Transverse differential confocal sensing measurement method and device with large linear response range

Abstract

The present invention relates to the field of optical precision measurement technology and provides a method and device for transverse differential confocal sensing measurement with a large linear response range. This method uses the optical axis as the dividing line in a traditional confocal detection optical path to block half of the detection beam, converting the defocus variation of the measured surface into transverse differential confocal (TDC) detection of the lateral off-axis variation of the light spot on the focal plane. The light spot on the focal plane is then segmented using a detector array, and adjacent intensity signals are differentially subtracted from each other to obtain multiple groups of normalized TDC response curves. These linear segments are then fitted and spliced ​​to obtain differential confocal characteristic curves with a large linear response range, achieving wide-range sensing measurement while maintaining high resolution. This method provides a new high-precision, large-scale, rapid sensing and scanning measurement method for fields such as precision component surface shape and three-dimensional surface structure.

Description

Technical Field

[0001] The present invention relates to the field of optical precision measurement technology, and in particular to a transverse differential confocal sensing measurement method and device with a large linear response range. Background Art

[0002] Precision optical components are core components that determine the performance of optical systems and are widely used in important fields such as super-resolution optical imaging, semiconductor testing, and medical testing. Surface errors in optical components directly impact the imaging quality and measurement accuracy of optical systems. Therefore, high-precision surface measurement technology for optical components is key to improving optical system performance.

[0003] There are many methods for high-precision measurement of optical component surfaces, which are generally divided into two categories based on the measurement principle: contact measurement and non-contact measurement. Contact measurement methods primarily use instruments such as three-coordinate machines and profilometers to directly acquire surface coordinate information of the optical component being measured. However, these methods are slow to scan and can easily damage the surface of the optical component, making them unsuitable for high-precision measurement at the nanometer level. Non-contact measurement methods are divided into interferometry and scanning methods. Interferometry uses interference patterns formed by the superposition of light waves to analyze the surface shape of the measured surface. It is one of the methods for high-precision measurement of complex optical surfaces, and mainly includes zero interferometry and non-zero interferometry. Zero interferometry utilizes zero-optical elements such as CGHs to precisely compensate for the distorted wavefront of free-form surfaces. However, CGHs are expensive to manufacture, and due to the one-to-one nature of the compensator and the device under test, they have limited measurement versatility and a limited dynamic range. Furthermore, when the surface gradient varies significantly, the CGH's lines become dense, increasing manufacturing difficulties. Non-zero interferometry extends the dynamic measurement range through methods such as subaperture splicing or shearing interferometry, eliminating the need for zero-optical elements to compensate for wavefront distortion. However, these methods suffer from significant return errors during measurement. Scanning methods include structured light, spectral confocal imaging, and differential confocal imaging. Structured light measurements primarily acquire three-dimensional information about an object through optical triangulation, which is susceptible to interference from ambient light and generally results in micron-level measurement accuracy. Spectral confocal imaging, based on confocal microscopy, uses wavelength to encode axial position. Combined with a scanning galvanometer or microlens array for two-dimensional scanning, it achieves 3D measurement of the surface profile of the optical component under test. However, its poor axial sensitivity and the accuracy of its peak intensity location algorithm limit its measurement resolution.

[0004] The differential confocal method generates a highly sensitive axial light intensity response curve by differentially subtracting the pre-focus and post-focus light intensity response signals, both of which have the same defocus value. Its zero crossing point precisely corresponds to the focal position of the measurement beam, resulting in high axial resolution and high sensitivity. The response curve exhibits excellent linearity near the zero crossing point, allowing the sample's defocus value to be directly calculated from the light intensity value. This avoids errors introduced by mechanical motion during axial point-by-point scanning, enabling high-precision, fast, fixed-focus measurements of surface features through three-dimensional scanning. However, the misalignment of the two detectors' mounting positions and inconsistent response characteristics can introduce measurement errors.

[0005] Wang et al. proposed a new transverse differential confocal (TDC) method for free-form surface measurement. This method utilizes a D-type aperture to transform the dual-path axial defocus detection into a single-path off-axis detection in the focal plane. This allows differential confocal measurement with only one detector, avoiding measurement errors introduced by misalignment between the two detectors. However, this method suffers from a relatively small axial linear response range of only 1.2 μm, making it incapable of achieving high-precision, "axial-scan-free" fast sensing measurements over a wide range. Qiu et al. doubled the axial linear response range by interpolating the two signals and dividing them by the higher signal between them, but at the expense of axial resolution. Shao et al. further improved this approach by proposing a dual differential confocal method that achieves a 6.7 μm linear measurement range with an objective lens NA of 0.3. However, this method significantly sacrifices axial resolution, and the introduction of three pinhole detectors in the optical path increases measurement errors. None of these methods achieves both a large linear response range and high-precision axial resolution, significantly limiting the application of differential confocal sensing technology in various fields. Summary of the Invention

[0006] In view of the above problems, the present invention proposes a transverse differential confocal sensing measurement method and device with a large linear response range to overcome the above problems.

[0007] One aspect of the present invention provides a method for measuring transverse differential confocal sensing with a large linear response range, the method comprising:

[0008] S1. Construct a transverse differential confocal detection optical path, wherein the transverse differential confocal detection optical path is obtained based on an improved confocal optical path, the confocal optical path includes a focused light beam converged by an objective lens and a detection light beam converged by a converging lens, the focused light beam converges on the surface to be measured, half of the detection light beam is blocked between the converging lens and the detection focal plane of the converging lens with the optical axis as the dividing line, and a plurality of physical pinholes symmetrical based on the optical axis are provided at the detection focal plane, a multi-pixel detector is provided behind the physical pinhole, and each pixel on the multi-pixel detector performs segmented detection on the light spot irradiated on the detection focal plane to obtain a transverse differential confocal detection optical path;

[0009] S2. Obtaining a light intensity response signal curve of each pixel, wherein the light intensity response signal curve of each pixel is used to characterize a corresponding relationship between the light intensity response signal of each pixel and a defocus amount of a measured surface, wherein the defocus amount of the measured surface is a distance between the measured surface and a focal plane of the objective lens;

[0010] S3. Perform differential subtraction normalization processing on the light intensity response signal curves of two adjacent pixels to obtain a normalized TDC response curve:

[0011]

[0012] Among them, I n is the light intensity response signal curve of the nth pixel, I D(n) is a normalized TDC response curve of the n-th pixel, wherein the normalized TDC response curve is a normalized lateral differential confocal response curve;

[0013] S4. Selecting and fitting a linear segment from the normalized TDC response curve of each pixel to obtain a normalized TDC response fitting curve, wherein the value of the axial coordinate corresponding to the minimum value of the fitted linear segment of the nth pixel is equal to the value of the axial coordinate corresponding to the maximum value of the fitted linear segment of the (n+1)th pixel;

[0014] S5. Splice the normalized TDC response fitting linear segments of any two adjacent pixels to obtain a large linear sensing interval response curve, perform sensor measurement on the measured surface based on the acquired detection light intensity response signals of different pixels and the response curve of the large linear sensing interval, and calculate the defocus amount of the measured surface relative to the focal plane of the objective lens.

[0015] Furthermore, after obtaining the large linear sensing interval response curve, the method includes: measuring the surface shape of the measured surface using a transverse differential confocal detection optical path;

[0016] The method of measuring the shape of the surface to be measured by using the transverse differential confocal detection optical path includes:

[0017] S11, detecting the target detection position on the measured surface, and obtaining the detection light intensity response signal of each pixel corresponding to the target detection position;

[0018] S12, calculating the defocus amount of the measured surface at the target position based on the response curve of the large linear sensing interval and the detection light intensity response signal of each pixel corresponding to the target position;

[0019] S13, scanning and measuring the surface to be measured along the transverse direction using the transverse differential confocal detection optical path, so that the target detection position of the surface to be measured moves along the transverse direction, and repeating the operations of steps S11 to S12 to obtain the defocus amount of the surface to be measured at different transverse positions;

[0020] S14. Based on the defocus amounts of the measured surface at different lateral positions, plot the axial fluctuation changes of the measured surface in the lateral direction to obtain the surface shape of the measured surface.

[0021] Furthermore, before executing step S5, the method further includes: determining an optimal off-axis amount of the physical pinhole;

[0022] Determining the optimal off-axis amount of the physical pinhole includes:

[0023] Selecting physical pinholes with different off-axis values, obtaining the slopes of the fitted linear segments at the zero-crossing points of the normalized TDC response curve corresponding to the physical pinholes with different off-axis values; and selecting the optimal off-axis value range at the peak position of the corresponding relationship curve between the off-axis value and the slope of the fitted linear segment at the zero-crossing point;

[0024] Selecting a quasi-optimal off-axis value within the optimal off-axis value range, and obtaining a normalized TDC response curve of adjacent pixels corresponding to each quasi-optimal off-axis value;

[0025] The optimal off-axis value of the physical pinhole is determined based on the overlap width of the normalized TDC response curves of adjacent pixels along the axial direction, wherein the overlap width satisfies the following condition: 0≤overlap width≤preset width, and the overlap width is inversely proportional to the linear measurement range.

[0026] Furthermore, obtaining the light intensity response signal curve of each pixel includes:

[0027] Collect the light intensity response signals of the target pixels when the measured surface is at different defocus amounts;

[0028] A light intensity response signal curve of the target pixel is drawn based on the light intensity response signal of the target pixel when the measured surface is at different defocus amounts.

[0029] Another aspect of the present invention provides a large linear response range lateral differential confocal sensing measurement device, the device comprising: a confocal optical path system, an aperture assembly being arranged between a converging mirror in a detection optical path of the confocal optical path system and its detection focal plane, the aperture assembly blocking half of the optical path of the detection light beam with the optical axis as the dividing line, and a plurality of physical pinholes symmetrical based on the optical axis being arranged at the detection focal plane of the detection light beam, a multi-pixel detector being arranged behind the physical pinholes, and each pixel on the multi-pixel detector performing segmented detection on the light spot irradiated on the detection focal plane.

[0030] Furthermore, the aperture assembly also includes an aperture and a relay lens. The aperture is used to block half of the detection light beam with the optical axis as the dividing line, and the relay lens is used to image and amplify the blocked detection light beam.

[0031] Furthermore, the confocal optical path system includes an objective lens, a quarter-wave plate, a polarization beam splitter and a converging lens arranged in sequence, and a laser and a collimator are also arranged next to the polarization beam splitter; wherein, the laser light source emitted by the laser forms a parallel light beam after passing through the collimator, the parallel light beam is reflected by the polarization beam splitter, and is converged by the quarter-wave plate and the objective lens in sequence to form a focused light beam, the focused light beam is irradiated onto the surface to be measured, and is reflected by the surface to be measured to form a reflected light beam, the reflected light beam passes through the objective lens, the quarter-wave plate and the converging lens in sequence, and the converging lens converges the reflected light beam to form a detection light beam; the detection light beam is focused onto a multi-pixel detector after passing through the aperture assembly, and the pixels on the multi-pixel detector segment and detect the focused light spot to obtain a large linear response curve corresponding to the defocus amount of the measured surface.

[0032] The present invention provides a large linear response range lateral differential confocal sensing measurement method and device. In the traditional confocal detection optical path, half of the detection light beam is blocked with the optical axis as the dividing line, and the defocus change of the measured surface is converted into the lateral off-axis change of the light spot on the detection focal plane, thereby realizing lateral differential detection; then the light spot on the detection focal plane is divided by a detector array, and multiple groups of normalized TDC response curves are obtained by differential subtraction between them, and their linear segments are fitted and spliced, thereby realizing large-scale sensing measurement while taking into account the resolution. It provides a new method of high-precision, large-scale, rapid sensing scanning measurement for the fields of precision component surface shape, three-dimensional surface structure, etc.

[0033] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are specifically listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Various other advantages and benefits will become apparent to those skilled in the art by reading the detailed description of the preferred embodiment below. The accompanying drawings are only for the purpose of illustrating the preferred embodiment and are not to be considered as limiting the present invention. In the accompanying drawings:

[0035] Figure 1 (a) Schematic diagram of the optical path of the transverse differential confocal sensing measurement method with a large linear response range;

[0036] Figure 1 (b) Schematic diagram of the light spot position on the detection focal plane when the measured surface is at different defocus values;

[0037] Figure 1 (c) Normalized TDC response fitting curves at different pixels;

[0038] Figure 1 (d) TDC response curve in the large linear sensing range;

[0039] Figure 2 It is the corresponding relationship diagram of the normalized TDC response curve and the light intensity response signal curve;

[0040] Figure 3 Fit the normalized TDC response curve for a pixel;

[0041] Figure 4 The normalized TDC response curve corresponding to different off-axis values;

[0042] Figure 5 is the corresponding relationship curve between the slope of the fitted linear segment of the normalized TDC response curve and the off-axis value;

[0043] Figure 6 This is a schematic diagram of the relationship between excessive overlap of fitted linear segments;

[0044] Figure 7 This is a schematic diagram of the relationship where the fitted linear segments cannot be effectively spliced;

[0045] Figure 8 The relationship curve between the slope of the normalized TDC response fitting curve and the goodness of fit as the fitting range coefficient increases;

[0046] Figure 9 R f The fitting results of the linear response range when =0.71;

[0047] Figure 10 is a spliced ​​sensing measurement curve of a specific embodiment;

[0048] Figure 11 The figure shows a comparison between the step measurement results of an embodiment of the present invention and the measurement results of a laser confocal microscope;

[0049] Figure 12 The diagram shows a comparison of step measurement results at different heights using the sensor of an embodiment of the present invention. DETAILED DESCRIPTION

[0050] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0051] Those skilled in the art will understand that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by those skilled in the art in the art to which the present invention pertains. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with those in the context of the prior art and, unless specifically defined, will not be interpreted in an idealized or overly formal sense.

[0052] Figure 1 FIG. 1 shows a schematic diagram of a large linear response range lateral differential confocal sensing measurement method according to an embodiment of the present invention, as shown in FIG. Figure 1 The optical system shown in (a) constructs a transverse differential confocal detection optical path for detecting the defocus and surface shape of the measured surface. This transverse differential confocal detection optical path is implemented based on the large linear response range transverse differential confocal sensing measurement device of an embodiment of the present invention. It should be noted that this device is an improvement on a traditional confocal optical path system.

[0053] Further, if Figure 1 As shown in (a), the confocal optical system for forming a confocal optical path includes an objective lens 1, a quarter-wave plate 2, a polarization beam splitter 3, and a converging lens 4, which are arranged in sequence. A laser 9 and a collimator 8 are also arranged next to the polarization beam splitter 3.

[0054] Furthermore, the laser light source emitted by the laser 9 forms a parallel beam after passing through the collimator 8, and the parallel beam is reflected by the polarization beam splitter 3 to form a p-light parallel beam, which passes through the quarter-wave plate 2 and the objective lens 1 in sequence. The quarter-wave plate 2 adjusts the polarization direction of the p-light polarized beam and then converges through the objective lens 1 to form a focused beam. The focused beam is irradiated to the surface to be measured and reflected by the surface to be measured to form a reflected beam. The reflected beam passes through the objective lens 1, the quarter-wave plate 2, and the converging lens 4 in sequence. The converging lens 4 converges the reflected beam to form a detection beam.

[0055] It should be noted that the confocal optical path system provided in the embodiment of the present invention is a specific embodiment, the purpose of which is to form a convergent light beam that converges to the surface to be measured and a detection light beam formed after the surface to be measured is reflected, and the convergent light beam and the detection light beam can achieve confocality.

[0056] Furthermore, in an embodiment of the present invention, Figure 1 As shown in (a), half of the detection beam is blocked with the optical axis as the dividing line before the detection focal plane of the detection beam and after the converging mirror 4 (i.e., between the converging mirror 4 and the detection focal plane of the converging mirror 4), and multiple physical pinholes 7 are provided at the detection focal plane symmetrically based on the optical axis to convert the defocus change of the measured surface into the lateral off-axis change of the light spot on the detection focal plane, thereby realizing lateral differential detection (such as Figure 1 (as shown in (b)).

[0057] Furthermore, a multi-pixel detector is provided behind the physical pinhole 7, and each pixel on the multi-pixel detector performs segmented detection on the light spot irradiated on the detection focal plane to obtain a light intensity response signal corresponding to each pixel, thereby forming a lateral differential confocal detection optical path of an embodiment of the present invention.

[0058] Furthermore, the present invention adopts an aperture component to block the detection light beam. The aperture component can be a D-shaped aperture 5. The aperture 5 is arranged close to the converging lens 4 and blocks half of the detection light beam with the optical axis as the dividing line.

[0059] As a preferred embodiment of the present invention, the aperture assembly includes an aperture 5 and a relay lens 6. Since the focused light spot at the focal plane of the converging mirror is small, the present invention introduces the relay lens 6 to amplify the detection light beam. It can be understood that the relay lens 6 causes the focal plane of the detection light beam to move back to the focal position of the relay lens 6.

[0060] As shown in the accompanying drawings, the defocus of the measured surface is the distance between the measured surface and the focal plane of the objective lens, which can be expressed as z0, z1 and z2. When the measured surface undergoes relative axial changes near the focus z0 and z2, the Airy disk of the focal plane will produce lateral off-axis movement. At the same time, each pixel on the detector divides and detects the corresponding light intensity response signal, converting the axial defocus detection into a lateral off-axis detection. Therefore, the present invention obtains multiple groups of normalized lateral differential confocal response curves (also expressed as normalized TDC response curves in this article) by performing differential subtraction normalization on the light intensity response signals of each adjacent pixel, and fits and splices their linear response ranges. According to the response light intensity within its linear region, the defocus change of the measured surface at this time can be accurately calculated, thereby realizing a large range of high-precision axial rapid sensing measurements.

[0061] The following describes the splicing principle of the fitting curves according to the embodiment of the present invention with reference to the accompanying drawings:

[0062] The optical path of the large linear response range lateral differential confocal sensing measurement in the embodiment of the present invention is based on the lateral differential confocal optical path, which detects the light intensity response signal I of two adjacent pixels on the plane. n and I n+1 Differential subtraction obtains the lateral differential confocal light intensity response signal I diff(n) , I diff(n) Satisfy the following calculation formula:

[0063] I diff(n) =I n -I n+1 (1)

[0064] Furthermore, in order to suppress the influence of reflectivity change caused by the flatness and local tilt of the measured surface, the embodiment of the present invention also normalizes the light intensity response signal to obtain a normalized TDC response curve I D(n) , the formula is as follows:

[0065]

[0066] Among them, I n is the light intensity response signal curve of the nth pixel, I D(n) is the normalized TDC response curve of the n-th pixel, where the normalized TDC response curve is also called the normalized lateral differential confocal light intensity response curve.

[0067] Furthermore, the corresponding relationship between the light intensity response signal curve of two adjacent pixels and the normalized TDC response curve is as follows: Figure 2 As shown, the normalized axial coordinate u of the measured surface can be expressed as:

[0068]

[0069] Where D is the beam aperture, f s is the focal length of the objective lens, z is the defocus of the measured surface, and λ is the laser wavelength.

[0070] Figure 3 The normalized TDC response curve corresponding to one pixel in the embodiment of the present invention is shown. Figure 3 It can be seen that the normalized TDC response curve has good linearity near the zero-crossing point. Therefore, the fitting linear interval of the part with good linearity in the normalized TDC response curve can be selected as the fitting linear segment, and linear fitting is performed to obtain the normalized TDC response fitting straight line. The linear function of the response light intensity of the pixel and the height change of the measured surface can be obtained, so that the defocus amount of the measured surface can be directly calculated according to the response light intensity, realizing high-precision and fast sensing measurement without "axial scanning".

[0071] Figure 1 (c) shows the principle diagram of splicing the normalized TDC response fitting curve in an embodiment of the present invention. As the Airy disk on the detection plane moves laterally off-axis, each pixel on the detector obtains a light intensity response signal in turn, where the light intensity response signal I generated by any three adjacent pixels is n , I n+1 and I n+2 Differential subtraction and normalization are performed in sequence to obtain two sets of adjacent normalized TDC response curves I D(n) and I D(n+1) , where M n N n and M n+1 N n+1 Respectively represent ID(n) and I D(n+1) The light intensity data range of the fitted linear segment is determined by selecting an appropriate fitting range coefficient R f To determine, its value range is [0,1], that is, the range of the light intensity data used for fitting is [R f ×max(I D(n) ),R f ×min(I D(n) )] and [R f ×max(I D(n+1) ),R f ×min(I D(n+1) )]. I D(n) and I D(n+1) The linear sensing measurement equation of the fitting area can be expressed as:

[0072]

[0073] where k (n) 、b n and k (n+1) 、b n+1 Represents linear segment I DL(n) and I DL(n+1) The linear fitting coefficient of DL(n) Minimum point R f ×min(I D(n) ) corresponds to the normalized horizontal coordinate u (n)min Equal to I DL(n+1) Maximum point R f ×max(I D(n+1) ) corresponds to the normalized horizontal coordinate u (n+1)max When , the stitching condition of the linear response range of two adjacent lateral differential confocal curves is met, and the normalized TDC response fitting linear segments are stitched together to obtain the following: Figure 1 (d) The large linear sensing range response curve is shown.

[0074] Therefore, based on the stitching principle of the above-mentioned fitting curves, an embodiment of the present invention provides a lateral differential confocal sensing measurement method with a large linear response range, which specifically includes the following steps:

[0075] S1. Construct a transverse differential confocal detection optical path, wherein the transverse differential confocal detection optical path is obtained based on an improved confocal optical path, the confocal optical path includes a focused light beam converged by an objective lens and a detection light beam converged by a converging lens, the focused light beam converges on the surface to be measured, half of the detection light beam is blocked between the converging lens and the detection focal plane of the converging lens with the optical axis as the dividing line, and a plurality of physical pinholes symmetrical based on the optical axis are provided at the detection focal plane, a multi-pixel detector is provided behind the physical pinhole, and each pixel on the multi-pixel detector performs segmented detection on the light spot irradiated on the detection focal plane to obtain a transverse differential confocal detection optical path;

[0076] S2. Obtaining a light intensity response signal curve of each pixel, wherein the light intensity response signal curve of each pixel is used to characterize a corresponding relationship between the light intensity response signal of each pixel and a defocus amount of a measured surface, wherein the defocus amount of the measured surface is the distance between the measured surface and a focal plane of the objective lens;

[0077] In the embodiment of the present invention, the longitudinal coordinate of the light intensity response signal curve of the pixel is the light intensity response signal, and the transverse coordinate is the normalized axial coordinate. The normalized axial coordinate u is as shown in formula (3).

[0078] S3. Perform differential subtraction normalization processing on the light intensity response signal curves of two adjacent pixels to obtain a normalized TDC response curve.

[0079] The specific implementation method of performing differential subtraction normalization processing on the light intensity response signal curves of two adjacent pixels has been described in the above embodiment and will not be repeated here. In addition, the normalized TDC response curve is the normalized lateral differential confocal response curve.

[0080] S4. Select a fitting linear segment for the normalized TDC response curve of each pixel and perform fitting to obtain a normalized TDC response fitting curve.

[0081] In the embodiment of the present invention, the value of the axial coordinate corresponding to the minimum value of the fitted linear segment of the nth pixel is equal to the value of the axial coordinate corresponding to the maximum value of the fitted linear segment of the n+1th pixel. DL(n) Minimum point R f ×min(I D(n) ) corresponds to the normalized horizontal coordinate u (n)min Equal to I DL(n+1) Maximum point R f ×max(I D(n+1) ) corresponds to the normalized horizontal coordinate u (n+1)max When , the stitching condition of the linear response range of two adjacent lateral differential confocal curves can be met.

[0082] S5. Splice the normalized TDC response fitting linear segments of any two adjacent pixels to obtain a large linear sensing interval response curve, perform sensor measurement on the measured surface based on the acquired detection light intensity response signals of different pixels and the response curve of the large linear sensing interval, and calculate the defocus amount of the measured surface relative to the focal plane of the objective lens.

[0083] It should be noted that in order to maintain high measurement resolution while expanding the linear response range as much as possible, it is necessary to comprehensively analyze the linear segment sensitivity S and the fitting range coefficient R f , goodness of fit R 2 and the influence of key parameters such as the linear measurement range to determine the optimal off-axis amount of the physical pinhole 7 in the measurement optical path and the optimal fitting range coefficient in the fitting process.

[0084] Furthermore, the optimal off-axis amount of the physical pinhole 7 in the measurement optical path is determined specifically as follows: a physical pinhole 7 with different off-axis amounts is selected, the slope of the normalized TDC response curve corresponding to the physical pinhole 7 at the zero-crossing point of the fitting linear segment is obtained when the physical pinhole 7 has different off-axis amounts, and the optimal off-axis amount value range is selected at the peak position of the slope of the zero-crossing point of the fitting linear segment; a quasi-optimal off-axis amount is selected within the optimal off-axis amount value range, and the normalized TDC response curve of the adjacent pixels corresponding to each quasi-optimal off-axis amount is obtained; the optimal off-axis amount of the physical pinhole 7 is determined based on the overlapping width of the normalized TDC response curves of the adjacent pixels along the axial direction, wherein the overlapping width satisfies the following conditions: 0≤overlap width≤preset width, and the overlapping width is inversely proportional to the linear measurement range.

[0085] In the embodiment of the present invention, by executing the operations from step S3 to step S4 , the slope of the normalized TDC response curve corresponding to each off-axis amount in the fitted linear segment can be obtained.

[0086] Specifically, according to the transverse differential confocal measurement principle, when the off-axis amount of the physical pinhole 7 changes, the slope of the transverse differential confocal response curve also changes, thereby affecting the axial resolution of the measurement. The slope of the normalized TDC response curve at the zero crossing point can be obtained by D(u) Taking the partial derivative of u we get:

[0087]

[0088] Figure 4 Shows different off-axis values ​​v M The normalized TDC response curve, Figure 5 The slope S of the zero-crossing point of the normalized TDC response curve fitting linear segment is shown as v M The change curve of Figure 5 It can be seen that as the off-axis distance v M As the value of v increases, the slope of the zero-crossing point increases first and then decreases.M =6.2, the normalized TDC response curve has the highest measurement sensitivity, but the linear response range is small at this time, which affects the splicing and expansion of adjacent linear segments.

[0089] Therefore, in the process of stitching, it is necessary to ensure that the linear segments of the normalized TDC response curves of adjacent pixels have an overlap width on the axis coordinate. If the overlap width is too wide or there is no overlap width, the stitching condition cannot be met.

[0090] Figure 6 、 Figure 7 Shows v M =3.0, v M = 5.0, we can find that when v M = 3.0, the linear part of the normalized TDC response curve is excessively overlapped, which is not conducive to range expansion; when v M =5.0, I D(n) The normalized horizontal coordinate corresponding to the minimum point is less than I D(n+1) The normalized horizontal coordinate corresponding to the maximum value point and the linear response range cannot be effectively spliced.

[0091] Therefore, in the specific embodiment of the present invention, the fitting range coefficient R is selected f =0.7, assuming the axial normalized scanning interval δ u Set to 0.01, through simulation we can see that when v M When it is less than 4.17, I DL(n) The normalized horizontal coordinate corresponding to the minimum point is equal to I DL(n+1) The normalized horizontal coordinate corresponding to the maximum point can satisfy the stitching of the linear response range of adjacent normalized lateral differential confocal response curves.

[0092] Furthermore, in an embodiment of the present invention, the optimal fitting range coefficient includes: obtaining a normalized TDC response curve of a pixel, fitting the fitted linear segment of the current normalized TDC response curve using different fitting range coefficients to obtain normalized TDC response fitting curves corresponding to different fitting range coefficients; calculating the fitting goodness of the normalized TDC response fitting curves corresponding to different fitting range coefficients; and determining the optimal fitting range coefficient based on the fitting goodness, the slope of the fitted linear segment, and the light intensity data range of the fitted linear segment.

[0093] Specifically, in order to ensure that the linear segment of the fitting has a high linearity, the present invention uses the goodness of fit coefficient R of the fitting curve to determine the linear segment of the fitting curve. 2 The linear response range is evaluated for straightness, and the calculation formula is as follows:

[0094]

[0095] Among them, SSR is the sum of squares of the differences between the fitted data and the mean of the original data, and SST is the sum of squares of the differences between the original data and the mean of the original data. 2 The value range of R is [0,1]. 2 The closer the value of is to 1, the better the fitting quality.

[0096] To further illustrate, the slope and goodness of fit of the linear segment under different fitting range coefficients are simulated and analyzed. The results are as follows Figure 8 As shown. The results show that the smaller the fitting range coefficient, that is, the smaller the fitting linear response range, the higher the slope of the fitting linear segment, the higher the axial resolution of the system, and the greater the goodness of fit. However, if the fitting data segment range is too small, random errors will be introduced. Therefore, in the embodiment of the present invention, the best fitting range coefficient can be determined based on the slope of the fitting linear segment and the fitting data segment range, wherein the slope of the fitting linear segment is inversely proportional to the best fitting range coefficient. Taking all factors into consideration, the best fitting range R in the preferred embodiment of the present invention is f Take 0.7.

[0097] According to the stitching condition of the linear response range of any two adjacent lateral differential confocal curves: I D(n) The normalized horizontal coordinate corresponding to the minimum point is equal to I D(n+1) The normalized horizontal coordinate corresponding to the maximum value point can be obtained by simulation calculation. f =0.71, the linear response range after splicing is the measurement sensitivity S LS =0.49,R 2 =0.9984. The fitting result of the linear response range is as follows: Figure 9 As shown, the normalized axial coordinate range of the linear response interval is approximately [-1.55, 1.55].

[0098] Furthermore, to more intuitively determine the correspondence between the light intensity response signals of different pixels and the surface shape measurement curve, in an embodiment of the present invention, after splicing the normalized TDC response fitting curves of any two adjacent pixels, the method further includes performing coordinate transformation on each of the spliced ​​fitting curves to obtain a curve representing the correspondence between the normalized differential values ​​of the light intensity response signals of each pixel and the off-axis amount, and using the curve representing the correspondence between the normalized differential values ​​of the light intensity response signals of each pixel and the off-axis amount as the surface shape measurement curve. Formula (3) shows the correspondence between the normalized axial coordinate u and the defocus amount z, and thus the axial coordinate can be converted into the defocus amount z. The specific conversion process will not be further described in this invention.

[0099] In a specific embodiment of the present invention, a silver-plated reflector is used as a sample to be measured to obtain the light intensity response signal curve of each pixel and the response curve of its large linear sensing range.

[0100] It should be noted that in actual operation, the axial movement of the focus of the focused light beam is achieved by driving the objective lens 1 to move, specifically, the objective lens 1 is driven to move in the axial direction to achieve axial scanning of the sample to be measured, and the light intensity response curves I1…I 12 The two-by-two differential subtraction normalization process is performed to obtain 11 groups of normalized TDC response curves I D1 …I D11 , the surface shape measurement curve after fitting, splicing and coordinate transformation is as follows Figure 10 shown.

[0101] Furthermore, the cubic fitting sensor measurement equation for each normalized TDC response curve can be expressed as follows:

[0102]

[0103] Furthermore, after obtaining the response curve of the large linear sensing range, the embodiment of the present invention can be used to detect the surface shape of the measured surface. Therefore, after step S5, the embodiment of the present invention further includes measuring the surface shape of the measured surface using a transverse differential confocal detection optical path. The specific method steps are as follows:

[0104] S11, detecting the target detection position on the measured surface, and obtaining the detection light intensity response signal of each pixel corresponding to the target detection position;

[0105] S12, calculating the defocus amount of the measured surface at the target position based on the response curve of the large linear sensing interval and the detection light intensity response signal of each pixel corresponding to the target position;

[0106] S13, scanning and measuring the surface to be measured along the transverse direction using the transverse differential confocal detection optical path, so that the target detection position of the surface to be measured moves along the transverse direction, and repeating the operations of steps S11 to S12 to obtain the defocus amount of the surface to be measured at different transverse positions;

[0107] S14. Based on the defocus amounts of the measured surface at different lateral positions, plot the axial fluctuation changes of the measured surface in the lateral direction to obtain the surface shape of the measured surface.

[0108] Furthermore, the present invention also verifies the measurement method. Specifically, the SHS series Si-5-MD standard step sample produced by AppliedNano is used as the measured sample. The step height reference value is 6.261 μm and the step period is 50 μm. The sample is measured using a laser confocal microscope (LCM) (Olympus LEXT OLS 4000) and the TDCS-LSMR measurement method of the present invention (hereinafter referred to as TDCS-LSMR). LCM requires axial tomography, and the measurement time is 17.8 s. Figure 11 As shown in (a), the step height measurement result is 6.033μm; Figure 11 As shown in (b), TDCS-LSMR has a large sensing measurement range. It only needs to scan one layer within the response range, without axial scanning. With the high-precision horizontal translation stage, the measurement time is 3.0s, and the step height measurement result is 6.037μm. Figure 11 (b) The profiles of the step standard measured by TDCS-LSMR (solid line) and LCM (dashed line) are also compared, and the two are very similar. This shows that TDCS-LSMR has good axial measurement performance and faster measurement speed over a large sensing range.

[0109] Furthermore, in order to verify the measurement performance and accuracy of the large sensing range of TDCS-LSMR, a Figure 12 In the experimental setup shown in (a), after the sensor is fixed and adjusted, the x-axis is driven to move horizontally to scan and measure the step sample. After the measurement is completed and reset, the VCM is driven to move vertically upward along the z-axis by 5μm. The above steps are repeated to scan and measure again. The experimental results of the large sensing measurement range of the step sample before and after the upward movement are shown in Figure 2. Figure 12 As shown in (b), the measurement results of the step standard before and after the upward movement are 6.034μm and 6.035μm, respectively, which are very close to the LCM. In addition, the total sensing measurement range before and after the upward movement reaches 11.034μm, which can realize a large range of "non-axial scanning" fast sensing measurement.

[0110] The present invention provides a large linear response range lateral differential confocal sensing measurement method and device. In the traditional confocal detection optical path, half of the detection light beam is blocked with the optical axis as the dividing line, and the defocus change of the measured surface is converted into the lateral off-axis change of the light spot on the detection focal plane, thereby realizing lateral differential detection; then the light spot on the detection focal plane is divided by a detector array, and multiple groups of normalized TDC response curves are obtained by differential subtraction between them, and their linear segments are fitted and spliced, thereby realizing large-scale sensing measurement while taking into account the resolution. It provides a new method of high-precision, large-scale, rapid sensing scanning measurement for the fields of precision component surface shape, three-dimensional surface structure, etc.

[0111] Furthermore, those skilled in the art will appreciate that although some embodiments herein include certain features included in other embodiments but not other features, combinations of features from different embodiments are intended to be within the scope of the present invention and to form different embodiments. For example, any of the claimed embodiments may be used in any combination.

[0112] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A transverse differential confocal sensing measurement method with a large linear response range, characterized in that: The method comprises: S1. Construct a transverse differential confocal detection optical path, wherein the transverse differential confocal detection optical path is obtained based on an improved confocal optical path, the confocal optical path includes a focused light beam converged by an objective lens and a detection light beam converged by a converging lens, the focused light beam converges on the surface to be measured, half of the detection light beam is blocked between the converging lens and the detection focal plane of the converging lens with the optical axis as the dividing line, and a plurality of physical pinholes symmetrical based on the optical axis are provided at the detection focal plane, a multi-pixel detector is provided behind the physical pinhole, and each pixel on the multi-pixel detector performs segmented detection on the light spot irradiated on the detection focal plane to obtain a transverse differential confocal detection optical path; S2. Obtaining a light intensity response signal curve of each pixel, wherein the light intensity response signal curve of each pixel is used to characterize a corresponding relationship between the light intensity response signal of each pixel and a defocus amount of a measured surface, wherein the defocus amount of the measured surface is a distance between the measured surface and a focal plane of the objective lens; S3. Perform differential subtraction normalization processing on the light intensity response signal curves of two adjacent pixels to obtain a normalized TDC response curve: Among them, I n is the light intensity response signal curve of the nth pixel, I D(n) is a normalized TDC response curve of the n-th pixel, wherein the normalized TDC response curve is a normalized lateral differential confocal response curve; S4. Selecting and fitting a linear segment from the normalized TDC response curve of each pixel to obtain a normalized TDC response fitting curve, wherein the value of the axial coordinate corresponding to the minimum value of the fitted linear segment of the nth pixel is equal to the value of the axial coordinate corresponding to the maximum value of the fitted linear segment of the (n+1)th pixel; S5. Splice the normalized TDC response fitting linear segments of any two adjacent pixels to obtain a large linear sensing interval response curve, perform sensor measurement on the measured surface based on the acquired detection light intensity response signals of different pixels and the response curve of the large linear sensing interval, and calculate the defocus amount of the measured surface relative to the focal plane of the objective lens.

2. The method according to claim 1, characterized in that After obtaining the large linear sensing interval response curve, the method includes: measuring the surface shape of the measured surface using a transverse differential confocal detection optical path; The method of measuring the shape of the surface to be measured by using the transverse differential confocal detection optical path includes: S11, detecting the target detection position on the measured surface, and obtaining the detection light intensity response signal of each pixel corresponding to the target detection position; S12, calculating the defocus amount of the measured surface at the target position based on the response curve of the large linear sensing interval and the detection light intensity response signal of each pixel corresponding to the target position; S13, scanning and measuring the surface to be measured along the transverse direction using the transverse differential confocal detection optical path, so that the target detection position of the surface to be measured moves along the transverse direction, and repeating the operations of steps S11 to S12 to obtain the defocus amount of the surface to be measured at different transverse positions; S14. Based on the defocus amounts of the measured surface at different lateral positions, plot the axial fluctuation changes of the measured surface in the lateral direction to obtain the surface shape of the measured surface.

3. The method according to claim 1 or 2, characterized in that Before executing step S5, the method further includes: determining an optimal off-axis amount of the physical pinhole; Determining the optimal off-axis amount of the physical pinhole includes: Selecting physical pinholes with different off-axis values, obtaining the slopes of the fitted linear segments at the zero-crossing points of the normalized TDC response curve corresponding to the physical pinholes with different off-axis values; and selecting the optimal off-axis value range at the peak position of the corresponding relationship curve between the off-axis value and the slope of the fitted linear segment at the zero-crossing point; Selecting a quasi-optimal off-axis value within the optimal off-axis value range, and obtaining a normalized TDC response curve of adjacent pixels corresponding to each quasi-optimal off-axis value; The optimal off-axis value of the physical pinhole is determined based on the overlap width of the normalized TDC response curves of adjacent pixels along the axial direction, wherein the overlap width satisfies the following condition: 0≤overlap width≤preset width, and the overlap width is inversely proportional to the linear measurement range.

4. The method according to claim 1 or 2, characterized in that The obtaining of the light intensity response signal curve of each pixel comprises: Collect the light intensity response signals of the target pixels when the measured surface is at different defocus amounts; A light intensity response signal curve of the target pixel is drawn based on the light intensity response signal of the target pixel when the measured surface is at different defocus amounts.

5. A lateral differential confocal sensing measurement device for the large linear response range lateral differential confocal sensing measurement method according to claim 1, characterized in that: The device includes: a confocal optical path system, wherein an aperture assembly is arranged between a converging mirror in a detection optical path of the confocal optical path system and its detection focal plane, the aperture assembly blocks half of the detection light beam with the optical axis as a dividing line, and a plurality of physical pinholes symmetrically based on the optical axis are arranged at the detection focal plane of the detection light beam, and a multi-pixel detector is arranged behind the physical pinholes, and each pixel on the multi-pixel detector divides and detects the light spot irradiated on the detection focal plane.

6. The lateral differential confocal sensing measurement device according to claim 5, characterized in that: The aperture assembly further comprises an aperture and a relay lens. The aperture is used to block half of the optical path of the detection light beam with the optical axis as the dividing line, and the relay lens is used to image and amplify the blocked detection light beam.

7. The lateral differential confocal sensing measurement device according to claim 5, characterized in that: The confocal optical path system includes an objective lens, a quarter-wave plate, a polarization beam splitter and a converging lens arranged in sequence, and a laser and a collimator are also arranged next to the polarization beam splitter. The laser light source emitted by the laser forms a parallel light beam after passing through the collimator. The parallel light beam is reflected by the polarization beam splitter and converged by the quarter-wave plate and the objective lens in sequence to form a focused light beam. The focused light beam is irradiated onto the surface to be measured and reflected by the surface to form a reflected light beam. The reflected light beam passes through the objective lens, the quarter-wave plate and the converging lens in sequence, and the converging lens converges the reflected light beam to form a detection light beam. The detection light beam is focused onto a multi-pixel detector after passing through an aperture assembly. The pixels on the multi-pixel detector segment and detect the focused light spot to obtain a large linear response curve corresponding to the defocus amount of the surface to be measured.

Images