A method, apparatus, device and medium for optimizing laser-induced spectroscopy

Abstract

The application provides a laser-induced breakdown spectroscopy optimization method, device, equipment and medium, and relates to the technical field of spectrum optimization, and comprises the following steps: obtaining original laser-induced breakdown spectroscopy data; constructing a covariance matrix based on the original laser-induced breakdown spectroscopy data; obtaining a target load matrix; calculating a corresponding wavelength contribution rate under each wavelength dimension based on the target load matrix to obtain a target wavelength contribution rate; and optimizing the original laser-induced breakdown spectroscopy data based on a feature optimization matrix to obtain optimized target spectrum data. By considering the importance of corresponding spectrum data under each wavelength dimension to the overall spectrum, the wavelength contribution rate is calculated, and the feature optimization matrix is constructed based on the normalized wavelength contribution rate, so that the original laser-induced spectrum data is optimized, the redundant features and interference signals in the original spectrum data are effectively inhibited, and the accuracy of detection and analysis experiments using the laser-induced spectrum data is improved.

Description

Technical Field

[0001] This invention relates to the field of spectral optimization technology, and more specifically, to methods, apparatus, equipment, and media for optimizing laser-induced breakdown spectra. Background Technology

[0002] Traditional laser-induced spectral optimization methods have certain limitations. Compared with traditional feature optimization methods, methods based on principal component analysis (PCA) loading have stronger noise suppression and feature enhancement capabilities, and can better adapt to laser-induced breakdown spectral data structures in various scenarios. However, the complexity of spectral signals and noise interference may lead to inaccurate and unstable features extracted by traditional PCA methods, reducing the reliability and accuracy of data analysis. At the same time, it ignores the elemental characterization capabilities of the spectrum itself. Therefore, it is necessary to improve the original PCA method to enhance the accuracy and stability of laser-induced spectral feature analysis. Summary of the Invention

[0003] The purpose of this invention is to provide a method, apparatus, device, and medium for optimizing laser-induced breakdown spectra, thereby improving the aforementioned problems. To achieve the above objective, the technical solution adopted by this invention is as follows:

[0004] In a first aspect, this application provides a method for optimizing laser-induced breakdown spectra, comprising:

[0005] Obtain raw laser-induced breakdown spectral data;

[0006] Based on the original laser-induced breakdown spectral data, a covariance matrix is ​​constructed, and the eigenvalues ​​and eigenvectors of the covariance matrix are calculated to obtain multiple target eigenvalues ​​and corresponding multiple target eigenvectors.

[0007] Based on the correspondence between the target feature values ​​and the target feature vectors, a load matrix is ​​constructed to obtain the target load matrix;

[0008] Based on the target payload matrix, the wavelength contribution rate corresponding to each wavelength dimension is calculated to obtain multiple target wavelength contribution rates. The wavelength contribution rate is used to characterize the importance of the spectral data corresponding to each wavelength dimension to the overall spectrum.

[0009] Based on the target wavelength contribution rate, a feature optimization matrix is ​​constructed, and the original laser-induced breakdown spectral data is optimized based on the feature optimization matrix to obtain the optimized target spectral data.

[0010] Secondly, this application also provides an apparatus for optimizing laser-induced breakdown spectra, comprising:

[0011] The acquisition unit is used to acquire raw laser-induced breakdown spectral data;

[0012] The first calculation unit is used to construct a covariance matrix based on the original laser-induced breakdown spectrum data, and calculate the eigenvalues ​​and eigenvectors of the covariance matrix to obtain multiple target eigenvalues ​​and corresponding multiple target eigenvectors.

[0013] A construction unit is used to construct a load matrix based on the correspondence between the target feature values ​​and the target feature vectors, thereby obtaining a target load matrix;

[0014] The first obtaining unit is used to calculate the wavelength contribution rate corresponding to each wavelength dimension based on the target load matrix, and obtain multiple target wavelength contribution rates. The wavelength contribution rate is used to characterize the importance of the spectral data corresponding to each wavelength dimension to the overall spectrum.

[0015] An optimization unit is used to construct a feature optimization matrix based on the target wavelength contribution rate, and to optimize the original laser-induced breakdown spectral data based on the feature optimization matrix to obtain optimized target spectral data.

[0016] Thirdly, this application also provides an optimization device for laser-induced breakdown spectrum, comprising:

[0017] Memory, used to store computer programs;

[0018] A processor is configured to implement the steps of the laser-induced breakdown spectrum optimization method when executing the computer program.

[0019] Fourthly, this application also provides a readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described optimization method based on laser-induced breakdown spectrum.

[0020] The beneficial effects of this invention are as follows:

[0021] This invention calculates the wavelength contribution rate by considering the importance of the spectral data corresponding to each wavelength dimension to the overall spectrum, and constructs a feature optimization matrix based on the normalized wavelength contribution rate. This optimizes the original laser-induced spectral data, effectively suppressing redundant features and interference signals in the original spectral data, and improving the accuracy of detection and analysis experiments using laser-induced light spectrum data.

[0022] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of the process for optimizing the laser-induced breakdown spectrum as described in an embodiment of the present invention;

[0025] Figure 2 This is a schematic diagram of the detection sample described in an embodiment of the present invention;

[0026] Figure 3 This is a schematic diagram of the laser-induced breakdown spectroscopy detection platform described in an embodiment of the present invention;

[0027] Figure 4 The original spectral data of the four different corrosion grades of the contact finger sheets described in the embodiments of the present invention;

[0028] Figure 5 This is a schematic diagram of the loadings of the first three principal components as described in this embodiment of the invention;

[0029] Figure 6 The optimized spectral data described in the embodiments of the present invention;

[0030] Figure 7 This is a schematic diagram of the structure of the laser-induced breakdown spectrum optimization device described in an embodiment of the present invention;

[0031] Figure 8 This is a schematic diagram of the structure of the laser-induced breakdown spectrum optimization device described in an embodiment of the present invention.

[0032] Marked in the image:

[0033] 10. Acquisition Unit; 20. First Calculation Unit; 30. Construction Unit; 40. First Obtaining Unit; 50. Optimization Unit;

[0034] 800. Optimization equipment for laser-induced breakdown spectrum; 801. Processor; 802. Memory; 803. Multimedia components; 804. I / O interface; 805. Communication components. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0036] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0037] Example 1:

[0038] This embodiment provides a method for optimizing laser-induced breakdown spectra.

[0039] See Figure 1 The figure shows that the method includes steps S10, S20, S30, S40 and S50.

[0040] Step S10. Obtain the raw laser-induced breakdown spectrum data;

[0041] Specifically, LIBS (Laser Induced Breakdown Spectroscopy) is a spectroscopic analysis method that uses laser pulses to induce plasma on the sample surface and analyzes the radiation spectrum of the plasma to determine the chemical composition and content of the sample.

[0042] This application embodiment uses LIBS technology to assess the corrosion status of electrical contact components inside the valve-side riser seat of an ultra-high voltage converter transformer. The test sample is a watch strap contact finger, such as... Figure 2 As shown, this is a typical electrical contact component inside the riser seat on the valve side of an ultra-high voltage converter transformer. Affected by many factors such as electricity, heat, force, chemical reaction, mechanical vibration, and eccentric unbalanced force, this watch band contact sample was disassembled and replaced after 11 years of service in the field due to corrosion and deterioration. A whole watch band contact contains multiple contact pieces, and the degree of corrosion of the contact pieces in different areas varies.

[0043] The deposition and growth scale of copper sulfide compounds, corrosion products, on the surface of the fingertip varied significantly depending on the degree of corrosion. With increasing corrosion, the physicochemical properties of the fingertip surface, such as elemental content, uniformity, and density, changed considerably. In particular, the changes in the copper, silver, and sulfur content on the fingertip surface were highly correlated with the degree of corrosion. The corrosion levels of the fingertip were classified into four categories, and samples from each of the four corrosion levels were selected for LIBS analysis.

[0044] The testing process is based on a self-built LIBS testing platform, the structure of which is shown in the diagram below. Figure 3 As shown, the system includes a reflector, focusing lens, fiber optic probe, three-dimensional stage, grating spectrometer, laser controller, pulsed laser, digital delayed pulse generator, computer, and enhanced charge-coupled device. During the detection process, the single-pulse energy of the laser was set to 60 mJ, the working wavelength to 1064 nm, and the duration to 6 ns. The LIBS spectral signal acquisition delay time was 1 μs. After testing all the fingernail samples using the LIBS detection system, a dataset of 640 LIBS spectra corresponding to four levels of corrosion was obtained. The original laser-induced breakdown spectra are shown below. Figure 4 As shown.

[0045] Specifically, considering that numerous factors such as laser energy fluctuations, environmental interference, changes in the surface structure of the fingertip sample, instrument offset, and matrix effects can all affect the quality of the spectral signal during LIBS data acquisition for fingertip samples, the acquired LIBS data may contain noise components, impacting subsequent evaluation and analysis. Furthermore, the raw LIBS data covers a wide wavelength range, encompassing a large amount of information. Each data point corresponds to a specific wavelength spectral signal intensity value, but most bands in the spectrum do not contain discrete spectral lines representing characteristic elements of the sample surface, which can be considered redundant information, affecting the understanding and analysis of the spectral data. Therefore, to improve the accuracy and sensitivity of subsequent fingertip corrosion state assessment, it is necessary to preprocess the raw LIBS data to optimize and enhance spectral characteristics and amplify valuable signals in the spectrum.

[0046] Step S20. Construct a covariance matrix based on the original laser-induced breakdown spectrum data, and calculate the eigenvalues ​​and eigenvectors of the covariance matrix to obtain multiple target eigenvalues ​​and corresponding multiple target eigenvectors;

[0047] Specifically, each LIBS data entry is first set as a vector x. i =(x i1 ,x i2 ,......,x i1024This forms a 640×1024 original matrix X, as shown below, where m and n represent the number of LIBS data lines and the number of data points per spectral line, respectively, with values ​​of 640 and 1024:

[0048]

[0049] The mean and variance are standardized on the original spectral data vector using the following formula:

[0050]

[0051] Where, x' ij For standardized spectral data; x ij The original spectral data in the X matrix;

[0052] A matrix X' is constructed based on the standardized spectral data, and a covariance matrix C is constructed based on matrix X'. The calculation formula is as follows:

[0053]

[0054] Where C is the covariance matrix; X' is the matrix constructed from the standardized spectral data;

[0055] Based on the covariance matrix C, eigenvalues ​​and eigenvectors are calculated. Eigenvectors represent the basis vectors after transformation of the original spectral dimension space, while the corresponding eigenvalues ​​correspond to the degree of difference in the transformed spectral data under that basis dimension. The principal component analysis loading method does not change the original LIBS data itself; it merely provides a new perspective for better understanding and analyzing the data, concentrating more spectral data information into a few dimensions. The formulas for calculating eigenvalues ​​and eigenvectors are as follows:

[0056] Ca i =λ i a i

[0057] Where C is the covariance matrix; a i Let λ be the feature vector of the i-th basis dimension; i Let be the eigenvalues ​​of the i-th basis dimension; where λ i =(λ1,λ2,……,λ) n ), a i =(a i1 ,a i2 ,……,a in ), i = 1, 2, ..., n.

[0058] Step S30. Construct the load matrix based on the correspondence between the target eigenvalues ​​and the target eigenvectors to obtain the target load matrix;

[0059] Specifically, the target feature values ​​carry principal component identifiers, and the target feature vectors carry basis dimension identifiers and principal component identifiers.

[0060] Specifically, step S30 includes steps S31, S32, and S33:

[0061] Step S31. Calculate the square root of multiple target feature values ​​to obtain multiple calculated feature values. Each calculated feature value carries the same basis dimension identifier as the corresponding target feature value.

[0062] Step S32. Summate the calculated feature values ​​carrying the same basis dimension identifier with the target feature vector to obtain multiple loading vectors. The loading vectors carry the same basis dimension identifier and principal component identifier as the corresponding target feature vector.

[0063] Specifically, the principal component analysis loading method considers breaking the relationship between the principal component vector and the dimensionality of the original spectral data, and loads the loading vector b of the i-th principal component. i =(b i1 ,b i2 ,……,b in It is calculated by multiplying the corresponding eigenvector by the square root of the eigenvalue. The formula is:

[0064]

[0065] Among them, a i Let λ be the feature vector of the i-th basis dimension; i b is the eigenvalue of the i-th basis dimension; i Let be the loading vector of the i-th principal component, i = 1, 2, ..., n;

[0066] Each value within the loading vector represents the associated loading of the principal component with respect to each wavelength. The absolute value of each value, to some extent, characterizes the importance of different wavelength dimensions in the original spectral data for that principal component. Figure 5 As shown, the LIBS load diagrams for the first three principal components are shown. PC-1, PC-2, and PC-3 represent principal component 1, principal component 2, and principal component 3, respectively. It can be seen from the diagram that the loads of principal components 1 and 2 are mainly concentrated in the characteristic peak band region of the LIBS data, which plays a key role in the assessment of the degree of corrosion. The positive and negative signs of the loads also represent the positive and negative correlation between the intensity of the wavelength point and the principal component.

[0067] Step S33. Based on the basis dimension identifier and principal component identifier carried by the load vector, set multiple load vectors to the corresponding positions in the preset matrix to obtain the target load matrix;

[0068] Specifically, the load matrix is ​​constructed based on the load vector as follows:

[0069]

[0070] Where B is the load matrix; λ n a is the eigenvalue of the basis dimension of the nth wavelength; nn Let be the feature vector of the nth principal component corresponding to the nth basis dimension.

[0071] Step S40. Calculate the wavelength contribution rate corresponding to each wavelength dimension based on the target payload matrix to obtain multiple target wavelength contribution rates. The wavelength contribution rate is used to characterize the importance of the spectral data corresponding to each wavelength dimension to the overall spectrum.

[0072] Specifically, step S40 includes steps S41, S42, and S43:

[0073] Step S41. Sum the associated loads in all load vectors corresponding to the same wavelength dimension identifier in the target load matrix to obtain multiple integrated wavelength loads;

[0074] Specifically, by integrating the loadings of the same wavelength dimension of all principal components, the contribution of each wavelength dimension of the original spectral data to the overall data can be evaluated. The integration method is a sum of squares, and the calculation of the integrated wavelength loadings is as follows:

[0075]

[0076] Where, d j To integrate wavelength loads; λ i a is the eigenvalue of the i-th principal component; nn Let be the feature vector of the nth principal component corresponding to the nth basis dimension.

[0077] Step S42. Summate the multiple integrated wavelength loads to obtain the target result;

[0078] Step S43. Calculate the ratio of multiple integrated wavelength loads to the target result to obtain multiple wavelength contribution rates, which carry a basis dimension identifier.

[0079] Specifically, the formula for calculating the wavelength contribution rate is as follows:

[0080]

[0081] Among them, c j The wavelength contribution rate of the j-th basis dimension; λ i Let a be the eigenvalue of the i-th basis dimension; ij Let be the feature vector of the i-th principal component corresponding to the j-th basis dimension.

[0082] Step S50. Construct a feature optimization matrix based on the target wavelength contribution rate, and optimize the original laser-induced breakdown spectrum data based on the feature optimization matrix to obtain the optimized target spectrum data;

[0083] Specifically, step S50 includes steps S51, S52, S53, S54, S55, and S56:

[0084] Step S51. Determine the maximum and minimum values ​​from the multiple wavelength contribution rates, and use them as the maximum contribution rate and minimum contribution rate, respectively;

[0085] Step S52. Normalize the contribution rates of multiple wavelengths, the maximum contribution rate, and the minimum contribution rate, and construct a matrix to obtain the feature optimization matrix;

[0086] Specifically, step S52 includes steps S521, S522, S523, and S524:

[0087] Step S521. Calculate the difference between the contribution rate of multiple wavelengths and the minimum contribution rate to obtain multiple first differences. Each first difference carries a wavelength dimension identifier that is the same as the contribution rate of the corresponding wavelength.

[0088] Step S522. Calculate the difference between the maximum contribution rate and the minimum contribution rate, and use it as the second difference;

[0089] Step S523. Calculate the ratio of multiple first differences to second differences to obtain multiple target ratios. Each target ratio carries the same wavelength dimension identifier as the corresponding first difference.

[0090] Step S524. Based on the wavelength dimension identifier carried by the target ratio, set multiple target ratios at the corresponding positions of a preset diagonal matrix to obtain the feature optimization matrix.

[0091] Specifically, the feature optimization matrix is ​​as follows:

[0092]

[0093] Where G is the feature optimization matrix; c n MIN represents the wavelength contribution rate of the nth wavelength dimension; MAX represents the minimum wavelength contribution rate; and MIN represents the maximum wavelength contribution rate.

[0094] Step S53. Calculate the product of the covariance matrix and the feature optimization matrix to obtain the target matrix after spectral feature optimization and enhancement;

[0095] Specifically, the wavelength dimension contribution rate characterizes the importance of the corresponding wavelength dimension data to the overall spectrum. Based on this contribution rate, the original LIBS data is optimized and adjusted to highlight important features and suppress irrelevant features, ultimately achieving feature optimization and improvement of the LIBS data. The normalization calculation formula for the original spectral data is as follows:

[0096] Y = X * G

[0097] Where X is the original matrix; Y is the optimized target matrix; and G is the normalized matrix;

[0098] Step S54. Perform data restoration processing on the target matrix to obtain the spectral data to be processed;

[0099] Specifically, such as Figure 6 As shown, the optimized spectral data has a significantly improved signal-to-noise ratio compared to the original spectral data. At the same time, the fluctuation of spectral signal intensity has been significantly suppressed, and the relative standard deviation of the spectral data of the same corrosion grade of the finger sheet has been greatly reduced.

[0100] Step S55. Calculate the key point similarity between the spectral data to be processed and the original laser-induced breakdown spectral data to obtain the target similarity;

[0101] Specifically, step S55 includes steps S551, S552, S553, S554, S555, S556, and S557:

[0102] Step S551. Based on the key point detection algorithm, determine the key points in the original laser-induced breakdown spectral data and the target spectral data, and obtain multiple first key points and multiple second key points respectively;

[0103] Step S552. Calculate the feature descriptors of multiple first key points and multiple second key points to obtain multiple first feature descriptors and multiple second feature descriptors respectively;

[0104] Step S553. Based on the matching algorithm, combine multiple first feature descriptors and second feature descriptors to obtain multiple initial combinations;

[0105] Step S554. Calculate the difference in wavelength features corresponding to the first and second feature descriptors in the initial combination to obtain the first difference;

[0106] Step S555. Calculate the difference in light intensity features corresponding to the first and second feature descriptors in the initial combination to obtain the second difference;

[0107] Step S556. Calculate the similarity of the initial combination based on the first difference and the second difference to obtain the local similarity;

[0108] Step S557. Sum the local similarities of all initial combinations to obtain the target similarity;

[0109] Step S56. When the target similarity is greater than the preset threshold, the spectral data to be processed is used as the optimized target spectral data;

[0110] Specifically, considering that there may be differences between the optimized spectral data and the original spectral data, but some key feature points are still the same, the similarity between the optimized spectral data and the original spectral data is calculated to determine whether the optimized optical spectrum data only suppresses redundant and interference signals, but does not suppress important signals.

[0111] First, based on the keypoint detection algorithm, multiple keypoints are identified from the original spectral data and the optimized spectral data respectively. The keypoints in the original spectral data are then paired with the keypoints in the optimized spectral data, assuming that each pair of keypoints is located at the same position in both spectral data and that their corresponding horizontal and vertical coordinates are approximately the same. The similarity is then calculated based on the difference between the horizontal and vertical coordinates. When the similarity is higher than a set threshold, it is considered that the optimized spectral data has effectively suppressed redundant features and interference signals, thereby improving the accuracy and sensitivity of detection and analysis using LIBS technology.

[0112] Example 2:

[0113] like Figure 7 As shown, this embodiment provides an optimization device for laser-induced breakdown spectrum, the device comprising:

[0114] Acquisition unit 10 is used to acquire raw laser-induced breakdown spectral data;

[0115] The first calculation unit 20 is used to construct a covariance matrix based on the original laser-induced breakdown spectrum data, and calculate the eigenvalues ​​and eigenvectors of the covariance matrix to obtain multiple target eigenvalues ​​and corresponding multiple target eigenvectors.

[0116] The construction unit 30 is used to construct the load matrix based on the correspondence between the target feature values ​​and the target feature vectors, thereby obtaining the target load matrix;

[0117] The first unit 40 is used to calculate the wavelength contribution rate corresponding to each wavelength dimension based on the target load matrix, and obtain multiple target wavelength contribution rates. The wavelength contribution rate is used to characterize the importance of the spectral data corresponding to each wavelength dimension to the overall spectrum.

[0118] The optimization unit 50 is used to construct a feature optimization matrix based on the target wavelength contribution rate, and optimize the original laser-induced breakdown spectral data based on the feature optimization matrix to obtain optimized target spectral data.

[0119] In one specific embodiment disclosed in this application, the construction unit 30 includes:

[0120] The second calculation unit is used to calculate the square root of multiple target feature values ​​to obtain multiple calculated feature values. Each calculated feature value carries the same basis dimension identifier as the corresponding target feature value.

[0121] The third calculation unit is used to sum the calculated feature values ​​carrying the same basis dimension identifier with the target feature vector to obtain multiple load vectors. The load vectors carry the same basis dimension identifier and principal component identifier as the corresponding target feature vector.

[0122] The first setting unit is used to set multiple load vectors at corresponding positions in a preset matrix based on the basis dimension identifier and principal component identifier carried by the load vectors, thereby obtaining the target load matrix.

[0123] In one specific embodiment disclosed in this application, the first obtaining unit 40 includes:

[0124] The fourth calculation unit is used to sum the associated loads in all load vectors corresponding to the same wavelength dimension identifier in the target load matrix to obtain multiple integrated wavelength loads.

[0125] The fifth calculation unit is used to sum and calculate multiple integrated wavelength loads to obtain the target result;

[0126] The sixth calculation unit is used to calculate the ratio of multiple integrated wavelength loads to the target result, and obtain multiple wavelength contribution rates, which carry the basis dimension identifier.

[0127] In one specific embodiment disclosed in this application, the optimization unit 50 includes:

[0128] The first determining unit is used to determine the maximum and minimum values ​​from multiple wavelength contribution rates, which are respectively used as the maximum contribution rate and the minimum contribution rate.

[0129] The second obtaining unit is used to normalize the contribution rates of multiple wavelengths, the maximum contribution rate, and the minimum contribution rate and construct a matrix to obtain the feature optimization matrix;

[0130] The seventh calculation unit is used to calculate the product of the covariance matrix and the feature optimization matrix to obtain the target matrix after spectral feature optimization and enhancement.

[0131] The data restoration unit is used to perform data restoration processing on the target matrix to obtain the spectral data to be processed.

[0132] The eighth calculation unit is used to calculate the key point similarity between the spectral data to be processed and the original laser-induced breakdown spectral data to obtain the target similarity.

[0133] As a unit, it is used to use the spectral data to be processed as the optimized target spectral data when the target similarity is greater than a preset threshold.

[0134] In one specific embodiment disclosed in this application, the second obtaining unit includes:

[0135] The ninth calculation unit is used to calculate the difference between the contribution rate of multiple wavelengths and the minimum contribution rate, and obtain multiple first differences. Each first difference carries a wavelength dimension identifier that is the same as the contribution rate of the corresponding wavelength.

[0136] The tenth calculation unit is used to calculate the difference between the maximum contribution rate and the minimum contribution rate, which is used as the second difference.

[0137] The eleventh calculation unit is used to calculate the ratio of multiple first differences to second differences, and obtain multiple target ratios respectively. Each target ratio carries a wavelength dimension identifier that is the same as the corresponding first difference.

[0138] The second setting unit is used to set multiple target ratios at corresponding positions in a preset diagonal matrix based on the wavelength dimension identifier carried by the target ratio, thereby obtaining a feature optimization matrix.

[0139] In one specific embodiment disclosed in this application, the eighth computing unit includes:

[0140] The second determining unit is used to determine key points in the original laser-induced breakdown spectral data and target spectral data based on the key point detection algorithm, and obtain multiple first key points and multiple second key points respectively.

[0141] The twelfth calculation unit is used to calculate the feature descriptors of multiple first key points and multiple second key points, and obtain multiple first feature descriptors and multiple second feature descriptors respectively;

[0142] The combination unit is used to combine multiple first feature descriptors and second feature descriptors based on a matching algorithm to obtain multiple initial combinations;

[0143] The thirteenth calculation unit is used to calculate the difference in wavelength characteristics corresponding to the first feature descriptor and the second feature descriptor in the initial combination, and obtain the first difference;

[0144] The fourteenth calculation unit is used to calculate the difference in light intensity characteristics corresponding to the first feature descriptor and the second feature descriptor in the initial combination, and obtain the second difference;

[0145] The fifteenth calculation unit is used to calculate the similarity of the initial combination based on the first difference and the second difference, and to obtain the local similarity.

[0146] The sixteenth calculation unit is used to sum the local similarities of all initial combinations to obtain the target similarity.

[0147] It should be noted that the specific manner in which each module performs its operation in the apparatus described in the above embodiments has been described in detail in the embodiments of the method, and will not be elaborated here.

[0148] Example 3:

[0149] Corresponding to the above method embodiments, this embodiment also provides a laser-induced breakdown spectrum optimization device. The laser-induced breakdown spectrum optimization device described below and the laser-induced breakdown spectrum optimization method described above can be referred to each other.

[0150] Figure 8 This is a block diagram illustrating an optimization device 800 for laser-induced breakdown spectrum according to an exemplary embodiment. Figure 8 As shown, the laser-induced breakdown spectrum optimization device 800 may include a processor 801 and a memory 802. The laser-induced breakdown spectrum optimization device 800 may also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0151] The processor 801 controls the overall operation of the laser-induced breakdown spectrum optimization device 800 to complete all or part of the steps in the laser-induced breakdown spectrum optimization method described above. The memory 802 stores various types of data to support the operation of the laser-induced breakdown spectrum optimization device 800. This data may include, for example, instructions for any application or method operating on the laser-induced breakdown spectrum optimization device 800, as well as application-related data such as contact data, sent and received messages, images, audio, video, etc. The memory 802 can be implemented using any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touchscreen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in the memory 802 or transmitted via the communication component 805. The audio component also includes at least one speaker for outputting audio signals. I / O interface 804 provides an interface between processor 801 and other interface modules, such as keyboards, mice, and buttons. These buttons can be virtual or physical. Communication component 805 is used for wired or wireless communication between the laser-induced breakdown spectrum optimization device 800 and other devices. Wireless communication includes, for example, Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination thereof. Therefore, the corresponding communication component 805 may include a Wi-Fi module, a Bluetooth module, or an NFC module.

[0152] In an exemplary embodiment, the laser-induced breakdown spectrum optimization device 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the laser-induced breakdown spectrum optimization method described above.

[0153] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of the laser-induced breakdown spectrum optimization method described above. For example, the computer-readable storage medium may be the memory 802 including program instructions described above, which may be executed by the processor 801 of the laser-induced breakdown spectrum optimization device 800 to complete the laser-induced breakdown spectrum optimization method described above.

[0154] Example 4:

[0155] Corresponding to the above method embodiments, this embodiment also provides a readable storage medium. The readable storage medium described below and the laser-induced breakdown spectrum optimization method described above can be referred to in correspondence.

[0156] A readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the laser-induced breakdown spectrum optimization method described in the above method embodiments.

[0157] Specifically, the 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 any other readable storage medium capable of storing program code.

[0158] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0159] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for optimizing laser-induced breakdown spectra, characterized in that, include: Obtain raw laser-induced breakdown spectral data; Based on the original laser-induced breakdown spectral data, a covariance matrix is ​​constructed, and the eigenvalues ​​and eigenvectors of the covariance matrix are calculated to obtain multiple target eigenvalues ​​and corresponding multiple target eigenvectors. Based on the correspondence between the target feature values ​​and the target feature vectors, a load matrix is ​​constructed to obtain the target load matrix; Based on the target payload matrix, the wavelength contribution rate corresponding to each wavelength dimension is calculated to obtain multiple target wavelength contribution rates. The wavelength contribution rate is used to characterize the importance of the spectral data corresponding to each wavelength dimension to the overall spectrum. A feature optimization matrix is ​​constructed based on the target wavelength contribution rate, and the original laser-induced breakdown spectral data is optimized based on the feature optimization matrix to obtain the optimized target spectral data. Specifically, a feature optimization matrix is ​​constructed based on the target wavelength contribution rate, and the original laser-induced breakdown spectral data is optimized based on the feature optimization matrix to obtain optimized target spectral data, including: The maximum and minimum values ​​of the contribution rates from multiple wavelengths are determined and used as the maximum and minimum contribution rates, respectively. The contribution rates of multiple wavelengths, the maximum contribution rate, and the minimum contribution rate are normalized and a matrix is ​​constructed to obtain the feature optimization matrix; Calculate the product of the covariance matrix and the feature optimization matrix to obtain the target matrix after spectral feature optimization and enhancement; The target matrix is ​​subjected to data restoration processing to obtain the spectral data to be processed; Calculate the key point similarity between the spectral data to be processed and the original laser-induced breakdown spectral data to obtain the target similarity; When the target similarity is greater than a preset threshold, the spectral data to be processed is used as the optimized target spectral data; The feature optimization matrix is ​​obtained by normalizing the contribution rates of multiple wavelengths, the maximum contribution rate, and the minimum contribution rate, and constructing a matrix, including: Calculate the difference between multiple wavelength contribution rates and the minimum contribution rate to obtain multiple first differences, each of which carries a wavelength dimension identifier that is the same as the corresponding wavelength contribution rate. Calculate the difference between the maximum contribution rate and the minimum contribution rate, and use it as the second difference; Calculate the ratio of multiple first differences to second differences to obtain multiple target ratios, wherein each target ratio carries a wavelength dimension identifier that is the same as the corresponding first difference. Based on the wavelength dimension identifier carried by the target ratio, multiple target ratios are set at corresponding positions in a preset diagonal matrix to obtain the feature optimization matrix.

2. The method for optimizing the laser-induced breakdown spectrum according to claim 1, characterized in that... A loading matrix is ​​constructed based on the correspondence between the target feature values ​​and the target feature vectors to obtain the target loading matrix. The target feature values ​​carry principal component identifiers, and the target feature vectors carry basis dimension identifiers and principal component identifiers, including: Calculate the square root of multiple target feature values ​​to obtain multiple calculated feature values, wherein each calculated feature value carries the same basis dimension identifier as the corresponding target feature value; The calculated feature values ​​carrying the same basis dimension identifier are summed with the target feature vector to obtain multiple load vectors. The load vectors carry the same basis dimension identifier and principal component identifier as the corresponding target feature vector. Based on the basis dimension identifier and principal component identifier carried by the load vector, multiple load vectors are set in the corresponding positions of a preset matrix to obtain the target load matrix.

3. The method for optimizing the laser-induced breakdown spectrum according to claim 2, characterized in that... Based on the target payload matrix, the wavelength contribution rate corresponding to each wavelength dimension is calculated to obtain multiple target wavelength contribution rates, including: The associated loads in all load vectors corresponding to the same wavelength dimension identifier in the target load matrix are summed to obtain multiple integrated wavelength loads. The target result is obtained by summing multiple integrated wavelength loads. The ratios of multiple integrated wavelength loads to the target result are calculated to obtain multiple wavelength contribution rates, which carry a basis dimension identifier.

4. An optimization device for laser-induced breakdown spectrum, characterized in that, include: The acquisition unit is used to acquire raw laser-induced breakdown spectral data; The first calculation unit is used to construct a covariance matrix based on the original laser-induced breakdown spectrum data, and calculate the eigenvalues ​​and eigenvectors of the covariance matrix to obtain multiple target eigenvalues ​​and corresponding multiple target eigenvectors. A construction unit is used to construct a load matrix based on the correspondence between the target feature values ​​and the target feature vectors, thereby obtaining a target load matrix; The first obtaining unit is used to calculate the wavelength contribution rate corresponding to each wavelength dimension based on the target load matrix, and obtain multiple target wavelength contribution rates. The wavelength contribution rate is used to characterize the importance of the spectral data corresponding to each wavelength dimension to the overall spectrum. An optimization unit is used to construct a feature optimization matrix based on the target wavelength contribution rate, and to optimize the original laser-induced breakdown spectral data based on the feature optimization matrix to obtain optimized target spectral data. Specifically, a feature optimization matrix is ​​constructed based on the target wavelength contribution rate, and the original laser-induced breakdown spectral data is optimized based on the feature optimization matrix to obtain optimized target spectral data, including: The maximum and minimum values ​​of the contribution rates from multiple wavelengths are determined and used as the maximum and minimum contribution rates, respectively. The contribution rates of multiple wavelengths, the maximum contribution rate, and the minimum contribution rate are normalized and a matrix is ​​constructed to obtain the feature optimization matrix; Calculate the product of the covariance matrix and the feature optimization matrix to obtain the target matrix after spectral feature optimization and enhancement; The target matrix is ​​subjected to data restoration processing to obtain the spectral data to be processed; Calculate the key point similarity between the spectral data to be processed and the original laser-induced breakdown spectral data to obtain the target similarity; When the target similarity is greater than a preset threshold, the spectral data to be processed is used as the optimized target spectral data; The feature optimization matrix is ​​obtained by normalizing the contribution rates of multiple wavelengths, the maximum contribution rate, and the minimum contribution rate, and constructing a matrix, including: Calculate the difference between multiple wavelength contribution rates and the minimum contribution rate to obtain multiple first differences, each of which carries a wavelength dimension identifier that is the same as the corresponding wavelength contribution rate. Calculate the difference between the maximum contribution rate and the minimum contribution rate, and use it as the second difference; Calculate the ratio of multiple first differences to second differences to obtain multiple target ratios, wherein each target ratio carries a wavelength dimension identifier that is the same as the corresponding first difference. Based on the wavelength dimension identifier carried by the target ratio, multiple target ratios are set at corresponding positions in a preset diagonal matrix to obtain the feature optimization matrix.

5. The apparatus for optimizing the laser-induced breakdown spectrum according to claim 4, characterized in that, The building unit includes: The second calculation unit is used to calculate the square root of multiple target feature values ​​to obtain multiple calculated feature values, wherein the calculated feature values ​​carry the same basis dimension identifier as the corresponding target feature values; The third calculation unit is used to sum the calculated feature values ​​carrying the same basis dimension identifier with the target feature vector to obtain multiple load vectors. The load vectors carry the same basis dimension identifier and principal component identifier as the corresponding target feature vector. The first setting unit is used to set multiple load vectors at corresponding positions in a preset matrix based on the basis dimension identifier and principal component identifier carried by the load vectors, thereby obtaining the target load matrix.

6. The apparatus for optimizing the laser-induced breakdown spectrum according to claim 5, characterized in that, The first obtaining unit includes: The fourth calculation unit is used to sum the associated loads in all load vectors corresponding to the same wavelength dimension identifier in the target load matrix to obtain multiple integrated wavelength loads. The fifth calculation unit is used to sum and calculate multiple integrated wavelength loads to obtain the target result; The sixth calculation unit is used to calculate the ratio of multiple integrated wavelength loads to the target result, and obtain multiple wavelength contribution rates, wherein the wavelength contribution rate carries a wavelength dimension identifier.

7. An optimization device for laser-induced breakdown spectrum, characterized in that, include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of the method for optimizing the laser-induced breakdown spectrum as described in any one of claims 1 to 3.

8. A readable storage medium, characterized in that, The readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method for optimizing the laser-induced breakdown spectrum as described in any one of claims 1 to 3.

Images