A bimodal spectral peak wavelength positioning method, electronic equipment and storage medium

Abstract

This application discloses a method for locating the peak wavelength of a bimodal spectrum, an electronic device, and a storage medium, relating to the field of spectral confocal systems. The method includes: S1, performing a fast Fourier transform on the acquired spectral signal to obtain the amplitude spectrum and phase spectrum; S2, performing phase unwrapping processing on the phase spectrum to obtain continuous phase values; S3, based on the spectral characteristics of the amplitude spectrum, determining the periodic minimum points in the spectrum using a peak-finding algorithm, and calculating a first parameter representing the position difference between the two peaks; S4, based on the unwrapped phase spectrum, calculating a second parameter representing the center position of the two peaks; S5, calculating the precise wavelength positions of the two peaks simultaneously using the first and second parameters; S6, substituting the precise wavelength positions of the two peaks into a pre-stored wavelength-displacement calibration curve to obtain the displacement or thickness value of the measured object. The method improves resolution by extracting the joint features of the amplitude and phase spectra through frequency domain analysis.

Description

Technical Field

[0001] This application relates to the field of spectral confocal system technology, and in particular to a method for locating the peak wavelength of a dual-peak spectrum, an electronic device, and a storage medium. Background Technology

[0002] Spectral confocal sensors are a high-precision measurement technology widely used in industrial inspection, scientific research, and other fields. A spectral confocal sensor consists of modules such as a polychromatic light source, a spectrometer, a dispersive lens, and a confocal aperture. The polychromatic beam emitted from the point light source produces axial chromatic aberration after passing through the dispersive lens group. The dispersive objective focuses light of different wavelengths at different positions on the same optical axis. When the surface to be measured is placed within the focusing range of the axial polychromatic beam, only one specific wavelength of light is focused on the surface. Due to the effect of the confocal aperture, only this specific wavelength of light returns entirely to the spectrometer, while only a very small portion of the unfocused light returns, thus presenting a single-peak spectral signal on the spectrometer. When the height of the surface to be measured changes, the wavelength of the light focused on the surface also changes accordingly.

[0003] Traditional methods for extracting signal peaks include parabolic fitting, extremum methods, Gaussian fitting, and centroid methods. These fitting methods are mostly based on the shape characteristics of the signal. However, actual spectral signals are not perfectly Gaussian or sinc² distributions due to factors such as spectrometer grating diffraction, fiber coupling, and dispersive objectives, and they do not meet symmetry requirements. Furthermore, when multiple peaks are similarly shaped and overlapping (e.g., when measuring transparent glass), the resolution further decreases. Therefore, shape-based peak localization methods have significant limitations. Deconvolution can reduce peak width to increase peak resolution, but this method requires the system's point spread function, which is often difficult to obtain in practice. Blind deconvolution does not rely on a fixed point spread function, but adjusting the parameters is complex and unstable.

[0004] When a spectral confocal sensor detects transparent objects (such as thin films and glass), the resulting spectral signal often exhibits two distinct peaks. Typically, when the two peaks are far apart, the peak positions can be determined using traditional methods, but the resolution limit is low. Summary of the Invention

[0005] To address the aforementioned issues, this application provides a method for locating the peak wavelength of a bimodal spectrum, which solves the problems of low resolution and dependence on signal shape in traditional time-domain analysis methods when bimodal peaks overlap.

[0006] The first technical solution adopted in this application is: providing a method for locating the peak wavelength of a bimodal spectrum, including the following steps:

[0007] S1. Perform a fast Fourier transform on the acquired spectral signal to obtain the amplitude spectrum and phase spectrum of the spectral signal;

[0008] S2. Perform phase unwinding processing on the phase spectrum to obtain continuous phase values;

[0009] S3. Based on the spectral characteristics of the amplitude spectrum, the periodic minimum points in the spectrum are determined by the peak-finding algorithm, and the first parameter representing the difference in the positions of the two peaks is calculated.

[0010] S4. Based on the unwound phase spectrum, a second parameter representing the position of the bimodal center is calculated by weighted linear fitting and compensation correction.

[0011] S5. Based on the first parameter and the second parameter, calculate the precise wavelength positions of the two peaks simultaneously;

[0012] S6. Substitute the precise wavelength positions of the two peaks into the pre-stored wavelength-displacement calibration curve to obtain the displacement or thickness value of the object being measured.

[0013] In an optional embodiment, step S3, calculating the first parameter based on the spectral characteristics of the amplitude spectrum, includes:

[0014] Select the low-frequency band in the spectrum;

[0015] Differentiate the spectrum signal in the low-frequency band and locate the poles where the derivative is zero;

[0016] Calculate the second derivative at the pole to determine the minimum point;

[0017] The weight matrix is ​​set based on the coordinates of the local minimum point and the signal envelope information obtained through Hilbert transform.

[0018] The minimum point is linearly fitted using the weighted least squares method to obtain the fitting slope.

[0019] The bimodal position difference is calculated based on the fitted slope.

[0020] In an optional embodiment, the frequency position of the minimum point is inversely proportional to the difference in the positions of the two peaks.

[0021] In an optional embodiment, step S4, which calculates the second parameter based on the unwound phase spectrum, includes:

[0022] Select a continuous phase segment near zero frequency in the low-frequency band;

[0023] The continuous phases are linearly fitted using the least squares method, and weighted using a weight matrix to obtain the initial slope;

[0024] The initial slope is compensated and corrected by a compensation parameter to obtain the compensated slope.

[0025] The position of the center of the double peaks is calculated based on the compensated slope.

[0026] In an optional embodiment, the method for determining the compensation parameter includes:

[0027] When the distance between the two peak signals is greater than a preset threshold, the extreme positions of the two peaks are located by Gaussian fitting or centroid method.

[0028] The intensity ratio parameter of the two peaks is calculated based on the extreme value locations.

[0029] In an optional embodiment, compensating for the initial slope using compensation parameters includes:

[0030] A first-order linear approximation is performed on the nonlinear terms in the phase spectrum of the dual-pulse signal, that is, a Taylor expansion is performed near the zero frequency and higher-order nonlinear terms are ignored, and the initial slope obtained by fitting is compensated using the intensity ratio parameter.

[0031] In an optional embodiment, in step S2, the unwound phase spectrum is filtered to remove high-frequency noise;

[0032] The effective frequency band used for filtering is determined by the standard deviation of the spectrum.

[0033] In an optional embodiment, in step S5, the precise wavelength positions of the two peaks are calculated simultaneously in the following manner:

[0034] Subtracting half of the first parameter from the second parameter yields the wavelength position of one peak, and adding half of the first parameter to the second parameter yields the wavelength position of another peak.

[0035] The second technical solution adopted in this application is: providing an electronic device, the electronic device comprising: a memory and a processor coupled to each other, the processor being used to execute program instructions stored in the memory to implement the bi-peak spectral peak wavelength localization method of any of the preceding claims.

[0036] The third technical solution adopted in this application is: providing a computer-readable storage medium that stores program data, the program data being executable by a processor to implement the bi-peak spectral peak wavelength localization method of any of the preceding claims.

[0037] Due to the adoption of the above technical solution, this application has at least one of the following beneficial effects compared with the prior art:

[0038] 1. This invention breaks through the dependence of traditional time-domain analysis methods on signal shape. It extracts the joint features of amplitude spectrum and phase spectrum through frequency domain analysis, which can improve resolution even in the case of severe double-peak overlap.

[0039] 2. It is not dependent on the specific shape of the signal (such as Gaussian distribution, symmetry, etc.), and has stronger adaptability and generalization ability to different optical systems and measurement environments.

[0040] 3. It eliminates the need to accurately obtain the system's point spread function beforehand, thus avoiding complex calibration processes and the instability of blind deconvolution. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0042] Figure 1 A flowchart illustrating a method for locating the peak wavelength of a bimodal spectrum provided in an embodiment of this application;

[0043] Figure 2 This is a schematic diagram of the structure of a computer device according to an embodiment of this application;

[0044] Figure 3 This is a schematic diagram of the structure of an embodiment of the computer-readable storage medium of this application. Detailed Implementation

[0045] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It is understood that the specific embodiments described herein are only for explaining this application and not for limiting it. Furthermore, it should be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all structures. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0046] The terms "first," "second," etc., used in this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0047] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0048] In existing technologies, the resolution of bimodal spectral signals generated by spectral confocal sensors when measuring transparent materials often decreases due to peak aliasing, non-ideal distribution, and noise interference. Traditional peak localization methods based on temporal shape features (such as Gaussian fitting and centroid method) are highly dependent on signal shape and are difficult to accurately distinguish when the distance between the two peaks is close, resulting in reduced measurement accuracy. Furthermore, methods to improve resolution, such as deconvolution, require knowledge of the system's point spread function or complex parameter adjustments, limiting their practicality. In contrast, this application converts the signal to the frequency domain, uses the periodic minimum of the amplitude spectrum to determine the position difference between the two peaks, and combines weighted linear fitting and compensation correction of the phase spectrum to obtain the center position of the two peaks. Solving these simultaneous solutions achieves high resolution and high-precision localization, effectively overcoming the dependence on signal shape. Even in cases of severe bimodal aliasing, the peak wavelength can still be accurately extracted, significantly improving measurement reliability.

[0049] like Figure 1 As shown, Figure 1 A flowchart illustrating a bimodal spectral peak wavelength localization method provided in an embodiment of this application includes the following steps:

[0050] S1. Perform a Fast Fourier Transform (FFT) on the acquired spectral signal to obtain the amplitude and phase spectra; that is, use a spectral confocal sensor to acquire the spectral signal reflected by the object under test (such as transparent glass); perform a FFT on the acquired time-domain spectral signal to obtain its frequency domain representation. And calculate its amplitude spectrum. .

[0051] S2. Perform phase unwinding on the phase spectrum to obtain continuous phase values; then filter the unwound phase spectrum to remove high-frequency noise.

[0052] The effective frequency band used for filtering is determined by the standard deviation of the spectrum.

[0053] Since the effective information and main energy of a signal are usually concentrated in the low-frequency range, while the high-frequency range is often dominated by noise, the phase spectrum after unwrapping needs to be filtered. The effective frequency band is... ,in denoted as the standard deviation of the spectrum.

[0054] S3. Based on the spectral characteristics of the amplitude spectrum, the periodic minimum points in the spectrum are determined by a peak-finding algorithm, and the first parameter representing the difference in the positions of the two peaks is calculated; including the following steps:

[0055] Select the low-frequency band in the spectrum; differentiate the spectral signal in the low-frequency band and locate the poles where the derivative is zero; that is, for the amplitude spectrum of the selected low-frequency band... The derivative is calculated by taking the first derivative. The peak-finding algorithm is used to locate all points where the first derivative is zero. These points are the extreme points of the amplitude spectrum, including maximum and minimum points.

[0056] The second derivative is calculated at the poles to determine the local minimum. If the second derivative is greater than 0, the point is a local minimum. All local minimums that meet the conditions are selected to form a candidate set. Periodically occurring local minimums are spectral beat nodes caused by interference of double-peaked signals, and their position information represents the distance between the two peaks. The frequency position of the local minimum is inversely proportional to the difference in position between the two peaks. The theoretical basis for the inverse relationship between the frequency position of the local minimum and the difference in position between the two peaks is described in detail below:

[0057] In the frequency domain, the spectral signal is modeled as the convolution of two ideal pulse signals and the point spread function (PSF) of the optical system; that is, the spectral signal can be viewed as the convolution of a unit pulse signal and the PSF signal, and the bimodal signal can be considered as two signals with different intensities. Different locations The convolution of the pulse signal and the PSF signal results in the following expression for the dual-pulse signal:

[0058] ,in The Dirac function is expressed as follows:

[0059] ,in This represents the difference in position between the two peaks; the amplitude spectrum exhibits a modulated oscillation, with the minimum point occurring when the cosine function is -1, i.e., satisfying... ,in, Let k be the frequency at the minimum point, k = 0, 1, 2, ... Rearranging the above equation, we can obtain an explicit expression for the frequency at the minimum point. This clearly reveals the core principle that the frequency of the minimum point is inversely proportional to the difference in the positions of the two peaks; that is, when the difference in the positions of the two peaks... When the frequency is large (i.e., the two peaks are far apart in the time domain), the minimum points will appear densely in the low-frequency region; conversely, when the frequency is small... When the value is relatively small (i.e., the two peaks overlap), the minimum point will spread to the high-frequency region, and the distance between adjacent minimum points will increase.

[0060] To improve fitting accuracy, different weights need to be assigned to different minimum points. The weight matrix is ​​set based on the coordinates of the minimum points and the signal envelope information obtained through Hilbert transform, so as to reflect the relative intensity of the two peaks.

[0061] The envelope of the amplitude spectrum is extracted using Hilbert transform; the envelope reflects the overall attenuation trend of the spectrum. For each minimum point, its weight can be set to the square of the envelope value at that point or its reciprocal, to reduce the impact of minimum points in the high-frequency region with low signal-to-noise ratio on the fitting results, thereby improving the robustness of the algorithm.

[0062] The minimum point is linearly fitted using the weighted least squares method to obtain the fitting slope; the difference in bimodal positions is calculated based on the fitting slope; the weighted least squares method is then used to... A linear fit is performed, and the slope of the fitted line is K; according to the explicit expression for the frequency of the minimum point, the slope of the line is... ,Right now It can be obtained via 1 / K.

[0063] Traditional time-domain fitting methods suffer from a sharp drop in resolution or even fail when there is severe bimodal overlap and the spacing is smaller than the width of the system's point spread function. Inversion is achieved by analyzing the distribution of frequency-domain minima. Its resolution is not directly limited by the width of the time-domain peak. Even in cases where the two peaks severely overlap and appear to be a single peak in the time domain, as long as at least one valid local minimum can be detected in the frequency domain, the resolution can be calculated. .

[0064] However, based solely on location differences This is insufficient to determine the absolute positions of the two pulses, and the solution... and Another independent constraint is required, namely the center value of the two pulse positions.

[0065] S4. Based on the unwound phase spectrum, the second parameter representing the position of the bimodal center is calculated by weighted linear fitting combined with compensation correction, including the following steps:

[0066] A continuous phase segment near zero frequency in the low-frequency band is selected; the continuous phase is linearly fitted using the least squares method, and weighted using a weight matrix to obtain the initial slope; to improve the fitting quality, a weight matrix is ​​introduced; this weight can be consistent with the weight used in amplitude spectrum analysis to give higher weight to data points in the high signal-to-noise ratio band.

[0067] The initial phase slope was obtained by fitting using the weighted least squares method. When the two pulse intensities are equal, this slope directly corresponds to the center position of the double peaks. .

[0068] In real-world systems, the intensities of the two peaks are often not equal. This introduces a nonlinear term into the phase spectrum, resulting in an initial slope There is a deviation. Therefore, we introduce a compensation parameter for correction; the initial slope is corrected using this compensation parameter to obtain the compensated slope; the method for determining the compensation parameter includes:

[0069] When the distance between the two peaks is greater than a preset threshold, the extreme positions of the two peaks are located using Gaussian fitting or centroid method; the intensity ratio parameter of the two peaks is calculated based on the extreme positions. During system initialization or when the distance between the two peaks is sufficiently large to be reliably distinguishable by traditional methods (such as Gaussian fitting), the initial positions and intensity estimates of the two peaks are obtained. and Subsequently, the strength ratio parameter was calculated. The compensation parameter m is derived from this strength ratio parameter, and its specific form can be: Or a related function to characterize the extent to which the bimodal intensity asymmetry affects phase linearity.

[0070] A first-order linear approximation is performed on the nonlinear terms in the phase spectrum of the dual-pulse signal, that is, a Taylor expansion is performed near zero frequency and higher-order nonlinear terms are ignored, and the initial slope obtained by fitting is compensated using the intensity ratio parameter.

[0071] Using the determined compensation parameter m, the initial slope The correction is then made. The compensation formula is based on a first-order Taylor expansion of the nonlinear term of the phase function, which in this embodiment is:

[0072] ,in, This is the slope value after compensation.

[0073] The location of the bimodal center is calculated based on the compensated slope; that is, the location of the bimodal center. .

[0074] By introducing a compensation parameter directly related to the intensity ratio, this nonlinear bias is quantitatively corrected, ensuring that the measurement results of the bimodal center position are accurate even when... and It maintains extremely high accuracy even when there are significant differences, solving the positioning problem under asymmetric signals.

[0075] S5. Based on the first and second parameters, the precise wavelength positions of the two peaks are calculated simultaneously. The precise wavelength positions of the two peaks are calculated simultaneously in the following way:

[0076] Subtracting half of the first parameter from the second parameter yields the wavelength position of one peak, and adding half of the first parameter to the second parameter yields the wavelength position of the other peak.

[0077] Right now The following is an explanation using a specific calculation example:

[0078] Twin peaks center position =600.0nm, difference in position between the two peaks =4.0nm; then, .

[0079] S6. Substitute the precise wavelength positions of the two peaks into the pre-stored wavelength-displacement calibration curve to obtain the displacement or thickness value of the object being measured.

[0080] Regarding the above embodiments, this application provides a computer device; please refer to [link / reference]. Figure 2 , Figure 2 This is a schematic diagram of the structure of a computer device according to an embodiment of the present application. The computer device includes a memory and a processor, wherein the memory and the processor are coupled to each other. The memory stores program data, and the processor executes the program data to implement the steps of any embodiment of the bimodal spectral peak wavelength localization method described above.

[0081] In this embodiment, the processor may also be referred to as a CPU (Central Processing Unit). The processor may be an integrated circuit chip with signal processing capabilities. The processor may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. A general-purpose processor may be a microprocessor or any conventional processor.

[0082] The methods described in the above embodiments can be implemented as computer programs; therefore, this application proposes a computer-readable storage medium. Please refer to [link to relevant documentation]. Figure 3 , Figure 3 This is a schematic diagram of the structure of an embodiment of the computer-readable storage medium of this application. The computer-readable storage medium stores program data that can be executed by a processor to implement the steps of any embodiment of the bimodal spectral peak wavelength localization method described above.

[0083] In this embodiment, the computer-readable storage medium can be a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, or a medium capable of storing program data. Alternatively, it can be a server storing the program data, which can send the stored program data to other devices for execution or run the stored program data itself.

[0084] In the several embodiments provided in this application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0085] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0086] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0087] The above description is merely an embodiment of this application and does not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A bimodal spectral peak wavelength locating method, characterized in that, Includes the following steps: S1. Perform a fast Fourier transform on the acquired spectral signal to obtain the amplitude spectrum and phase spectrum of the spectral signal; S2. Perform phase unwinding processing on the phase spectrum to obtain continuous phase values; S3. Based on the spectral characteristics of the amplitude spectrum, the periodic minimum points in the spectrum are determined by the peak-finding algorithm, and the first parameter representing the difference in the positions of the two peaks is calculated. S4. Based on the unwound phase spectrum, a second parameter representing the position of the bimodal center is calculated by weighted linear fitting and compensation correction. S5. Based on the first parameter and the second parameter, calculate the precise wavelength positions of the two peaks simultaneously; S6. Substitute the precise wavelength positions of the two peaks into the pre-stored wavelength-displacement calibration curve to obtain the displacement or thickness value of the object being measured.

2. The bimodal spectral peak wavelength locating method of claim 1, wherein, The calculation of the first parameter based on the spectral characteristics of the amplitude spectrum in step S3 includes: Select the low-frequency band in the spectrum; Differentiate the spectrum signal in the low-frequency band and locate the poles where the derivative is zero; Calculate the second derivative at the pole to determine the minimum point; The weight matrix is ​​set based on the coordinates of the local minimum point and the signal envelope information obtained through Hilbert transform. The minimum point is linearly fitted using the weighted least squares method to obtain the fitting slope. The bimodal position difference is calculated based on the fitted slope.

3. The bimodal spectral peak wavelength locating method of claim 2, wherein, The frequency position of the minimum point is inversely proportional to the difference in the positions of the two peaks.

4. The bimodal spectral peak wavelength locating method of claim 1, wherein, The calculation of the second parameter based on the unwound phase spectrum in step S4 includes: Select a continuous phase segment near zero frequency in the low-frequency band; The continuous phases are linearly fitted using the least squares method, and weighted using a weight matrix to obtain the initial slope; The initial slope is compensated and corrected by a compensation parameter to obtain the compensated slope. The position of the center of the double peaks is calculated based on the compensated slope.

5. The bimodal spectral peak wavelength locating method of claim 4, wherein, The method for determining the compensation parameters includes: When the distance between the two peak signals is greater than a preset threshold, the extreme positions of the two peaks are located by Gaussian fitting or centroid method. The intensity ratio parameter of the two peaks is calculated based on the extreme value locations.

6. The bimodal spectral peak wavelength locating method of claim 5, wherein, The initial slope is compensated and corrected using compensation parameters, including: A first-order linear approximation is performed on the nonlinear terms in the phase spectrum of the dual-pulse signal, that is, a Taylor expansion is performed near the zero frequency and higher-order nonlinear terms are ignored, and the initial slope obtained by fitting is compensated using the intensity ratio parameter.

7. The method for locating the peak wavelength of a bimodal spectrum according to claim 1, characterized in that, In step S2, the unwound phase spectrum is filtered to remove high-frequency noise. The effective frequency band used for filtering is determined by the standard deviation of the spectrum.

8. The method for locating the peak wavelength of a bimodal spectrum according to claim 1, characterized in that, In step S5, the precise wavelength positions of the two peaks are calculated simultaneously in the following way: Subtracting half of the first parameter from the second parameter yields the wavelength position of one peak, and adding half of the first parameter to the second parameter yields the wavelength position of another peak.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor coupled to each other, the processor being configured to execute program instructions stored in the memory to implement the bi-peak spectral peak wavelength localization method as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program data that can be executed by a processor to implement the bi-peak spectral peak wavelength localization method as described in any one of claims 1-8.

Images