An optical measurement data analysis method, analysis system and electronic device

Abstract

The application provides an optical measurement data analysis method, comprising: performing spectrum analysis on a plurality of detection samples to obtain a deviation index and a correlation coefficient of each detection sample; setting a spectrum deviation index parameter and a spectrum correlation coefficient parameter, and calculating a spectrum matching degree index of each detection sample, so that the spectrum matching degree index of each detection sample is greater than a preset spectrum matching degree index threshold; based on the deviation index parameter and the correlation coefficient parameter, performing spectrum analysis on a to-be-tested piece to calculate a spectrum matching degree index of the to-be-tested piece, and judging whether the to-be-tested piece is a normal piece. The application firstly includes the deviation index and the correlation coefficient in the spectrum matching degree index, so that the reliability of a measurement result can be evaluated from two aspects of an absolute deviation of a measurement spectrum and a simulation spectrum and a corresponding case of a wave peak and a wave trough, the spectrum matching degree of the to-be-tested piece is judged through the spectrum matching degree index, and the accuracy and the robustness of the spectrum matching degree are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of optical scatterometry, and in particular to an optical measurement data analysis method, an analysis system and an electronic device. BACKGROUND

[0002] The optical scatterometry method, also known as the optical critical dimension (OCD) measurement method, has the advantages of fast speed, low cost, non-contact and non-destructive compared with microscopic morphology measurement means such as scanning electron microscope and atomic force microscope, and thus has been widely used in the field of advanced process online monitoring. The measurement signal obtained by the optical scatterometry technology is only a set of spectral signals about the distribution of incident wavelength or incident angle, such as reflectivity, ellipsometric parameters and Mueller matrix, and a certain data analysis means is needed to extract the sample parameters to be measured from the measurement signal. After the extraction of the parameters to be measured, the reliability of the obtained results needs to be judged by matching the theoretical spectrum and the measured spectrum, and there are two commonly used evaluation indexes: deviation index and correlation coefficient.

[0003] The first index, deviation index (DI), reflects the absolute deviation of the measured spectrum and the theoretical spectrum, and the smaller the absolute deviation, the higher the matching degree of the measured spectrum and the theoretical spectrum. The second index, correlation coefficient, is used to reflect the alignment degree of the peaks and valleys of the measured spectrum and the theoretical spectrum, and the larger the correlation coefficient, the higher the alignment degree of the peaks and valleys of the two.

[0004] However, the deviation index and the correlation coefficient are two independent evaluation standards, and the traditional analysis method of optical scatterometry only uses one of the two indexes to evaluate the reliability of the measurement results. If only the correlation coefficient index is used, the peaks and valleys of the measured spectrum and the theoretical spectrum do not match when the deviation index is different. If only the deviation index is used, the peaks of the measured spectrum and the theoretical spectrum will be misaligned when the correlation coefficient is different. Therefore, the traditional analysis method cannot obtain correct evaluation results and cannot judge whether the measurement results are reliable, resulting in poor robustness of the measurement, and thus the accuracy of the judgment of the semiconductor processing yield cannot be accurately judged. SUMMARY

[0005] The present application provides an optical measurement data analysis method, an analysis system and an electronic device to solve the defects in the prior art, and creatively proposes a new spectral matching degree index, which includes both the deviation index and the correlation coefficient, improves the accuracy and universality of spectral analysis, can judge whether the semiconductor processing process meets the standard based on the abnormality of the measurement results, and greatly improves the accuracy of the judgment of the semiconductor processing yield.

[0006] The application provides an optical measurement data analysis method, comprising:

[0007] spectrum analysis is performed on the multiple detection samples to obtain a spectrum deviation index and a spectrum correlation coefficient of each detection sample;

[0008] spectrum deviation index parameters and spectrum correlation coefficient parameters are set up, a spectrum matching degree index mathematical model is created in combination of the spectrum deviation index, the spectrum deviation index parameters, the spectrum correlation coefficient and the spectrum correlation coefficient parameters, and the spectrum matching degree index value of each detection sample is calculated based on the mathematical model and is greater than a preset spectrum matching degree index threshold value; spectrum analysis is performed on a to-be-tested piece to obtain a spectrum deviation index and a spectrum correlation coefficient of the to-be-tested piece, and the spectrum matching degree index value of the to-be-tested piece is calculated based on the mathematical model; if the spectrum matching degree index value of the to-be-tested piece is greater than the preset spectrum matching degree index threshold value, the to-be-tested piece is a normal piece; otherwise, the to-be-tested piece is an abnormal piece.

[0009] Preferably, the spectrum analysis on the multiple detection samples to obtain a spectrum deviation index and a spectrum correlation coefficient of each detection sample comprises:

[0010] a physical model of the multiple detection samples is established to obtain a theoretical spectrum of the detection samples; and the multiple detection samples are measured to obtain a measured spectrum of the detection samples;

[0011] a spectrum deviation index and a spectrum correlation coefficient of the theoretical spectrum and the measured spectrum of each to-be-tested sample are calculated;

[0012] The type of the theoretical spectrum and the measured spectrum is consistent, and the type of the theoretical spectrum and the measured spectrum comprises one or more combinations of reflectivity, transmissivity, ellipsometric parameters and Mueller matrix signals.

[0013] According to the optical measurement method provided by the application, after the spectrum deviation index and the spectrum correlation coefficient of each detection sample are obtained, the method comprises:

[0014] preset deviation index threshold values and correlation coefficient threshold values, for any one detection sample, if the spectrum deviation index of the detection sample is less than the deviation index threshold value and the spectrum correlation coefficient of the detection sample is greater than the correlation coefficient threshold value, the detection sample and the spectrum deviation index and the spectrum correlation coefficient thereof are retained, otherwise, the detection sample and the spectrum deviation index and the spectrum correlation coefficient thereof are rejected. Further, in the spectrum matching degree index mathematical model, the spectrum matching degree index and the spectrum correlation coefficient are in a monotonically increasing relationship, and the spectrum matching degree index and the spectrum deviation index are in a monotonically decreasing relationship;

[0015] The range of the spectrum matching degree index is [0, 1].

[0016] Optionally, the spectral matching degree index mathematical model comprises:

[0017] GOF=(αρ) βM

[0018] or GOF=sin(αρ)cos(βM);

[0019] wherein GOF is the spectral matching degree index, α is the spectral correlation coefficient parameter, β is the spectral deviation index parameter, ρ is the spectral correlation coefficient, and M is the spectral deviation index.

[0020] Further, the spectral deviation index of the reserved detection sample is normalized to obtain a normalized spectral deviation index, comprising:

[0021] The value of the spectral correlation coefficient is set to 1;

[0022] The value of the spectral matching degree index is set to a spectral matching degree index threshold, so that the following formula is satisfied:

[0023]

[0024] The range of the normalized spectral deviation index is The spectral deviation index of the reserved detection sample is normalized to

[0025] wherein GOF is the spectral matching degree index, GOFt o l is the spectral matching degree index threshold, α0 is the value of the spectral correlation coefficient parameter, β0 is the value of the spectral deviation index parameter, ρ is the spectral correlation coefficient, is the maximum value of the normalized spectral deviation index.

[0026] According to the optical measurement method provided by the application, the spectral deviation index parameter and the spectral correlation coefficient parameter are obtained, comprising:

[0027] The initial value of the spectral deviation index parameter and the initial value of the spectral correlation coefficient parameter are preset, the spectral matching degree index of each detection sample is calculated and compared with the spectral matching degree index threshold;

[0028] If the spectral matching degree index of any one of the detection samples is not greater than the spectral matching degree index threshold, the initial value of the spectral deviation index parameter and the initial value of the spectral correlation coefficient parameter are updated, the spectral matching degree index of each detection sample is recalculated and compared with the spectral matching degree index threshold;

[0029] Until the spectral matching degree index of each of the detection samples is greater than the spectral matching degree index threshold value, output the initial value of the updated spectral deviation index parameter and the initial value of the spectral correlation coefficient parameter as the value of the spectral deviation index parameter and the value of the spectral correlation coefficient parameter respectively.

[0030] Optionally, the spectral deviation index includes a mean square error or a mean square root error or a mean absolute error between the theoretical spectrum and the measured spectrum of the detection sample or the to-be-detected member.

[0031] The spectral correlation coefficient includes a Pearson correlation coefficient or a Spearman correlation coefficient between the theoretical spectrum and the measured spectrum of the detection sample or the to-be-detected member.

[0032] The present application also provides an optical scattering measurement data analysis system, comprising a sample analysis module, a spectral matching degree index mathematical model creation module and an analysis module, wherein:

[0033] The sample analysis module is used for performing spectral analysis on a plurality of detection samples to obtain a spectral deviation index and a spectral correlation coefficient of each of the detection samples.

[0034] The spectral matching degree index mathematical model creation module is used for obtaining a spectral deviation index parameter and a spectral correlation coefficient parameter, creating a spectral matching degree index mathematical model in combination with the spectral deviation index, the spectral deviation index parameter, the spectral correlation coefficient and the spectral correlation coefficient parameter, and calculating a spectral matching degree index of each of the detection samples based on the mathematical model, so that the spectral matching degree index value of each of the detection samples is greater than a preset spectral matching degree index threshold value.

[0035] The analysis module is used for performing spectral analysis on a to-be-detected member to obtain a spectral deviation index and a spectral correlation coefficient of the to-be-detected member, calculating a spectral matching degree index value of the to-be-detected member based on the mathematical model, and if the spectral matching degree index value of the to-be-detected member is greater than the preset spectral matching degree index threshold value, the to-be-detected member is a normal member, otherwise, the to-be-detected member is an abnormal member.

[0036] The present application also provides an electronic device, comprising a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor implements the steps of any of the above methods when executing the program.

[0037] The present invention provides an optical measurement data analysis method, analysis system and electronic equipment. By analyzing the test sample, the spectral deviation index and spectral correlation coefficient of the sample are obtained, and the spectral deviation index parameters and spectral correlation coefficient parameters are defined. The spectral deviation index and spectral correlation coefficient are innovatively incorporated into the spectral matching index. Therefore, the reliability of the measurement results can be evaluated from two aspects: the absolute deviation between the measured spectrum and the simulated spectrum and the correspondence between the peaks and troughs. The spectral matching degree of the test piece is judged by the spectral matching degree index, thereby improving the accuracy and robustness of the spectral matching judgment, thereby improving the accuracy of the semiconductor processing yield judgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0039] Figure 1 1 is a flow chart of the optical measurement data analysis method provided by the present invention;

[0040] Figure 2 It is a structural schematic diagram of the optical measurement data analysis system provided by the present invention;

[0041] Figure 3 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0042] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0043] In one embodiment, Figure 1 As shown, the present invention provides an optical measurement data analysis method, comprising the steps of:

[0044] Performing spectral analysis on a plurality of test samples to obtain a spectral deviation index and a spectral correlation coefficient of each of the test samples;

[0045] The spectral deviation index, the spectral deviation index parameter, the spectral correlation coefficient and the spectral correlation coefficient parameter are combined to create a mathematical model of a spectral matching degree index, and the mathematical model is used to calculate the spectral matching degree index value of each detection sample, and the spectral matching degree index value of each detection sample is greater than a preset spectral matching degree index threshold value;

[0046] The spectral analysis is performed on the to-be-tested sample to obtain the spectral deviation index and the spectral correlation coefficient of the to-be-tested sample, and the spectral matching degree index value of the to-be-tested sample is calculated based on the mathematical model. If the spectral matching degree index value of the to-be-tested sample is greater than the preset spectral matching degree index threshold value, the to-be-tested sample is a normal sample; otherwise, the to-be-tested sample is an abnormal sample.

[0047] The spectral analysis is performed on the to-be-tested sample to obtain the spectral deviation index and the spectral correlation coefficient of the to-be-tested sample, and the spectral matching degree index value of the to-be-tested sample is calculated based on the mathematical model. If the spectral matching degree index value of the to-be-tested sample is greater than the preset spectral matching degree index threshold value, the to-be-tested sample is a normal sample; otherwise, the to-be-tested sample is an abnormal sample.

[0048] It can be understood that, for the convenience of description, the deviation index described in the present application should be regarded as a spectral deviation index, and the correlation coefficient is also a spectral correlation coefficient. Correspondingly, the deviation index parameter, the correlation coefficient parameter and the corresponding threshold value are also spectral correlation data.

[0049] The physical model of the detection sample is established to obtain the theoretical spectrum of the detection sample; the detection sample is measured to obtain a plurality of measurement spectra.

[0050] It should be noted that the establishment of the physical model includes: based on the design information of the detection sample, such as the film layer stacking structure, the nominal value of the film layer thickness, the structure width, or the information determined in advance through other measurement methods, for example, the refractive index of the material used in the to-be-tested structure, the physical model of the detection sample is established;

[0051] The physical model is based on Maxwell's equations or a simplified form thereof, and is used to realize the calculation from the feature parameters (geometric parameters, material refractive index, etc.) of the to-be-tested sample to the scattered spectrum. Common modeling methods include but are not limited to 4*4 matrix method, rigorous coupled wave analysis (Rigorous Coupled-Wave Analysis), finite element method (Finite Element Method), boundary element method (Boundary Element Method), etc.

[0052] The deviation index and the correlation coefficient of the theoretical spectrum and the measurement spectrum are calculated through the spectral analysis;

[0053] The measurement spectrum analysis refers to adjusting the characteristic parameters (geometric parameters, material refractive index, etc.) of the detected sample in the physical model so that the simulation spectrum calculated based on the physical model matches the measurement spectrum. The analysis methods used include but are not limited to nonlinear fitting algorithms (such as gradient descent method, Newton iteration method, Levenberg-Marquardt method, etc.), library search method, machine learning method, etc.

[0054] The types of the theoretical spectrum and the measurement spectrum are consistent, and the types of the theoretical spectrum and the measurement spectrum include one or a combination of more of reflectivity, transmissivity, ellipsometric parameters, and Mueller matrix signals.

[0055] The deviation index includes but is not limited to mean squared error (MSE), root mean squared error (RMSE), mean absolute error (MAE), etc.; the correlation coefficient includes but is not limited to Pearson correlation coefficient and Spearman correlation coefficient, etc.; the present application does not limit this.

[0056] Further, the method comprises the steps of:

[0057] The initial value of the preset deviation index parameter and the initial value of the correlation coefficient parameter are calculated, and the spectral matching degree index of each detected sample is compared with the spectral matching degree index threshold value;

[0058] If the spectral matching degree evaluation of any one of the detected samples is not greater than the spectral matching degree index threshold value, the initial value of the deviation index parameter and the initial value of the correlation coefficient parameter are updated by a preset algorithm, the spectral matching degree index of each detected sample is recalculated and compared with the spectral matching degree index threshold value;

[0059] The initial value of the deviation index parameter and the initial value of the correlation coefficient parameter are continuously optimized; until the spectral matching degree index of each detected sample is greater than the spectral matching degree index threshold value, the updated initial value of the deviation index parameter and the initial value of the correlation coefficient parameter are output as the final preferred value of the deviation index parameter and the value of the correlation coefficient parameter. Generally, the initial value of the preset deviation index parameter approaches 0, and the initial value of the preset correlation coefficient parameter approaches 1. In some embodiments, the range of the initial value of the preset deviation index parameter is (0, 0.5), preferably (0, 0.2), and the range of the initial value of the preset correlation coefficient parameter is (0.5, 1.5), preferably (0.8, 1.2); for example, the initial value of the preset deviation index parameter is 0.05, and the initial value of the preset correlation coefficient parameter is 0.95.

[0060] Specifically, the spectral matching degree index is marked as GOF (Goodness Of Fitting), the deviation index is marked as M, the correlation coefficient is marked as p, the initial value of the correlation coefficient parameter is a0, and the initial value of the deviation index parameter is b0.

[0061] Specifically, the method comprises the following steps:

[0062] a spectral matching degree index threshold GOF is given tol ; the initial value of the preset deviation index parameter b0 and the initial value of the correlation coefficient parameter a0 are given.

[0063] Based on the deviation index and the correlation coefficient of the plurality of detection samples, the initial value of the deviation index parameter and the initial value of the correlation coefficient parameter, the mathematical model of the spectral matching degree index is substituted to calculate the spectral matching degree index GOF of each detection sample.

[0064] The spectral matching degree index GOF of each detection sample is compared with the spectral matching degree index threshold GOF tol , and if there is any one spectral matching degree index less than or equal to the spectral matching degree index threshold GOF tol , the initial value of the correlation coefficient parameter and the initial value of the deviation index parameter are updated by a preset algorithm.

[0065] It should be noted that the preset algorithm for updating the initial value of the correlation coefficient parameter and the initial value of the deviation index parameter includes but is not limited to global optimization algorithms (such as genetic algorithm, particle swarm algorithm, ant colony algorithm, etc.) and traversal method. The correlation coefficient parameter and the deviation index parameter are sampled by the traversal algorithm, and then the optimal parameter value is selected from the results.

[0066] After updating the initial value of the correlation coefficient parameter and the initial value of the deviation index parameter, the spectral matching degree index GOF of each detection sample is recalculated, and if all the spectral matching degree indexes GOF are greater than the spectral matching degree index threshold GOF tol , the updated initial value of the correlation coefficient parameter and the initial value of the deviation index parameter are output as the value of the correlation coefficient parameter and the value of the deviation index parameter, respectively; and the corresponding correlation coefficient parameter and deviation index parameter are used as the optimal parameters a and b.

[0067] Further, the spectral matching degree index of the to-be-tested piece is calculated based on the optimal parameters, and whether the to-be-tested piece is a normal piece is determined, comprising the following steps:

[0068] Based on the same physical model, the theoretical spectrum of the to-be-tested piece is obtained, and the to-be-tested piece is measured to obtain a measured spectrum; the spectral analysis is performed, and the deviation index and the correlation coefficient of the to-be-tested piece are calculated.

[0069] Based on the preferred parameters α and β obtained in the above steps, the spectral matching degree index of the to-be-tested piece is calculated by substituting the calculation formula of the spectral matching degree index mathematical model.

[0070] If the spectral matching degree index of the to-be-tested piece is greater than the spectral matching degree index threshold value, the to-be-tested piece is a normal piece; otherwise, the to-be-tested piece is an abnormal piece.

[0071] In a preferred embodiment, the optimization calculation of the preferred value β of the preset deviation index parameter and the preferred value α of the correlation coefficient parameter further comprises:

[0072] The preset deviation index threshold value and the correlation coefficient threshold value, for any one detection sample, if the deviation index of the detection sample is less than the deviation index threshold value and the correlation coefficient of the detection sample is greater than the correlation coefficient threshold value, the detection sample and its corresponding data are retained; otherwise, the detection sample is regarded as an abnormal sample, and the detection sample and its corresponding data are removed.

[0073] That is, for data of any one detection sample:

[0074] The deviation index of the detection sample is greater than or equal to the deviation index threshold value;

[0075] The correlation coefficient of the detection sample is less than or equal to the correlation coefficient threshold value;

[0076] As long as one of the above conditions is met, the data of the detection sample is regarded as abnormal data, and the abnormal data is removed;

[0077] Preferably, the 3-sigma criterion, Tukey box plot method and other abnormal data detection methods can be used to remove the analysis abnormal detection sample and its corresponding data.

[0078] It should be noted that the correlation coefficient is used to reflect the alignment degree of the characteristic measurement spectrum and the theoretical spectrum peak and valley, and the value range is [0, 1].

[0079] After analyzing N detection samples, N deviation indexes M i and correlation coefficients ρ i (i=1, 2,..., N) are obtained; after removing abnormal data, for the remaining L detection samples, L deviation indexes M' j and correlation coefficients ρ' j (j=1, 2,..., L), L≤N are obtained.

[0080] Further, the deviation indexes of the retained detection samples are normalized to obtain normalized deviation indexes including:

[0081] The initial value α0 of the preset deviation index parameter and the initial value β0 of the correlation coefficient parameter;

[0082] Assuming that the theoretical spectrum and the measured spectrum are completely correlated at this time, the value of the correlation coefficient is set to 1;

[0083] The value of the spectral matching degree index is set to the spectral matching degree index threshold, so that the formula is satisfied:

[0084]

[0085] The range of the normalized deviation index is The deviation index of the reserved detection sample is normalized to The normalization method includes but is not limited to linear function normalization, zero mean standardization, inverse tangent function conversion and the like, and the specific method of normalization is not limited.

[0086] Wherein, GOF is the spectral matching degree index, GOF tol is the spectral matching degree index threshold, α0 is the initial value of the correlation coefficient parameter, β0 is the initial value of the deviation index parameter, ρ is the correlation coefficient, is the maximum value of the normalized deviation index.

[0087] It should be noted that the calculation formula of the spectral matching degree index can be as follows:

[0088] Or

[0089]

[0090] Wherein, the correlation coefficient preferred parameter α is the coefficient of the correlation coefficient ρ, and the deviation index preferred parameter β is the coefficient of the normalized deviation index .

[0091] The specific form of the calculation formula of the spectral matching degree index is not limited, and the calculation formula of the spectral matching degree index should satisfy the following conditions:

[0092] (1) The spectral matching degree index GOF and the correlation coefficient ρ are in a monotonically increasing relationship, and the deviation index is in a monotonically decreasing relationship;

[0093] (2) The range of the spectral matching degree index is [0, 1].

[0094] It should be noted that the spectral matching degree index GOF is a function related to the correlation coefficient, the bias index and the correlation coefficient parameter, the bias index parameter, here only part of the calculation form that GOF may have is listed, and it does not constrain the scope of application of the disclosed method; in fact, any GOF calculation method that meets the above conditions is within the protection scope of the present application.

[0095] As an example, when the calculation formula of the spectral matching degree index is The maximum value of the normalized bias index Should meet:

[0096]

[0097] That is Thus the normalized numerical range of the bias index is Further complete the normalization of the bias index.

[0098] Further, preferably, the spectral matching degree index of the to-be-tested member is calculated based on the preferred parameters, and it is judged whether the to-be-tested member is a normal member, comprising the steps that:

[0099] The theoretical spectrum of the to-be-tested member is obtained based on the same physical model, and the to-be-tested member is measured to obtain a measured spectrum; spectral analysis is performed, and the bias index and the correlation coefficient of the to-be-tested member are calculated;

[0100] At this time, according to the normalized numerical range of the bias index The bias index of the to-be-tested member is normalized to obtain a normalized bias index;

[0101] The spectral matching degree index of the to-be-tested member is calculated based on the preferred parameters α and β obtained in the above steps and the calculation formula of the spectral matching degree index;

[0102] If the spectral matching degree index of the to-be-tested member is greater than the spectral matching degree index threshold, it is judged that the to-be-tested member is a normal member, otherwise, the to-be-tested member is an abnormal member.

[0103] By normalizing the bias index to the above range, the initial value α0 of the correlation coefficient parameter and the initial value β0 of the bias index parameter are closer to the preferred values of the two parameters, so as to maximize the reduction of iteration speed or the reduction of traversal times, to quickly optimize the correlation coefficient parameter and the bias index parameter, thereby determining the correlation coefficient parameter and the bias index parameter.

[0104] Based on the normalization method of the above embodiment, the iteration speed or the traversal times can be minimized, so as to quickly obtain the preferred deviation index parameter and the correlation coefficient parameter, and then the abnormal sample can be quickly screened from the detection result, and the accuracy of the semiconductor processing yield judgment can be improved.

[0105] In another aspect, the present application also provides an optical measurement data analysis system, such as Figure 2 As shown, the optical scattering measurement data analysis system described below can be referred to the optical measurement method described above, and specifically includes:

[0106] The sample analysis module is configured to perform spectral analysis on the plurality of detection samples, and obtain the spectral deviation index and the spectral correlation coefficient of each detection sample.

[0107] The spectral matching degree index mathematical model creation module is configured to obtain the spectral deviation index parameter and the spectral correlation coefficient parameter, and create a spectral matching degree index mathematical model based on the spectral deviation index, the spectral deviation index parameter, the spectral correlation coefficient and the spectral correlation coefficient parameter. The mathematical model is used to calculate the spectral matching degree index of each detection sample, so that the spectral matching degree index value of each detection sample is greater than a preset spectral matching degree index threshold. The analysis module is configured to perform spectral analysis on the to-be-tested sample, and obtain the spectral deviation index and the spectral correlation coefficient of the to-be-tested sample. The spectral matching degree index value of the to-be-tested sample is calculated based on the mathematical model. If the spectral matching degree index value of the to-be-tested sample is greater than the preset spectral matching degree index threshold, the to-be-tested sample is a normal sample. Otherwise, the to-be-tested sample is an abnormal sample.

[0108] Figure 3 An example of an entity structure diagram of an electronic device is shown in FIG. 1. Figure 3As shown, the electronic device can include a processor 310, a communications interface 320, a memory 330, and a communications bus 340, wherein the processor 310, the communications interface 320, and the memory 330 complete mutual communication through the communications bus 340. The processor 310 can invoke a logical instruction in the memory 330 to execute an optical measurement data analysis method, including steps of: S1, performing spectral analysis on a plurality of detection samples to obtain a spectral deviation index and a spectral correlation coefficient of each detection sample; S2, setting a spectral deviation index parameter and a spectral correlation coefficient parameter, creating a spectral matching degree index mathematical model in combination with the spectral deviation index, the spectral deviation index parameter, the spectral correlation coefficient, and the spectral correlation coefficient parameter, and calculating the spectral matching degree index value of each detection sample based on the mathematical model so that the spectral matching degree index value of each detection sample is greater than a preset spectral matching degree index threshold; S3, performing spectral analysis on a to-be-tested piece to obtain a spectral deviation index and a spectral correlation coefficient of the to-be-tested piece, and calculating a spectral matching degree index value of the to-be-tested piece based on the mathematical model, wherein if the spectral matching degree index value of the to-be-tested piece is greater than the preset spectral matching degree index threshold, the to-be-tested piece is a normal piece; otherwise, the to-be-tested piece is an abnormal piece.

[0109] In addition, the logical instruction in the memory 330 described above can be implemented in the form of a software function unit and sold or used as an independent product, and can be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various program code storage media.

[0110] In another aspect, the present application also provides a computer program product, which comprises a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions that, when executed by a computer, enable the computer to perform an optical measurement data analysis method, comprising: S1, performing spectral analysis on a plurality of detection samples to obtain a spectral deviation index and a spectral correlation coefficient of each detection sample; S2, setting a spectral deviation index parameter and a spectral correlation coefficient parameter, creating a spectral matching degree index mathematical model in combination with the spectral deviation index, the spectral deviation index parameter, the spectral correlation coefficient and the spectral correlation coefficient parameter, and calculating the spectral matching degree index value of each detection sample based on the mathematical model so that the spectral matching degree index value of each detection sample is greater than a preset spectral matching degree index threshold value; S3, performing spectral analysis on a to-be-tested piece to obtain a spectral deviation index and a spectral correlation coefficient of the to-be-tested piece, and calculating a spectral matching degree index value of the to-be-tested piece based on the mathematical model, if the spectral matching degree index value of the to-be-tested piece is greater than the preset spectral matching degree index threshold value, the to-be-tested piece is a normal piece; otherwise, the to-be-tested piece is an abnormal piece.

[0111] In another aspect, the present application also provides a non-transitory computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement an optical measurement data analysis method, comprising: S1, performing spectral analysis on a plurality of detection samples to obtain a spectral deviation index and a spectral correlation coefficient of each detection sample; S2, setting a spectral deviation index parameter and a spectral correlation coefficient parameter, creating a spectral matching degree index mathematical model in combination with the spectral deviation index, the spectral deviation index parameter, the spectral correlation coefficient and the spectral correlation coefficient parameter, and calculating the spectral matching degree index value of each detection sample based on the mathematical model so that the spectral matching degree index value of each detection sample is greater than a preset spectral matching degree index threshold value; S3, performing spectral analysis on a to-be-tested piece to obtain a spectral deviation index and a spectral correlation coefficient of the to-be-tested piece, and calculating a spectral matching degree index value of the to-be-tested piece based on the mathematical model, if the spectral matching degree index value of the to-be-tested piece is greater than the preset spectral matching degree index threshold value, the to-be-tested piece is a normal piece; otherwise, the to-be-tested piece is an abnormal piece.

[0112] The apparatus embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the present embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0113] Those skilled in the art can clearly understand the technical solutions of the various embodiments from the above description of the embodiments, and the various embodiments can be implemented by means of software with the necessary general hardware platforms, and of course, can also be implemented by hardware. Based on such understanding, the above technical solutions, essentially or in other words, the part of the prior art that makes a contribution, can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, and the like, and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0114] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for some technical features therein; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method of analyzing optical measurement data, characterized by, The method comprises the following steps: S1, performing spectral analysis on a plurality of detection samples to obtain a spectral deviation index and a spectral correlation coefficient of each detection sample; S2, setting a spectral deviation index parameter and a spectral correlation coefficient parameter, combining the spectral deviation index, the spectral deviation index parameter, the spectral correlation coefficient and the spectral correlation coefficient parameter to create a spectral matching degree index mathematical model, and calculating the spectral matching degree index value of each detection sample based on the mathematical model, so that the spectral matching degree index value of each detection sample is greater than a preset spectral matching degree index threshold value; S3, performing spectral analysis on a test piece to obtain a spectral deviation index and a spectral correlation coefficient of the test piece, and calculating the spectral matching degree index value of the test piece based on the mathematical model, if the spectral matching degree index value of the test piece is greater than the preset spectral matching degree index threshold value, the test piece is a normal piece; otherwise, the test piece is an abnormal piece; In the spectral matching degree index mathematical model, the spectral matching degree index and the spectral correlation coefficient are in a monotonically increasing relationship, and the spectral matching degree index and the spectral deviation index are in a monotonically decreasing relationship; the range of the spectral matching degree index is [0, 1].

2. The method of claim 1, wherein, The S1 comprises: establishing a physical model of the plurality of detection samples to obtain a theoretical spectrum of the detection sample; measuring a plurality of the detection samples to obtain a measured spectrum of the detection sample; calculating the spectral deviation index and the spectral correlation coefficient of each test sample; wherein the type of the theoretical spectrum and the measured spectrum is consistent, and the type of the theoretical spectrum and the measured spectrum includes one or a combination of more of reflectivity, transmittance, ellipsometric parameters, and Mueller matrix signals.

3. The method of claim 1, wherein, The S1 comprises: presetting a deviation index threshold value and a correlation coefficient threshold value, for any one detection sample, if the spectral deviation index of the detection sample is less than the deviation index threshold value and the spectral correlation coefficient of the detection sample is greater than the correlation coefficient threshold value, the detection sample and its spectral deviation index and spectral correlation coefficient are retained, otherwise, the detection sample and its spectral deviation index and spectral correlation coefficient are rejected.

4. The method of claim 1, wherein, The spectral matching degree index mathematical model comprises: or ; Wherein, GOF is the spectral matching index, a is the spectral correlation coefficient parameter, and β is the spectral deviation index parameter, is the spectral correlation coefficient, and M is the spectral deviation index.

5. The method of claim 1, wherein, normalizing the spectral deviation index of the detection sample to obtain a normalized spectral deviation index, including: setting the value of the spectral correlation coefficient to 1; setting the value of the spectral matching degree index to the spectral matching degree index threshold value, so as to satisfy the formula: ； The normalized spectral deviation index ranges from [0, 1] and normalizes the spectral deviation index of the preserved test sample to [0, 1]. ] will be normalized to [0, 1]. ] will be normalized to [0, 1]. wherein GOF is a spectral matching index, is a threshold value of the spectral matching index, is a value of a spectral correlation coefficient parameter, is a value of a spectral deviation index parameter, is the spectral correlation coefficient, is a maximum value of the normalized spectral deviation index.

6. The method of claim 1, wherein The S2 comprises: presetting the initial value of the spectral deviation index parameter and the initial value of the spectral correlation coefficient parameter, calculating the spectral matching degree index of each detection sample and comparing it with the spectral matching degree index threshold value; if the spectral matching degree index of any one detection sample is not greater than the spectral matching degree index threshold value, updating the initial value of the spectral deviation index parameter and the initial value of the spectral correlation coefficient parameter, recalculating the spectral matching degree index of each detection sample and comparing it with the spectral matching degree index threshold value; Until the spectral matching degree index of each of the detection samples is greater than the spectral matching degree index threshold, output the initial value of the updated spectral deviation index parameter and the initial value of the spectral correlation coefficient parameter as the value of the spectral deviation index parameter and the value of the spectral correlation coefficient parameter respectively.

7. The optical measurement data analysis method according to any one of claims 1-6, characterized in that, the spectral deviation index comprises a mean square error or a mean square root error or a mean absolute error between a theoretical spectrum and a measured spectrum of the detection sample or the to-be-measured member; the spectral correlation coefficient comprises a Pearson correlation coefficient or a Spearman correlation coefficient between the theoretical spectrum and the measured spectrum of the detection sample or the to-be-measured member.

8. An optical measurement data analysis system, characterized by Comprise: a sample analysis module for performing spectral analysis on a plurality of detection samples to obtain a spectral deviation index and a spectral correlation coefficient of each of the detection samples; a spectral matching degree index mathematical model creation module for obtaining a spectral deviation index parameter and a spectral correlation coefficient parameter, creating a spectral matching degree index mathematical model in combination with the spectral deviation index, the spectral deviation index parameter, the spectral correlation coefficient and the spectral correlation coefficient parameter, and calculating a spectral matching degree index of each of the detection samples based on the mathematical model, so that the spectral matching degree index value of each detection sample is greater than a preset spectral matching degree index threshold; an analysis module for performing spectral analysis on a to-be-measured member to obtain a spectral deviation index and a spectral correlation coefficient of the to-be-measured member, calculating a spectral matching degree index value of the to-be-measured member based on the mathematical model, and if the spectral matching degree index value of the to-be-measured member is greater than the preset spectral matching degree index threshold, the to-be-measured member is a normal member; otherwise, the to-be-measured member is an abnormal member; wherein in the spectral matching degree index mathematical model, the spectral matching degree index and the spectral correlation coefficient are in a monotonically increasing relationship, and the spectral matching degree index and the spectral deviation index are in a monotonically decreasing relationship; the range of the spectral matching degree index is [0, 1].

9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to implement the steps of the method of any one of claims 1 to 7.

Images