (R / T) two-dimensional array multifocal light spots with different focusing states; where C is the number of columns of micromirror units in the digital micromirror array, and R is the number of rows of micromirror units in the digital micromirror array. The dichroic mirror splits the reflected light from the sample into two paths: the first path of reflected light is incident on the first filter, and the second path of reflected light is incident on the second filter. Adjust the filtering range of the first filter so that the first filter can only pass light in the first band. The first camera collects the first reflected light after passing through the first filter to obtain a confocal image of the first band. Adjust the filtering range of the second filter so that it can only pass light in the second band. Then, use the second camera to collect the second reflected light after passing through the second filter to obtain a confocal image of the second band. The micromirror array, which controls the digital micromirror array, forms a wide-field illumination light on the sample under test; The dichroic mirror splits the reflected light from the sample into two paths: the first path of reflected light is incident on the first filter, and the second path of reflected light is incident on the second filter. Adjust the filtering range of the first filter so that the first filter can only pass light in the third band. The first camera collects the first reflected light after passing through the first filter to obtain a wide field image of the third band. Adjust the filtering range of the second filter so that it can only pass light in the fourth band. Then, use the second camera to collect the second reflected light after passing through the second filter to obtain a wide-field image of the fourth band.

5. The 3D measurement error correction method based on differential parallel confocal focusing according to claim 1, characterized in that, The initial first band correction image is normalized using the following formula: ； in, This is the correction image for the first band. This is the initial first band correction image. For the first band, i ∈[1,2,3,..., N ], N The total number of bands; The method for normalizing the initial second-band correction image is the same as the method for normalizing the initial first-band correction image.

6. A 3D measurement error correction system based on differential parallel confocal focusing, applied to the 3D measurement error correction method based on differential parallel confocal focusing as described in any one of claims 1-5, characterized in that, The 3D measurement error correction system based on differential parallel confocal focusing includes: The image acquisition unit is used to acquire confocal images of the test sample under the first band illumination and the second band illumination, and wide-field images of the test sample under the third band illumination and the fourth band illumination, respectively, to obtain the first band confocal image, the second band confocal image, the third band wide-field image, and the fourth band wide-field image; wherein, the first band ≠ the second band, and the third and fourth bands are arbitrary bands; The first correction unit is connected to the image acquisition unit and is used to obtain the first band correction map based on the first band confocal map and the third band wide field map. The second correction unit is connected to the image acquisition unit and is used to obtain the second band correction map based on the second band confocal map and the fourth band wide field map. A differential signal determination unit is connected to the first correction unit and the second correction unit respectively, and is used to subtract the first band correction diagram from the second band correction diagram to obtain a differential signal diagram; A height measurement unit, connected to the differential signal determination unit, is used to convert the differential signal map into a height map according to a pre-determined differential signal axial response curve, so as to measure the surface height of the sample to be tested; the differential signal axial response curve is the relationship curve between the differential signal and the sample surface height.

Images