A dynamic optical thickness measurement method and system based on multi-band guidance

Abstract

This invention discloses a dynamic optical thickness measurement method and system based on multi-band guidance. The system uses a 1310nm wide-spectrum SLD light source and a 1550nm tunable narrowband laser to form a dual-band interference optical path. The beam is combined and split by a wavelength division multiplexer. The core of the system is the use of an FPGA as a heterogeneous computing unit to process the dual interference signals in parallel: using the wide-spectrum signal for coarse positioning of the optical thickness measurement range; using the narrowband signal to achieve sub-nanometer resolution fine measurement; and innovatively introducing a dynamic weighted fusion strategy based on signal-to-noise ratio and the degree of vibration of the measurement results to adaptively weight and fuse the coarse and fine measurement results, which significantly improves the accuracy, anti-interference ability and real-time performance of optical element thickness and spacing measurement in complex industrial environments.

Description

Technical Field

[0001] This invention relates to the field of optical thickness measurement technology, specifically to a dynamic optical thickness measurement method and system based on multi-band guidance, suitable for nanometer-level measurement of the center thickness of optical lenses and the thickness of multiple mirror surfaces in complex environments. Background Technology

[0002] Traditional lens thickness measurement uses an envelope detection algorithm, which is affected by spectral leakage, resulting in a positioning error of ±0.5μm. At the same time, it uses a single band for detection, which has low positioning accuracy, weak anti-interference ability, and is easily affected by environmental vibration, which can cause the signal-to-noise ratio to deteriorate by more than 30%, thereby increasing the detection error. It also has weak real-time detection capability and cannot achieve accurate measurement of optical thickness. Summary of the Invention

[0003] Technical Objective: To address the shortcomings of existing optical thickness measurement methods, this invention discloses a dynamic optical thickness measurement method and system based on multi-band guidance. By combining wide-spectrum coarse positioning with narrow-band fine measurement, the problem of inaccurate positioning in a single band is solved, resulting in an optical thickness measurement accuracy better than ±50nm. Furthermore, multi-band fusion and dynamic weight optimization calculation improve the robustness of the measurement system to environmental vibrations and low signal-to-noise ratio conditions; the measurement error is reduced by more than 30% as external interference changes.

[0004] Technical solution: To achieve the above technical objectives, the present invention adopts the following technical solution:

[0005] A dynamic optical thickness measurement method based on multi-band guidance includes the following steps:

[0006] S01. A broadband light source is used as the coarse positioning light source, and the light wave emitted by a narrowband tunable laser is used as the fine measurement light source. An interference optical path is constructed using the coarse positioning light source, the fine measurement light source, the sample to be tested, a reference mirror, and a reflector. After the coarse positioning light source is split, it illuminates the sample to be tested and the reference mirror respectively. After the fine measurement light source is split, it illuminates the reference mirror and the reflector respectively. The returned light rays interfere with each other to form interference light of corresponding wavelength bands.

[0007] S02. The returned interference light is split into beams according to wavelength, and the broadband interference light signal and narrowband interference light signal after beam splitting are converted into electrical signals respectively;

[0008] S03. Use the broadband interference electrical signal corresponding to the broadband interference optical signal to roughly locate the optical thickness measurement range and confirm the optical thickness measurement range;

[0009] S04. Based on the confirmed optical thickness measurement range, use the narrowband interference electrical signal corresponding to the narrowband interference optical signal to perform distance conversion, obtain an accurate optical path difference measurement value, and use the optical path difference measurement value as the optical thickness value.

[0010] Preferably, when using broadband interferometric electrical signals for coarse positioning, the present invention first uses broadband interferometric electrical signals to coarsely locate the peak region, then extracts the envelope signal through Hilbert transform, and determines the accurate position of the envelope peak, thereby achieving coarse positioning of the optical thickness measurement range.

[0011] Preferably, when determining the envelope peak position, the present invention performs the determination of the acquired broadband interference signal I. 1310 (z) Perform Hilbert transform to obtain the analytic signal H[I 1310 [(z)], calculate the signal envelope E1(z): ;

[0012] A sliding window mean filter is applied to the envelope signal E1(z) to eliminate high-frequency noise. A third-order polynomial is used to fit the envelope peak region. The location of the envelope peak is determined by solving for the points where the first derivative of the fitting function is zero. Based on the location of the envelope peak, the optical measurement values ​​of the coarsely positioned light source are obtained through distance conversion. .

[0013] Preferably, in the mean filtering process, the window size is 11-21 points. After filtering, the maximum value of the envelope is found, and the 50 data points before and after the maximum value are taken as the envelope peak region. The least squares method is used to perform a third-order polynomial. Fitting; by solving the first derivative of the fitting function The root is found, and the root within the envelope peak region is selected to obtain the accurate location of the envelope peak.

[0014] Preferably, when using narrowband interferometric signals for distance conversion, the present invention locates the sampling interval where the zero-crossing point of the signal is located by the sign change of the signal. Within the sampling interval, the three-point interpolation method is used to determine the position of the zero-crossing point and the phase corresponding to the zero-crossing point. where n is the refractive index. The x-coordinate value corresponds to the zero-crossing point; an FFT transform is performed on the narrowband interference electrical signal, and the dominant frequency f is obtained by peak finding. peak Then, the absolute estimate of the optical path difference can be calculated. Determine the interference order m; precisely measure the optical values ​​of the light source. , This indicates the wavelength of the precision-measured light source.

[0015] Preferably, the process of determining the zero-crossing position within the sampling interval using the three-point interpolation method of the present invention includes: sequentially selecting three sampling points (z0, z ...) within the sampling interval. k-1 ,I k-1 ),(z k ,I k ),(z k+1 ,Ik+1 By fitting a quadratic curve I(z) to three sampling points and solving for the roots of I(z)=0, the position of the zero-crossing point can be obtained. .

[0016] Preferably, in step S04 of the present invention, the optical measurement values ​​of the coarse positioning light source are assigned based on the signal-to-noise ratio of the broadband interference signal and the narrowband interference signal, as well as the vibration degree of the narrowband interference signal. Optical measurement values ​​of precision light source Different weights are used for heterogeneous weighting, and the final weighted result is used as the actual optical thickness measurement value; actual optical thickness measurement value ,in =1, The calculated weights are the optical measurements corresponding to the coarse positioning light source. The calculation weights are the optical measurement values ​​corresponding to the precision light source.

[0017] Preferably, the present invention increases the signal-to-noise ratio (SNR) when the narrowband interference signal vibration increases drastically or the signal-to-noise ratio (SNR2) of the narrowband interference signal is lower than a threshold. Conversely, it increases. .

[0018] This invention discloses an optical thickness measurement system using the aforementioned optical thickness measurement method. The system includes a light source module, an optical path module, a detection module, and a data acquisition and processing module. The light source module includes a broadband light source and a narrowband tunable laser. The optical path module includes an optical fiber coupler and a wavelength division multiplexer. The optical fiber coupler splits the light beam from the light source module, and the wavelength division multiplexer combines the light emitted by the broadband light source and the narrowband tunable laser to illuminate a reference mirror. After reflection by the reference mirror, the light beam forms an interference optical path with other light beams, generating an interference optical signal. The detection module uses a photodetector to convert the interference optical signal into a corresponding electrical signal. The data acquisition and processing module uses an FPGA chip to acquire and process the electrical signal, calculate, and output the optical thickness.

[0019] Preferably, the measurement system of the present invention further includes a dynamic weight fusion module, which is equipped with a signal quality evaluation unit. The signal quality evaluation unit calculates the signal-to-noise ratio of the broadband interferometric electrical signal and the narrowband interferometric electrical signal and uses optical measurement values ​​acquired at different times. The degree of vibration imparts optical measurements to the coarse positioning light source. Optical measurement values ​​of precision light source Different weights are used to perform heterogeneous weighted calculations to obtain the final optical thickness measurement value.

[0020] Beneficial Effects: The dynamic optical thickness measurement method and system based on multi-band guidance disclosed in this invention have the following beneficial effects:

[0021] 1. This invention uses a broadband light source for coarse positioning and a narrowband tunable laser to achieve precise measurement, which can solve the problem of inaccurate positioning in a single band in the prior art; the final optical thickness measurement accuracy is better than ±50nm; at the same time, the measurement method can be applied to the measurement of various structures such as lens center thickness, multi-mirror spacing, and optical thin films; the system structure is expandable and adaptable to different bands and measurement objects.

[0022] 2. Based on the signal-to-noise ratio of the signal and the degree of vibration of the measurement data, this invention performs heterogeneous weighted calculation on the optical measurement values ​​of the coarse positioning light source and the fine measurement light source, which improves the robustness to environmental vibration and low signal-to-noise ratio conditions, reduces the influence of the external environment on the accuracy of optical thickness measurement, and improves the accuracy of the results.

[0023] 3. In the coarse positioning process, this invention uses third-order polynomial fitting to determine the accurate envelope peak position, thereby improving the positioning accuracy. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0025] Figure 1 This is a block diagram of the measurement system structure according to an embodiment of the present invention; Detailed Implementation

[0026] Reference will now be made in detail to embodiments of the present disclosure, one or more of which are set forth herein. Each embodiment and example is provided by way of explanation of the apparatus, composition, and materials of the present disclosure, and not by way of limitation. Rather, the following description provides convenient illustrations for implementing exemplary embodiments of the present disclosure. Indeed, it will be apparent to those skilled in the art that various modifications and variations can be made to the teachings of the present disclosure without departing from the scope or spirit of the present disclosure.

[0027] This invention discloses a dynamic optical thickness measurement method based on multi-band guidance, comprising the following steps:

[0028] S01. A broadband light source is used as the coarse positioning light source, and the light wave emitted by a narrowband tunable laser is used as the fine measurement light source. An interference optical path is constructed using the coarse positioning light source, the fine measurement light source, the sample to be tested, a reference mirror, and a reflector. After the coarse positioning light source is split, it illuminates the sample to be tested and the reference mirror respectively. After the fine measurement light source is split, it illuminates the reference mirror and the reflector respectively. The returned light rays interfere with each other to form interference light of corresponding wavelength bands.

[0029] In an embodiment of the present invention, the coarse positioning light source uses an SLD1310 broadband light source to form a wide envelope signal to achieve a wide range of optical thickness search and preliminary positioning. The narrowband tunable laser has a wavelength of 1550nm and uses high coherence interference to achieve precise nanometer-level thickness measurement.

[0030] S02. The returned interference light is split into beams according to wavelength. The broadband interference light signal and the narrowband interference light signal after beam splitting are converted into electrical signals respectively. The corresponding electrical signals are processed and analyzed to obtain the required data for accurate measurement of optical thickness.

[0031] S03. Use the broadband interference electrical signal corresponding to the broadband interference optical signal to roughly locate the optical thickness measurement range and confirm the optical thickness measurement range.

[0032] When using broadband interferometric electrical signals for coarse localization, the peak region is first coarsely located using the broadband interferometric electrical signals, and then the envelope signal is extracted using Hilbert transform. When using the SLD1310 broadband light source, the acquired broadband interferometric electrical signals I... 1310 Perform Hilbert transform on (z) to calculate the signal envelope E1(z): A sliding window mean filter is applied to the envelope signal E1(z) to eliminate high-frequency noise. A third-order polynomial is used to fit the envelope peak region. The position of the envelope peak is determined by solving for the points where the first derivative of the fitting function is zero. Since the envelope peak is time-series data, the optical measurement value of the coarse positioning light source can be obtained by distance conversion based on the position of the envelope peak and the magnetic grating pulse read by ADC conversion. It also determines the accurate position of the envelope peak, thereby enabling coarse positioning of the optical thickness measurement range.

[0033] In this invention, the window size is 11-21 points during mean filtering. After filtering, the maximum value of the envelope is found, and the 50 data points before and after the maximum value are taken as the envelope peak region. The least squares method is used to perform a third-order polynomial. Fitting; by solving the first derivative of the fitting function The root is found, and the root within the envelope peak region is selected to obtain the accurate position of the envelope peak. Compared with the direct peak finding algorithm, the positioning accuracy of this invention is better than ±0.3μm, providing a more accurate search range for the fine measurement stage and improving measurement efficiency.

[0034] S04. Based on the confirmed optical thickness measurement range, use the narrowband interference electrical signal corresponding to the narrowband interference optical signal to perform distance conversion, obtain an accurate optical path difference measurement value, and use the optical path difference measurement value as the optical thickness value.

[0035] When performing distance conversion using narrowband interferometric electrical signals, this invention performs phase demodulation on the narrowband interferometric electrical signals. The sampling interval containing the zero-crossing point of the signal is located by the sign change of the signal. Within the sampling interval, a three-point interpolation method is used to determine the position of the zero-crossing point and the corresponding phase. where n is the refractive index. The zero-crossing point represents the abscissa value. The zero-crossing point position determines the zero-phase point of the sine wave of the acquired narrowband interferometric signal. After acquiring both broadband and narrowband interferometric signals, the coarse positioning data may be non-integer periodic data extracted from the narrowband interferometric signal. The zero-crossing point allows for further subdivision of the non-integer periodic phase, improving the accuracy of thickness measurement. Simultaneously, Xu Ao performs an FFT transform on the narrowband interferometric signal and obtains the dominant frequency f by peak finding. peak The peak-finding method uses a sub-pixel peak-finding algorithm (Gaussian fitting method) to obtain accurate frequencies, and then calculates the absolute estimate of the optical path difference. , Provides lower precision absolute distance, through The interference order m can be determined; the optical measurement value of the precision light source is obtained by combining the zero-crossing position and the interference order. , The wavelength of the precision measurement light source is indicated. In this embodiment of the invention, the wavelength is 1550nm. However, other narrowband lights can also be selected for optical thickness measurement as needed.

[0036] The process of determining the zero-crossing point position within the sampling interval using the three-point interpolation method in this invention includes: sequentially selecting three sampling points (z0, z ...) within the sampling interval. k-1 ,I k-1 ),(z k ,I k ),(z k+1 ,I k+1 The zero-crossing position is at z k-1 -z k+1 Within the range; by fitting a quadratic curve I(z) through three sampling points and solving for the roots of I(z)=0, the position of the zero-crossing point is obtained. .

[0037] Meanwhile, to further improve the accuracy of the detection results and avoid the results being significantly affected by external environmental interference, in step S04 of this invention, different weights are assigned to the optical measurement values ​​of the coarse positioning light source and the fine measurement light source based on the signal-to-noise ratio of the broadband interference signal and the narrowband interference signal, as well as the vibration degree of the narrowband interference signal, to perform heterogeneous weighting. The final weighted result is used as the actual optical thickness measurement value. ,in =1, The calculated weights are the optical measurements corresponding to the coarse positioning light source. The calculation weights are the optical measurement values ​​corresponding to the precision light source.

[0038] Weights can be assigned using a pre-defined weight lookup table. After acquiring broadband and narrowband interference signals, signal power and noise power are calculated within the peak region to obtain the signal-to-noise ratios (SNR1 and SNR2) for the broadband and narrowband interference signals, respectively. The vibration level of the narrowband interference signal is calculated using the standard deviation of values ​​from multiple consecutive measurement time points. This standard deviation is used as an indicator of the vibration level, Jitter. For example, Jitter > J... th or SNR2 <SNR th Then w fine =0.6. If Jitter <J th And SNR2>SNR th Then w fine =0.95. J th For the set vibration threshold, SNR th To set the signal-to-noise ratio (SNR) threshold, a weight lookup table can be preset based on the required detection accuracy. For different operating conditions, weights can be interpolated to select the appropriate weight value under each condition, ensuring the accuracy of the final result. Alternatively, optimization can be based on machine learning algorithms. For example, a pre-trained lightweight neural network model can be embedded in the FPGA, taking SNR1, SNR2, and jitter as inputs, and directly outputting the optimal fusion weight w. fine This can further enhance the system's adaptability.

[0039] This invention increases the signal-to-noise ratio (SNR) of narrowband interference electrical signals when the vibration of the narrowband interference electrical signal increases drastically or the SNR² of the narrowband interference electrical signal pair is lower than a threshold. Conversely, if the vibration of the narrowband interference signal decreases or the signal-to-noise ratio (SNR) of the narrowband interference signal pair is higher than the threshold, then the amplitude increases. .

[0040] like Figure 1As shown, this invention discloses an optical thickness measurement system using the aforementioned optical thickness measurement method. The system includes a light source module, an optical path module, a detection module, and a data acquisition and processing module. The light source module includes a broadband light source and a narrowband tunable laser. The broadband light source is an SLD1310 light source, and the narrowband tunable laser emits light waves with a wavelength of 1550nm as the 1550nm light source. The optical path module includes an optical fiber coupler and a wavelength division multiplexer (WDM). The optical fiber coupler splits the light beam from the light source module, and the WDM combines the light emitted by the broadband light source and the narrowband tunable laser to illuminate a reference mirror. After reflection by the reference mirror, the light beam forms an interference optical path with other beams, generating an interference optical signal. The detection module uses a photodetector to convert the interference optical signal into a corresponding electrical signal. The data acquisition and processing module uses an FPGA chip to acquire and process the electrical signal, calculate, and output the optical thickness.

[0041] like Figure 1 As shown, the SLD1310 light source is split into two beams, which are respectively connected to the sample under test and the reference mirror of the reference arm. The returned interference light is received by a photodetector and converted into a broadband interference electrical signal. The measurement light from the 1550 light source is also split into two beams by a fiber coupler, which are respectively connected to a fiber optic mirror and a beam from the SLD1310 light source combined by a WDM and then connected to the reference mirror of the reference arm. The returned interference light is received by a photodetector and converted into a narrowband interference electrical signal.

[0042] The FPGA core board for the FPGA chip can use a Xilinx Kintex-7 series KC705 evaluation board or a self-made board with equivalent performance as the core processing unit; the ADC (sampling) module uses a high-speed, high-resolution ADC chip with a maximum sampling rate of 250MSPS. The analog signals output from the two photodetectors are passed through an anti-aliasing filter and then connected to the two channels of the ADC for synchronous sampling, acquiring the corresponding electrical signals for analysis and processing.

[0043] The measurement system of this invention also includes a dynamic weight fusion module, which is equipped with a signal quality evaluation unit. The signal quality evaluation unit calculates the signal-to-noise ratio of the broadband interferometric electrical signal and the narrowband interferometric electrical signal, and uses the optical measurement values ​​of the precision measurement light source acquired at different times. The degree of vibration imparts optical measurements to the coarse positioning light source. Optical measurement values ​​of precision light source Different weights are used to perform heterogeneous weighted calculations to obtain the final optical thickness measurement value.

[0044] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A dynamic optical thickness measurement method based on multi-band guidance, characterized in that, Including the following steps: S01. A broadband light source is used as the coarse positioning light source, and the light wave emitted by a narrowband tunable laser is used as the fine measurement light source. An interference optical path is constructed using the coarse positioning light source, the fine measurement light source, the sample to be tested, a reference mirror, and a reflector. After the coarse positioning light source is split, it illuminates the sample to be tested and the reference mirror respectively. After the fine measurement light source is split, it illuminates the reference mirror and the reflector respectively. The returned light rays interfere with each other to form interference light of corresponding wavelength bands. S02. The returned interference light is split into beams according to wavelength, and the broadband interference light signal and narrowband interference light signal after beam splitting are converted into electrical signals respectively; S03. Use the broadband interference electrical signal corresponding to the broadband interference optical signal to roughly locate the optical thickness measurement range and confirm the optical thickness measurement range; S04. Based on the confirmed optical thickness measurement range, use the narrowband interference electrical signal corresponding to the narrowband interference optical signal to perform distance conversion, obtain an accurate optical path difference measurement value, and use the optical path difference measurement value as the optical thickness value. When using broadband interferometric electrical signals for coarse localization, the peak region is first coarsely located using broadband interferometric electrical signals, and then the envelope signal is extracted using Hilbert transform to determine the accurate position of the envelope peak, thereby achieving coarse localization of the optical thickness measurement range. When determining the envelope peak position, the acquired broadband interferometric electrical signal I 1310 (z) Perform Hilbert transform to obtain the analytic signal H[I 1310 [(z)], calculate the signal envelope E1(z): ; A sliding window mean filter is applied to the envelope signal E1(z). A third-order polynomial is used to fit the envelope peak region. The location of the envelope peak is determined by solving for the points where the first derivative of the fitting function is zero. Based on the location of the envelope peak, the optical measurement values ​​of the coarsely located light source are obtained through distance conversion. ; When performing distance conversion using narrowband interferometric signals, the sampling interval containing the zero-crossing point of the signal is located by observing the sign change of the signal. Within the sampling interval, the position of the zero-crossing point is determined using three-point interpolation, and the phase corresponding to the zero-crossing point is also determined. where n is the refractive index. The x-coordinate value corresponds to the zero-crossing point; an FFT transform is performed on the narrowband interference electrical signal, and the dominant frequency f is obtained by peak finding. peak Then, the absolute estimate of the optical path difference can be calculated. Determine the interference order m; precisely measure the optical values ​​of the light source. , Indicates the wavelength of the precision measurement light source; In step S04, the optical measurement values ​​of the coarse positioning light source are assigned based on the signal-to-noise ratio of the broadband interference signal and the narrowband interference signal, as well as the vibration degree of the narrowband interference signal. Optical measurement values ​​of precision light source Different weights are used for heterogeneous weighting, and the final weighted result is used as the actual optical thickness measurement value; Actual optical thickness measurement value ,in =1, The calculated weights are the optical measurements corresponding to the coarse positioning light source. The calculation weights are the optical measurement values ​​corresponding to the precision light source.

2. The dynamic optical thickness measurement method based on multi-band guidance according to claim 1, characterized in that, In mean filtering, the window size is 11-21 points. After filtering, the maximum value of the envelope is found, and the 50 data points before and after the maximum value are taken as the envelope peak region. The least squares method is used to perform a third-order polynomial. Fitting; by solving the first derivative of the fitting function The root is found, and the root within the envelope peak region is selected to obtain the accurate location of the envelope peak.

3. The dynamic optical thickness measurement method based on multi-band guidance according to claim 1, characterized in that, The process of determining the zero-crossing point position using three-point interpolation within the sampling interval includes: sequentially selecting three sampling points (z0, z ...) within the sampling interval. k-1 I k-1 ),(z k I k ),(z k+1 I k+1 By fitting a quadratic curve I(z) to three sampling points and solving for the roots of I(z)=0, the position of the zero-crossing point can be obtained. .

4. The dynamic optical thickness measurement method based on multi-band guidance according to claim 1, characterized in that, When the vibration of the narrowband interference signal increases drastically or the signal-to-noise ratio (SNR)² of the narrowband interference signal falls below a threshold, increase... Conversely, it increases. .

5. An optical thickness measurement system, using the optical thickness measurement method according to any one of claims 1-4, characterized in that, The system includes a light source module, an optical path module, a detection module, and a data acquisition and processing module. The light source module includes a broadband light source and a narrowband tunable laser. The optical path module includes an optical fiber coupler and a wavelength division multiplexer. The optical fiber coupler splits the light beam from the light source module, and the wavelength division multiplexer combines the light beams emitted by the broadband light source and the narrowband tunable laser and directs them onto a reference mirror. After being reflected by the reference mirror, the beams combine with other beams to form an interference optical path, generating an interference optical signal. The detection module uses a photodetector to convert the interference optical signal into a corresponding electrical signal. The data acquisition and processing module uses an FPGA chip to acquire and process electrical signals, calculate and output optical thickness.

6. The optical thickness measurement system according to claim 5, characterized in that, The measurement system also includes a dynamic weight fusion module, which is equipped with a signal quality evaluation unit. This unit calculates the signal-to-noise ratio of the broadband and narrowband interferometric electrical signals and uses optical measurements from a precision measurement light source acquired at different times. The degree of vibration imparts optical measurements to the coarse positioning light source. Optical measurement values ​​of precision light source Different weights are used to perform heterogeneous weighted calculations to obtain the final optical thickness measurement value.

Images