Pteris vittata arsenic content quantitative detection method, device, equipment, medium and product

Abstract

The invention discloses a method, a device, equipment, a medium and a product for quantitatively detecting arsenic content of ciliate desert-grass, and relates to the technical field of environmental pollution element monitoring, the method comprises the following steps: constructing a sample spectrum database of ciliate desert-grass; obtaining an LIBS spectrum matrix and an XRF spectrum matrix of the to-be-detected ciliate desert-grass; based on the sample spectrum database, screening spectral intensities corresponding to spectral wavelengths in the LIBS spectral matrix and the XRF spectral matrix of the to-be-detected ciliate desert-grass to obtain screened LIBS spectral intensity and XRF spectral intensity; establishing a ciliate desert-grass arsenic content quantitative model by adopting a partial least square method; establishing a precise arsenic content detection model of the ciliate desert-grass by adopting a fusion strategy based on error reciprocal weighting; based on the screened LIBS spectral intensity and XRF spectral intensity, the ciliate desert-grass arsenic content quantitative model and the ciliate desert-grass arsenic content accurate detection model are adopted to obtain a ciliate desert-grass arsenic content predicted value, and the ciliate desert-grass arsenic content detection efficiency and accuracy can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of environmental pollution element monitoring, and particularly to a method, device, equipment, medium and product for quantitatively detecting the arsenic content in Pteris vittata. Background Art

[0002] Arsenic (As) pollution widely exists in soil, and its carcinogenic characteristics have become an important environmental problem. Pteris vittata, as the first plant species confirmed to have arsenic hyperaccumulation ability. Detecting the arsenic content in the whole plant of Pteris vittata is of great significance for deeply exploring its arsenic hyperaccumulation ability and optimizing its application in contaminated soil remediation. Pteris vittata has the ability to absorb and accumulate a large amount of arsenic in different parts. Therefore, accurately analyzing the distribution law of arsenic in the plant is of core significance for revealing its unique absorption mechanism, scientifically evaluating the remediation efficiency and formulating optimized treatment strategies.

[0003] Theoretically, any atomic spectrometer system can be used for the detection and analysis of arsenic. At present, there are various methods for detecting and quantitatively analyzing arsenic in plants, including inductively coupled plasma mass spectrometry (ICP-MS), colorimetry, electrophoresis, anodic stripping voltammetry, biological detection methods, atomic fluorescence spectrometry, etc. Although laboratory-level detection methods usually have high precision, the sample processing process is complex and time-consuming, resulting in untimely detection of arsenic content in Pteris vittata, which not only is not conducive to timely grasping the progress of soil remediation, but also affects the timeliness of soil remediation decision-making. Summary of the Invention

[0004] The purpose of the present application is to provide a method, device, equipment, medium and product for quantitatively detecting the arsenic content in Pteris vittata, which can improve the efficiency and accuracy of detecting the arsenic content in Pteris vittata.

[0005] To achieve the above purpose, the present application provides the following solutions:

[0006] In a first aspect, the present application provides a method for quantitatively detecting the arsenic content in Pteris vittata, including:

[0007] Constructing a sample spectral database of Pteris vittata; the sample spectral database includes: sample LIBS spectral wavelengths and sample XRF spectral wavelengths;

[0008] Obtaining a LIBS spectral matrix and an XRF spectral matrix of the Pteris vittata to be measured; the spectral matrix includes: spectral intensities corresponding to multiple spectral wavelengths;

[0009] Based on the sample spectral database, screening the spectral intensities corresponding to the spectral wavelengths in the LIBS spectral matrix and the XRF spectral matrix of the Pteris vittata to be measured, to obtain the screened LIBS spectral intensities and XRF spectral intensities;

[0010] A quantitative model for the arsenic content of Pteris vittata L. is established by using partial least squares method;

[0011] An accurate detection model for the arsenic content of Pteris vittata L. is established by using a fusion strategy based on the reciprocal of error weighted;

[0012] Based on the screened LIBS spectral intensity and XRF spectral intensity, using the quantitative model for the arsenic content of Pteris vittata L., the predicted value of the arsenic content of Pteris vittata L. based on LIBS technology and the predicted value of the arsenic content of Pteris vittata L. based on XRF technology are obtained;

[0013] Based on the predicted value of the arsenic content of Pteris vittata L. based on LIBS technology and the predicted value of the arsenic content of Pteris vittata L. based on XRF technology, the predicted value of the arsenic content of Pteris vittata L. is calculated by using the accurate detection model for the arsenic content of Pteris vittata L.

[0014] In a second aspect, the present application provides a quantitative detection system for the arsenic content of Pteris vittata L., including:

[0015] A spectral database construction module, configured to construct a sample spectral database of Pteris vittata L.; the sample spectral database includes: sample LIBS spectral wavelengths and sample XRF spectral wavelengths;

[0016] An acquisition module, configured to acquire a LIBS spectral matrix and an XRF spectral matrix of the Pteris vittata L. to be measured; the spectral matrix includes: spectral intensities corresponding to multiple spectral wavelengths;

[0017] A screening module, configured to screen the spectral intensities corresponding to the spectral wavelengths in the LIBS spectral matrix and the XRF spectral matrix of the Pteris vittata L. to be measured based on the sample spectral database, so as to obtain the screened LIBS spectral intensity and XRF spectral intensity;

[0018] A quantitative model construction module, configured to establish a quantitative model for the arsenic content of Pteris vittata L. by using partial least squares method;

[0019] An accurate detection model construction module, configured to establish an accurate detection model for the arsenic content of Pteris vittata L. by using a fusion strategy based on the reciprocal of error weighted;

[0020] A quantitative model prediction module, configured to input the screened LIBS spectral matrix and XRF spectral matrix into the quantitative model for the arsenic content of Pteris vittata L., so as to obtain the predicted value of the arsenic content of Pteris vittata L. based on LIBS technology and the predicted value of the arsenic content of Pteris vittata L. based on XRF technology;

[0021] An accurate detection model prediction module, configured to calculate the predicted value of the arsenic content of Pteris vittata L. by using the accurate detection model for the arsenic content of Pteris vittata L. based on the predicted value of the arsenic content of Pteris vittata L. based on LIBS technology and the predicted value of the arsenic content of Pteris vittata L. based on XRF technology.

[0022] In a third aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method for quantitatively detecting the arsenic content in centipede grass as described in the first aspect above.

[0023] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for quantitatively detecting arsenic content in centipede grass as described in the first aspect above.

[0024] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the method for quantitatively detecting arsenic content in centipede grass as described in the first aspect above.

[0025] According to the specific embodiments provided in this application, this application has the following technical effects:

[0026] The present application provides a method, device, equipment, medium and product for quantitative detection of arsenic content in centipede grass, which adopts LIBS and XRF spectral technologies without the need for complex pre-treatment processes such as sample digestion required by traditional analytical methods. By obtaining the LIBS spectrum and XRF spectrum of centipede grass, a mathematical model can be further used to obtain the predicted value of arsenic content in centipede grass, and the prediction of arsenic content in centipede grass to be tested is completed. The whole process, through model calculation, greatly reduces the time to obtain the predicted value of arsenic content in centipede grass, and improves the efficiency of arsenic content detection in centipede grass; through the fusion strategy of inverse error weighting, a centipede grass arsenic content accurate detection model with both rapid response and high precision is constructed, and the high efficiency and high accuracy determination of arsenic content in centipede grass is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0028] Figure 1 A schematic diagram of a flow chart of a method for quantitatively detecting arsenic content in centipede grass provided in one embodiment of the present application;

[0029] Figure 2 A schematic diagram of the LIBS spectrum of a sample of centipede grass provided in an embodiment of the present application;

[0030] Figure 3 A schematic diagram of the XRF spectrum of a sample of centipede grass provided in an embodiment of the present application;

[0031] Figure 4 Schematic diagram of arsenic content in sample Pteris vittata L. provided by an embodiment of the present application;

[0032] Figure 5 Schematic diagram of the structure of a computer device provided by an embodiment of the present application. Detailed implementation manners

[0033] Next, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0034] The Laser-induced breakdown spectroscopy (LIBS) technology adopted in the present application is a newly emerging and powerful analysis technology, which is analyzed based on the optical emission lines generated after laser ablation. This technology uses a single laser pulse to qualitatively and quantitatively analyze the elemental composition of aerosol, liquid, gas and solid samples. When a high-energy pulsed laser beam is focused on the surface of the sample, a laser spark will be formed, generating a high-temperature plasma, thereby evaporating, atomizing and exciting a small part of the substance in the sample. After the plasma cools down, the electrons return to the ground state and emit light with characteristic wavelengths. In the past decade, the LIBS technology has been widely used to directly analyze solid samples of plant materials (such as leaves, roots, fruits, etc.). The main advantages of the LIBS technology are simple equipment, only requiring a laser, a high-sensitivity spectrometer and a computer for data acquisition, and fast detection speed, usually only taking a few seconds.

[0035] The X-ray Fluorescence Spectrometer (XRF) technology adopted in the present application is an efficient on-site arsenic detection method and is also one of the few technologies that can directly detect the arsenic content in solid matrices (such as soil) without water extraction. XRF identifies elements by detecting the characteristic wavelengths emitted after the sample is irradiated by X-rays. The primary X-rays emitted by XRF will excite the target atoms, causing them to release secondary X-rays (fluorescence). The detector can accurately determine the elemental composition of the sample by analyzing the energy spectrum of these fluorescences. When analyzing arsenic in an aqueous solution, the XRF method needs to preconcentrate the sample on a suitable solid matrix first, and the detection limit can be as low as 50 ppb. The XRF technology is easy to operate, does not require sample digestion or irradiation, does not damage the sample during the detection process, and the sample can be recovered after analysis, so it is suitable for the study of toxicity or inorganic elements in plant materials.

[0036] To make the above objects, features, and advantages of the present application more apparent and understandable, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0037] In an exemplary embodiment, as Figure 1 shown, a method for quantitatively detecting the arsenic content of Pteris vittata is provided. This method is executed by a computer device, and specifically, it can be executed alone by a computer device such as a terminal or a server, or jointly executed by a terminal and a server. In the embodiments of the present application, taking the application of this method to Figure 1 the server in

[0038] as an example, the following steps 1 to 7 are included. Among them:

[0039] Specifically, constructing a sample spectral database of Pteris vittata specifically includes the following steps:

[0040] Step 11: Prepare a tablet sample from the sample Pteris vittata.

[0041] Specifically, use the Pteris vittata plants cultivated with arsenic-containing nutrient solution, collect the root / leaf samples of Pteris vittata, and prepare a tablet sample.

[0042] Step 12: Use a laser-induced breakdown spectroscopy system to obtain the initial LIBS spectrum of the tablet sample.

[0043] Specifically, use laser-induced breakdown spectroscopy technology to collect the spectrum of the Pteris vittata sample and construct the initial LIBS spectrum X LIBS =[x1,x2,…,x p n×p (n is the number of samples, p is the wavelength, x1 represents the spectral intensity value at the 1st wavelength, x2 represents the spectral intensity value at the 2nd wavelength, x p represents the spectral intensity value at the pth wavelength).

[0044] Step 13: Use an X-ray fluorescence spectroscopy system to obtain the initial XRF spectrum of the tablet sample; the spectrum includes: wavelength and the corresponding spectral intensity.

[0045] Specifically, use X-ray fluorescence technology to collect the spectrum of the Pteris vittata sample and construct the initial XRF spectrum X XRF =[x1,x2,…,x q n×q (n is the number of samples, q is the wavelength, x1 represents the spectral intensity value at the 1st wavelength, x2 represents the spectral intensity value at the 2nd wavelength, x q represents the spectral intensity value at the qth wavelength). ​​

[0046] Step 14: Process the initial LIBS spectrum by area normalization to obtain the processed LIBS spectrum.

[0047] Step 15: Use the competitive adaptive reweighted sampling method to extract features from the processed LIBS spectrum and the initial XRF spectrum respectively, obtaining the sample LIBS spectrum wavelengths and their corresponding spectral intensities, as well as the sample XRF spectrum wavelengths and their corresponding spectral intensities.

[0048] Specifically, preprocess the initial LIBS spectrum by area normalization without preprocessing the initial XRF spectrum. Use competitive adaptive reweighted sampling (CARS) to extract the characteristic bands of the processed LIBS spectrum and the initial XRF spectrum, and extract the characteristic bands most closely associated with arsenic from them to optimize the feature dimension and eliminate redundant variables, forming the screened spectral matrix X selected =[x1,x2,…,x p1 n×p1 (n is the number of samples, p1 is the number of wavelengths after screening, x1 represents the spectral intensity value at the first wavelength, x2 represents the spectral intensity value at the second wavelength, and x p1 represents the spectral intensity value at the p1th wavelength).

[0049] Step 16: Based on the sample LIBS spectrum wavelengths and the sample XRF spectrum wavelengths, construct a sample spectral database of Pteris vittata.

[0050] Step 2: Obtain the LIBS spectral matrix and the XRF spectral matrix of the Pteris vittata to be measured; the spectral matrix includes: spectral intensities corresponding to multiple spectral wavelengths.

[0051] Step 3: Based on the sample spectral database, screen the spectral intensities corresponding to the spectral wavelengths in the LIBS spectral matrix and the XRF spectral matrix of the Pteris vittata to be measured, obtaining the screened LIBS spectral intensities and XRF spectral intensities.

[0052] Step 4: Establish a quantitative model for the arsenic content of Pteris vittata using partial least squares method. Among them, the quantitative model for the arsenic content of Pteris vittata includes the LIBS quantitative model for the arsenic content of Pteris vittata and the XRF quantitative model for the arsenic content of Pteris vittata. Specifically, it includes the following steps:

[0053] Step 41: Use inductively coupled plasma mass spectrometry to measure the true arsenic element content of the sample Pteris vittata to obtain the concentration matrix of the sample Pteris vittata.

[0054] ​Specifically, the true value of the arsenic content in the Pteris vittata sample (i.e., the true arsenic content in the Pteris vittata sample) was measured using inductively coupled plasma mass spectrometry, and the concentration matrix C of the Pteris vittata sample was constructed. rel =[C] m (where m is the number of samples, and C is the true arsenic content of a single sample (in mg / kg)).

[0055] Step 42: Based on the spectral intensities corresponding to the wavelengths of the LIBS spectra of the samples, construct the LIBS spectral matrix of the samples.

[0056] Step 43: Based on the spectral intensities corresponding to the wavelengths of the XRF spectra of the samples, construct the XRF spectral matrix of the samples.

[0057] Step 44: Based on the LIBS spectral matrix and the XRF spectral matrix of the samples, construct the training set and the prediction set.

[0058] Step 45: Using the LIBS spectral matrix of the samples in the training set as the independent variable and the concentration matrix of the Pteris vittata samples as the dependent variable, adopt the least squares regression method to obtain the quantitative model for the arsenic content in Pteris vittata by LIBS.

[0059] Specifically, the expression of the quantitative model for the arsenic content in Pteris vittata by LIBS is:

[0060]

[0061] where C LIBS is the predicted value of the arsenic content in Pteris vittata based on LIBS technology; is the fitting coefficient of the i-th LIBS wavelength; is the spectral intensity corresponding to the i-th LIBS wavelength; a is the intercept of the quantitative model for the arsenic content in Pteris vittata by LIBS.

[0062] Step 47: Using the XRF spectral matrix of the samples in the training set as the independent variable and the concentration matrix of the Pteris vittata samples as the dependent variable, adopt the least squares regression method to obtain the quantitative model for the arsenic content in Pteris vittata by XRF.

[0063] Specifically, the expression of the quantitative model for the arsenic content in Pteris vittata by XRF is:

[0064]

[0065] where C XRF is the predicted value of the arsenic content in Pteris vittata based on XRF technology; is the fitting coefficient of the j-th XRF wavelength; is the spectral intensity corresponding to the j-th XRF wavelength; b is the intercept of the quantitative model for the arsenic content in Pteris vittata by XRF.

[0066] Step 5: Establish an accurate detection model for arsenic content in Pteris vittata using a fusion strategy based on the reciprocal of error weighting. The specific steps are as follows:

[0067] Step 51: Based on the prediction set, use the quantitative model for arsenic content in Pteris vittata to obtain the root mean square error based on LIBS technology and the root mean square error based on XRF technology.

[0068] Step 52: Normalize the reciprocal of the root mean square error based on LIBS technology and the reciprocal of the root mean square error based on XRF technology to obtain the weight based on LIBS technology and the weight based on XRF technology.

[0069] Step 53: Based on the weight based on LIBS technology and the weight based on XRF technology, construct an accurate detection model for arsenic content in Pteris vittata.

[0070] Specifically, the expression of the accurate detection model for arsenic content in Pteris vittata is:

[0071] C final =w′ LIBS ×C LIBS +w' XRF ×C XRF 。

[0072]

[0073] Among them, C final is the predicted value of arsenic content in Pteris vittata; w′ LIBS is the weight based on LIBS technology; C LIBS is the predicted value of arsenic content in Pteris vittata based on LIBS technology; w′ XRF is the weight based on XRF technology; C XRF is the predicted value of arsenic content in Pteris vittata based on XRF technology; w LIBS is the reciprocal of the root mean square error based on LIBS technology; w XRF is the reciprocal of the root mean square error based on XRF technology; RMSE LIBS is the root mean square error based on LIBS technology; RMSE XRF is the root mean square error based on XRF technology.

[0074] Step 6: Based on the filtered LIBS spectral intensity and XRF spectral intensity, use the quantitative model for arsenic content in Pteris vittata to obtain the predicted value of arsenic content in Pteris vittata based on LIBS technology and the predicted value of arsenic content in Pteris vittata based on XRF technology.

[0075] Step 7: Based on the predicted value of arsenic content in Pteris vittata based on LIBS technology and the predicted value of arsenic content in Pteris vittata based on XRF technology, use the accurate detection model for arsenic content in Pteris vittata to calculate the predicted value of arsenic content in Pteris vittata.

[0076] For the above-mentioned quantitative detection method of arsenic content in Pteris vittata, a specific embodiment is provided, and the steps are as follows:

[0077] Step S1: Prepare a pressed tablet of Pteris vittata sample (i.e., the pressed tablet sample).

[0078] Specifically, select 30 Pteris vittata plants with uniform height (about 10 cm) and transfer them to a 1 L black container (one plant per pot). Cultivate the plants using 0.2 times Hoagland's nutrient solution (HNS), add arsenate (disodium hydrogen arsenate) with a concentration of 50 μM - 500 μM to the hydroponic solution. After 14 days of growth, take out the plants and wash the roots with ice-cold phosphate buffer (1 mM Na2HPO4, 0.5 mM Ca(NO3)2, and 10 mM MES) and ultrapure water to remove the adsorbed As on the surface. Collect 60 Pteris vittata samples (30 root samples and 30 leaf samples). Take 0.2 g of the dried, ground, and sieved samples and press them into flat and tight tablets with a uniform thickness of 1.5 mm using a tablet press. Three replicated sample tablets are made for each sample for the collection of LIBS and XRF spectra, totaling 180 samples.

[0079] Step S2: Obtain the LIBS spectral data of Pteris vittata (i.e., the initial LIBS spectrum).

[0080] Specifically, use a laser-induced breakdown spectroscopy system to obtain the LIBS spectral data of Pteris vittata samples. An Nd:YAG pulsed solid-state laser (Vlite-200, Beamtech, Beijing, China) is used as the ablation source, emitting laser pulses with a wavelength of 532 nm and a pulse width of 8 ns. The laser energy is set to 90 mJ, and the frequency is set to 1 Hz. The laser is transmitted through an optical system and focused by a plano-convex lens with a focal length of 100 mm about 2 mm below the pressed tablet sample, thereby generating a stable plasma. During the measurement, the delay time and gate width are set to 2.7 μs and 16 μs, respectively. The spectral acquisition uses a spectrometer (SR500i, Andor, Belfast, UK), covering a wavelength range of 220 to 240 nm, and the signal detection is completed by an intensified charge-coupled device (ICCD, DH334T-18F-03, Andor, Belfast, UK). The timing control between the laser and the ICCD detector is performed by a digital delay generator (DG645, Stanford Research Systems, California, USA). In the 4 mm × 4 mm area of each pressed tablet sample, each point is ablated 5 times at a step of 1 mm to obtain 80 spectra, and the average value of the spectra is used as the LIBS spectral data of the sample. Construct the LIBS spectral data X LIBS =[x1,x2,…,x 1024180×1024 (180 is the number of samples, 1024 is the wavelength, x1 represents the spectral intensity value at the 1st wavelength, x2 represents the spectral intensity value at the 2nd wavelength, x 1024 represents the spectral intensity at the 1024th wavelength), and its LIBS spectrum is as Figure 2 shown.

[0081] Step S3: Obtain the Pteris vittata XRF spectral data (i.e., the initial XRF spectrum).

[0082] Specifically, use X-ray fluorescence technology to obtain the Pteris vittata sample XRF spectral data. A portable XRF analyzer (Olympus, Vanta M Series, USA) is used to scan the samples. The instrument uses a voltage of 8 - 50 kV as the excitation source. During measurement, the measurement window is directly close to the pressed sample tablet for scanning, and the single scan time is 60 seconds. Each sample is scanned 3 times repeatedly, and the average value of the spectra is used as the XRF spectral data of the sample. Construct the XRF spectral data X XRF = [x1, x2, …, x 2048 180×2048 (180 is the number of samples, 2048 is the wavelength, x1 represents the spectral intensity value at the 1st wavelength, x2 represents the spectral intensity value at the 2nd wavelength, x 1024 represents the spectral intensity at the 2048th wavelength), and its XRF spectrum is as Figure 3 shown.

[0083] Step S4: Determine the true value of the arsenic element content in Pteris vittata.

[0084] Specifically, the true value of the arsenic element content in Pteris vittata (i.e., arsenic concentration) is as Figure 4 shown. The air-dried sample of Pteris vittata (about 0.50 g) is digested according to the USEPA3050B method using a HNO3 - H2O2 mixed reagent on a graphite digestion instrument (EnvironmentalExpress, Mt. Pleasant, SC). After digestion is completed, the sample of Pteris vittata is filtered through a 0.45 μm filter membrane, and then the inductively coupled plasma mass spectrometer (ICP - MS, NexION 300X, Perkin Elmer, Waltham, MA) is used to determine the true arsenic element content of the sample of Pteris vittata, and a concentration matrix C of the sample of Pteris vittata is constructed rel = [C] m (m is the number of samples, C is the arsenic element content of a single sample (in mg / kg)).

[0085] Step S5: Spectral preprocessing, use the competitive adaptive reweighted sampling method to screen the LIBS and XRF characteristic wavelengths.

[0086] ​​Specifically, all LIBS spectral data and XRF spectral data are divided into three data sets in a ratio of 3:1:1: a training set, a prediction set, and a validation set. Among them, 108 samples are used for the training set, 36 samples are used for the prediction set, and 36 samples are used for the validation set. After comparing and validating various preprocessing methods, area normalization is finally selected for LIBS spectra to reduce the intensity deviation caused by measurement conditions and sample differences; while no preprocessing is performed on XRF spectra.

[0087] Furthermore, the competitive adaptive reweighted sampling method is used to screen the characteristic wavelengths of LIBS and XRF. First, an initial linear regression model is constructed based on all spectral bands. The normalized LIBS spectra and XRF spectra are used as independent variables, and the concentration of the sample Pteris vittata is used as the dependent variable to evaluate the correlation between each band and the arsenic content. The absolute value of the regression coefficient of each wavelength is calculated as the wavelength weight, and the wavelengths are gradually screened through multiple Monte Carlo samplings and adaptive reweighting to form a screened spectral matrix. The matrix after screening the LIBS spectra (i.e., the wavelengths of the sample LIBS spectra and the corresponding spectral intensities) is X LIBS-selected =[x1,x2,…,x 115 180×115 (180 is the number of samples, 115 is the number of screened wavelengths), and the screened wavelengths are shown in Table 1; the matrix after screening the XRF spectra (i.e., the wavelengths of the sample XRF spectra and the corresponding spectral intensities) is X XRF-selected =[x1,x2,…,x 148 180×148 (180 is the number of samples, 148 is the number of screened wavelengths), and the screened wavelengths are shown in Table 2.

[0088] Table 1 Wavelengths after screening the LIBS spectra

[0089] Wavelength Wavelength Wavelength Wavelength Wavelength Wavelength Wavelength Wavelength 219.74 219.76 219.78 219.80 220.02 220.05 220.13 220.29 220.38 220.46 220.71 220.77 221.04 221.14 222.77 223.01 223.03 223.59 223.78 223.96 224.27 224.29 224.41 224.54 224.93 225.11 225.32 225.42 225.65 225.73 226.08 226.20 226.57 226.72 226.84 226.86 226.96 226.98 227.09 227.27 227.62 228.38 228.54 228.81 229.00 229.20 229.43 229.49 230.23 230.35 230.47 230.53 230.70 230.72 230.88 231.04 231.07 231.29 231.37 231.58 231.62 231.72 231.74 231.80 232.03 232.07 232.19 232.58 232.89 232.99 233.19 233.42 233.91 234.20 234.28 234.30 234.56 235.03 235.07 235.26 235.28 235.50 235.63 235.69 235.75 235.77 235.93 236.22 236.50 236.81 236.87 237.11 237.16 237.24 237.36 237.54 237.56 237.85 238.01 238.58 238.62 238.66 238.75 238.77 238.87 238.89 238.93 238.99 239.05 239.48 239.93 240.07 240.15 240.35 240.41

[0090] Table 2 Wavelengths after screening the XRF spectra

[0091] Wavelength Wavelength Wavelength Wavelength Wavelength Wavelength Wavelength Wavelength Wavelength Wavelength 1.58 1.76 2.28 2.30 2.32 2.34 2.36 2.78 3.00 3.48 3.52 3.54 3.56 3.62 3.66 3.68 3.72 3.76 3.96 3.98 4.42 4.52 4.82 4.84 4.86 5.40 5.58 5.84 5.98 6.08 6.32 6.36 6.46 6.56 6.58 6.66 6.76 6.84 6.98 7.00 7.04 7.12 7.24 7.30 7.36 7.38 7.44 7.46 7.52 7.64 7.78 8.02 8.06 8.18 8.22 8.48 8.54 8.58 8.64 8.66 8.68 8.92 8.94 8.96 9.04 9.12 9.16 9.28 9.32 9.40 9.52 9.58 9.60 9.64 9.72 9.82 9.84 9.86 9.90 9.92 9.94 10.02 10.06 10.16 10.28 10.34 10.40 10.42 10.46 10.48 10.54 10.56 10.62 10.68 10.74 10.76 10.78 10.88 10.90 10.92 10.94 10.98 11.14 11.16 11.20 11.30 11.32 11.36 11.42 11.46 11.54 11.58 11.60 11.72 11.76 11.86 11.88 11.94 12.10 12.12 12.22 12.28 12.34 12.40 12.48 12.72 12.74 12.84 12.88 12.92 13.06 13.08 13.10 13.16 13.18 13.32 13.42 13.48 13.62 13.64 13.80 13.88 14.00 14.04 14.10 14.32 14.42 14.60

[0092] Step S6: A quantitative model for the arsenic content of Pteris vittata is initially established using the partial least squares method.

[0093] Specifically, the partial least squares regression (PLSR) method is used for the screened spectral data. X LIBS-selected 、X XRF-selected are used as independent variables, and the concentration matrix C of the sample Pteris vittata rel ​​Taking [dependent variable], a calibration model was constructed between the spectral intensity of the spectral characteristic band and the arsenic element content. The fitting formula for modeling using the PLSR model with LIBS spectra is where, β l represents the fitting coefficient of the l-th LIBS wavelength, x l represents the spectral intensity corresponding to the l-th LIBS wavelength, a is the intercept term (the value is 610.9971), and the fitting parameters x l and β l The specific values are listed in Table 3. Similarly, the fitting formula for modeling using the PLSR model with XRF spectra is where, β r represents the fitting coefficient of the r-th XRF wavelength, x r represents the spectral intensity corresponding to the r-th XRF wavelength, b is the intercept term (the value is 511.2326), and the fitting parameters x r and β r The specific values are listed in Table 4.

[0094] Table 3 Fitting parameters of the PLSR model for LIBS spectra after feature screening (x l and β l )

[0095] <![CDATA[x l > <![CDATA[β l > <![CDATA[x l > <![CDATA[β l > <![CDATA[x l > <![CDATA[β l > <![CDATA[x l > <![CDATA[β l > <![CDATA[x 10 > -0.0001 <![CDATA[x 11 > -0.0001 <![CDATA[x 12 > -0.0001 <![CDATA[x 35 > -0.0001 <![CDATA[x 36 > -0.0001 <![CDATA[x 37 > -0.0001 <![CDATA[x 38 > -0.0001 <![CDATA[x 39 > -0.0001 <![CDATA[x 40 > -0.0001 <![CDATA[x 41 > -0.0001 <![CDATA[x 47 > 0.0002 <![CDATA[x 48 > 0.0002 <![CDATA[x 50 > -0.0001 <![CDATA[x 51 > -0.0001 <![CDATA[x 52 > -0.0001 <![CDATA[x 71 > -0.0001 <![CDATA[x 72 > -0.0001 <![CDATA[x 73 > -0.0003 <![CDATA[x 74 > -0.0001 <![CDATA[x 75 > -0.0001 <![CDATA[x 76 > -0.0001 <![CDATA[x 77 > -0.0001 <![CDATA[x 78 > -0.0005 <![CDATA[x 79 > -0.0003 <![CDATA[x 80 > -0.0002 <![CDATA[x 81 > -0.0002 <![CDATA[x 82 > -0.0001 <![CDATA[x 83 > 0.0001 <![CDATA[x 87 > -0.0001 <![CDATA[x 88 > -0.0001 <![CDATA[x 89 > -0.0003 <![CDATA[x 90 > -0.0001 <![CDATA[x 91 > -0.0001 <![CDATA[x 92 > -0.0003 <![CDATA[x 93 > -0.0003 <![CDATA[x 94 > -0.0002 <![CDATA[x 95 > -0.0003 <![CDATA[x 96 > -0.0001 <![CDATA[x 97 > -0.0001 <![CDATA[x 98 > -0.0004 <![CDATA[x 99 > -0.0004 <![CDATA[x 100 > -0.0005 <![CDATA[x 101 > -0.0006 <![CDATA[x 102 > -0.0008 <![CDATA[x 103 > -0.0012 <![CDATA[x 101 > -0.0012 <![CDATA[x 105 > -0.0008 <![CDATA[x 106 > -0.0006 <![CDATA[x 107 > -0.0004 <![CDATA[x 108 > -0.0002 <![CDATA[x 109 > -0.0002 <![CDATA[x 110 > -0.0004 <![CDATA[x 111 > -0.0003 <![CDATA[x 112 > -0.0009 <![CDATA[x 113 > -0.0011 <![CDATA[x 114 > -0.0005 <![CDATA[x 115 > -0.0006

[0096] Table 4 Fitting parameters of the PLSR model for XRF spectra after feature screening (x r and β r )

[0097] <![CDATA[x r > <![CDATA[β r > <![CDATA[x r > <![CDATA[β r > <![CDATA[x r > <![CDATA[β r > <![CDATA[x r > <![CDATA[β r > <![CDATA[x 14 > -0.0002 <![CDATA[x 15 > -0.0008 <![CDATA[x 16 > -0.0011 <![CDATA[x 17 > -0.0012 <![CDATA[x 18 > -0.0006 <![CDATA[x 19 > -0.0001 <![CDATA[x 20 > -0.0001 <![CDATA[x 22 > -0.0001 <![CDATA[x 31 > -0.0011 <![CDATA[x 32 > -0.0027 <![CDATA[x 33 > -0.0030 <![CDATA[x 34 > -0.0002 <![CDATA[x 39 > -0.0002 <![CDATA[x 40 > -0.0003 <![CDATA[x 41 > -0.0006 <![CDATA[x 42 > -0.0004 <![CDATA[x 75 > 0.0002 <![CDATA[x 76 > 0.0001 <![CDATA[x 77 > 0.0001 <![CDATA[x 78 > 0.0001 <![CDATA[x 79 > 0.0001 <![CDATA[x 80 > 0.0001 <![CDATA[x 81 > 0.0001 <![CDATA[x 82 > 0.0001 <![CDATA[x 87 > 0.0002 <![CDATA[x 88 > 0.0004 <![CDATA[x 89 > 0.0007 <![CDATA[x 90 > 0.0009 <![CDATA[x 91 > 0.0013 <![CDATA[x 92 > 0.0013 <![CDATA[x 93 > 0.0008 <![CDATA[x 94 > 0.0003 <![CDATA[x 95 > 0.0001 <![CDATA[x 96 > 0.0001 <![CDATA[x 97 > 0.0001 <![CDATA[x 98 > 0.0001 <![CDATA[x 99 > 0.0001 <![CDATA[x 100 > 0.0001 <![CDATA[x 101 > 0.0001 <![CDATA[x 102 > 0.0001 <![CDATA[x 103 > 0.0001 <![CDATA[x 104 > 0.0001 <![CDATA[x 105 > 0.0001 <![CDATA[x 106 > 0.0001 <![CDATA[x 111 > 0.0001 <![CDATA[x 112 > 0.0001 <![CDATA[x 113 > 0.0002 <![CDATA[x 114 > 0.0002

[0098] Using the coefficient of determination (R 2 C represents the coefficient of determination of the training set, R 2 P represents the coefficient of determination of the prediction set), root mean square error (RMSE C represents the training set error, RMSE P represents the prediction set error), and average relative error (ARE C represents the average relative error of the training set, ARE P represents the average relative error of the prediction set) of the training set and prediction set to comprehensively evaluate the performance of the model. After CARS screened the characteristic spectra, for the LIBS spectra, the R 2 C of the training set and the R 2 P of the prediction set were 0.9995 and 0.9612 respectively, the RMSE P of the prediction set was 60.39 mg kg-1, and the AREP is 17%; for the XRF spectrum, R 2 C and R 2 P are 0.9999 and 0.9686 respectively, and the RMSE of the prediction set P is 54.13 mg kg-1, and the ARE P is 16%.

[0099] Step S7: Use the fusion strategy based on the reciprocal of the error weighted to establish an accurate detection model for the arsenic content of Pteris vittata L..

[0100] Use two technical means to initially model the reciprocal of the root mean square error of the prediction set as the weight. The smaller the error of the model, the better its prediction performance. The calculation formula is as follows:

[0101]

[0102] Further normalize the weights to ensure that the sum of the weights of the two technologies is 1, ensure that the relative ratio of the two weights is retained, and at the same time, the contribution of a certain technology will not be amplified or reduced. The calculation formula is as follows:

[0103]

[0104] Use the normalized weights to perform weighted fusion on the prediction results of LIBS and XRF. The calculation formula is as follows:

[0105] C final = 0.4727 × C LIBS + 0.5273 × C XRF .

[0106] Similarly, the established weighted fusion model needs to be verified on the validation set to evaluate its prediction accuracy. Use the coefficient of determination (R 2 C represents the coefficient of determination of the training set, R 2 P represents the coefficient of determination of the prediction set), root mean square error (RMSE C represents the training set error, RMSE P represents the prediction error), and average relative error (ARE C represents the average relative error of the training set, ARE P represents the average relative error of the prediction) to comprehensively evaluate the performance of the model. After the two technology decision fusion modeling is completed, the R 2 C of the training set and the R 2 PReaching 0.9999 and 0.9797 respectively, the RMSE of the prediction set P is 43.58 mg / kg -1 , and the ARE P is 10%.

[0107] In this embodiment, it can be seen from the modeling results that the fusion strategy based on the reciprocal weighting of errors can effectively improve the detection accuracy of the arsenic content in Pteris vittata, can effectively weaken the interference of the matrix effect on the detection, and improve the accuracy, stability and generalization performance of the detection model.

[0108] The beneficial effects of a method for quantitatively detecting the arsenic content in Pteris vittata proposed in this application are mainly manifested in:

[0109] (1) Innovatively combining two spectral technologies, LIBS and XRF, for detecting the arsenic content in Pteris vittata, screening the characteristic wavelengths of spectral data by the competitive adaptive reweighted sampling (CARS) method, optimizing the spectral information dimension, and eliminating redundant variables to retain the characteristic bands highly correlated with the arsenic content in Pteris vittata. This characteristic screening strategy significantly improves the modeling efficiency and prediction performance of the model.

[0110] (2) Using the partial least squares regression method (PLSR) to preliminarily quantitatively model the LIBS and XRF spectral data respectively. By comparing the prediction performance of the models, evaluating the advantages and disadvantages of the two spectral technologies in arsenic content detection, and providing a scientific basis for subsequent data fusion.

[0111] (3) A reciprocal weighting strategy based on the root mean square error (RMSE) of the test set is proposed to allocate weights to the prediction results of the LIBS and XRF spectra, and weighted fusion is achieved through normalization processing. This strategy comprehensively utilizes the LIBS and XRF data to construct an accurate detection model of comprehensive spectral information, significantly improving the detection sensitivity and stability of the arsenic content in Pteris vittata.

[0112] Based on the same inventive concept, the embodiment of this application also provides a system for quantitatively detecting the arsenic content in Pteris vittata. The implementation solutions provided by this system to solve problems are similar to the implementation solutions recorded in the above method. Therefore, the specific limitations in one or more embodiments of the system for quantitatively detecting the arsenic content in Pteris vittata provided below can refer to the limitations on the method for quantitatively detecting the arsenic content in Pteris vittata above, and will not be elaborated here.

[0113] In an exemplary embodiment, a system for quantitatively detecting the arsenic content in Pteris vittata is provided, including:

[0114] A spectral database construction module for constructing a sample spectral database of Pteris vittata; the sample spectral database includes: sample LIBS spectral wavelengths and sample XRF spectral wavelengths.

[0115] An acquisition module for acquiring the LIBS spectral matrix and the XRF spectral matrix of the Pteris vittata to be measured; the spectral matrix includes spectral intensities corresponding to multiple spectral wavelengths.

[0116] A screening module for screening the spectral intensities corresponding to the spectral wavelengths in the LIBS spectral matrix and the XRF spectral matrix of the Pteris vittata to be measured based on a sample spectral database to obtain the screened LIBS spectral intensity and XRF spectral intensity.

[0117] A quantitative model construction module for establishing a quantitative model for the arsenic content of Pteris vittata using the partial least squares method.

[0118] A precise detection model construction module for establishing a precise detection model for the arsenic content of Pteris vittata using a fusion strategy based on the reciprocal of the error weighted.

[0119] A quantitative model prediction module for inputting the screened LIBS spectral matrix and XRF spectral matrix into the quantitative model for the arsenic content of Pteris vittata to obtain the predicted value of the arsenic content of Pteris vittata based on the LIBS technology and the predicted value of the arsenic content of Pteris vittata based on the XRF technology.

[0120] A precise detection model prediction module for calculating the predicted value of the arsenic content of Pteris vittata using the precise detection model for the arsenic content of Pteris vittata based on the predicted value of the arsenic content of Pteris vittata based on the LIBS technology and the predicted value of the arsenic content of Pteris vittata based on the XRF technology.

[0121] In an exemplary embodiment, a computer device is provided. The computer device can be a server or a terminal, and its internal structure diagram can be as Figure 5 shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the predicted value of the arsenic content of Pteris vittata. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a method for quantitatively detecting the arsenic content of Pteris vittata.

[0122] Those skilled in the art can understand, Figure 5The structure shown is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different component layout. In an exemplary embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0123] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0124] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0125] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0126] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memories can include read-only memory (ROM), magnetic tapes, floppy disks, flash memories, optical memories, high-density embedded non-volatile memories, resistive random access memories (ReRAMs), magnetoresistive random access memories (MRAMs), ferroelectric random access memories (FRAMs), phase change memories (PCMs), graphene memories, etc. Volatile memories can include random access memories (RAMs) or external caches, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0127] The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchains, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logics, data processing logics based on quantum computing, etc., without limitation.

[0128] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0129] In this article, specific examples are used to elaborate on the principles and implementation manners of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A method for quantitatively detecting arsenic content in centipede grass, characterized in that: The method for quantitatively detecting the arsenic content of centipede grass comprises: Constructing a sample spectrum database of centipede grass; the sample spectrum database includes: sample LIBS spectrum wavelength and sample XRF spectrum wavelength; Obtaining a LIBS spectrum matrix and an XRF spectrum matrix of the centipede grass to be tested; the spectrum matrix includes: spectrum intensities corresponding to a plurality of spectrum wavelengths; Based on the sample spectrum database, the spectral intensities corresponding to the spectral wavelengths in the LIBS spectrum matrix and the XRF spectrum matrix of the centipede grass to be tested are screened to obtain the screened LIBS spectrum intensity and XRF spectrum intensity; The partial least square method was used to establish a quantitative model for arsenic content in Psoralea corylifolia. A fusion strategy based on inverse error weighting was used to establish an accurate detection model for arsenic content in Centipedegrass; Based on the screened LIBS spectrum intensity and XRF spectrum intensity, the arsenic content quantitative model of Scolopendra subtilis was used to obtain the predicted value of arsenic content of Scolopendra subtilis based on LIBS technology and the predicted value of arsenic content of Scolopendra subtilis based on XRF technology. Based on the predicted value of the arsenic content of centipede grass based on the LIBS technology and the predicted value of the arsenic content of centipede grass based on the XRF technology, the predicted value of the arsenic content of centipede grass is calculated using a precise detection model for the arsenic content of centipede grass.

2. The method for quantitatively detecting arsenic content in Centipede Grass according to claim 1, characterized in that: Construct a sample spectral database of centipede grass, including: Prepare compressed tablet samples by sampling centipede grass; The initial LIBS spectra of the pressed pellet samples were obtained using a laser-induced breakdown spectroscopy system; An X-ray fluorescence spectrometer system is used to obtain an initial XRF spectrum of the pressed sample; the spectrum includes: a wavelength and a corresponding spectral intensity; The initial LIBS spectrum was processed by area normalization to obtain the processed LIBS spectrum; The competitive adaptive reweighted sampling method is used to extract features from the processed LIBS spectrum and the initial XRF spectrum, respectively, to obtain the sample LIBS spectrum wavelength and the corresponding spectrum intensity, as well as the sample XRF spectrum wavelength and the corresponding spectrum intensity; Based on the sample LIBS spectrum wavelength and the sample XRF spectrum wavelength, a sample spectrum database of centipede grass is constructed.

3. The method for quantitatively detecting arsenic content in Centipede Grass according to claim 2, characterized in that: The arsenic content quantitative model of centipede grass includes a LIBS arsenic content quantitative model of centipede grass and an XRF arsenic content quantitative model of centipede grass; The partial least squares method was used to establish a quantitative model for arsenic content in centipede grass, including: The real arsenic content of the sample centipede grass was measured by inductively coupled plasma mass spectrometry, and the concentration matrix of the sample centipede grass was obtained. Based on the spectral intensity corresponding to the wavelength of the sample LIBS spectrum, a sample LIBS spectrum matrix is ​​constructed; Based on the spectral intensity corresponding to the wavelength of the sample XRF spectrum, a sample XRF spectrum matrix is ​​constructed; Based on the sample LIBS spectrum matrix and the sample XRF spectrum matrix, construct a training set and a prediction set; The LIBS spectrum matrix of the samples in the training set is used as an independent variable, the concentration matrix of the sample centipede grass is used as a dependent variable, and the least squares regression method is used to obtain a LIBS centipede grass arsenic content quantitative model; The sample XRF spectrum matrix in the training set is used as the independent variable, the concentration matrix of the sample centipede grass is used as the dependent variable, and the least squares regression method is used to obtain the XRF centipede grass arsenic content quantitative model.

4. The method for quantitatively detecting arsenic content in Centipede Grass according to claim 3, characterized in that: A fusion strategy based on inverse error weighting was used to establish an accurate detection model for arsenic content in centipede grass, including: Based on the prediction set, a quantitative model for arsenic content in centipede grass was used to obtain the root mean square error based on LIBS technology and the root mean square error based on XRF technology; Normalizing the reciprocal of the root mean square error based on the LIBS technology and the reciprocal of the root mean square error based on the XRF technology to obtain a weight based on the LIBS technology and a weight based on the XRF technology; Based on the weights based on the LIBS technology and the weights based on the XRF technology, a precise detection model for the arsenic content in Scutellaria baicalensis is constructed.

5. The method for quantitatively detecting arsenic content in Centipede Grass according to claim 3, characterized in that: The expression of the LIBS arsenic content quantitative model of centipede grass is: Among them, C LIBS The predicted value of arsenic content in Scolopendra subspinipes based on LIBS technology; is the fitting coefficient of the i-th LIBS wavelength; is the spectral intensity corresponding to the i-th LIBS wavelength; a is the intercept of the LIBS arsenic content quantitative model of centipede grass; The expression of the XRF centipede grass arsenic content quantitative model is: Among them, C XRF The predicted value of arsenic content in Scutellaria baicalensis based on XRF technology; is the fitting coefficient of the jth XRF wavelength; is the spectral intensity corresponding to the jth XRF wavelength; b is the intercept of the XRF arsenic content quantitative model of centipede grass.

6. The method for quantitatively detecting arsenic content in Centipede Grass according to claim 1, characterized in that: The expression of the accurate detection model of arsenic content in centipede grass is: C final =w′ LIBS ×C LIBS +w' XRF ×C XRF ; Among them, C final is the predicted value of arsenic content in centipede grass; w′ LIBS is the weight based on LIBS technology; C LIBS is the predicted value of arsenic content in Centipede grass based on LIBS technology; w' XRF is the weight based on XRF technology; C XRF is the predicted value of arsenic content in Centipede Grass based on XRF technology; LIBS is the inverse of the root mean square error based on LIBS technology; w XRF RMSE is the inverse of the root mean square error based on XRF technology; LIBS is the root mean square error based on LIBS technology; RMSE XRF is the root mean square error based on XRF technology.

7. A quantitative detection system for arsenic content in centipede grass, characterized in that: The method for quantitatively detecting arsenic content in centipede grass as described in any one of claims 1 to 6, wherein the system for quantitatively detecting arsenic content in centipede grass comprises: A spectrum database construction module is used to construct a sample spectrum database of centipede grass; the sample spectrum database includes: sample LIBS spectrum wavelength and sample XRF spectrum wavelength; An acquisition module is used to acquire a LIBS spectrum matrix and an XRF spectrum matrix of the centipede grass to be tested; the spectrum matrix includes: spectrum intensities corresponding to a plurality of spectrum wavelengths; A screening module is used to screen the spectral intensities corresponding to the spectral wavelengths in the LIBS spectral matrix and the XRF spectral matrix of the centipede grass to be tested based on the sample spectral database to obtain the screened LIBS spectral intensity and XRF spectral intensity; The quantitative model building module is used to establish a quantitative model for arsenic content in centipede grass using the partial least squares method; The module for building an accurate detection model is used to establish an accurate detection model for arsenic content in Scolopendra subspinipes by adopting a fusion strategy based on inverse error weighting; A quantitative model prediction module is used to input the screened LIBS spectrum matrix and XRF spectrum matrix into the arsenic content quantitative model of Scolopendra subtilis to obtain the arsenic content prediction value of Scolopendra subtilis based on LIBS technology and the arsenic content prediction value of Scolopendra subtilis based on XRF technology; The accurate detection model prediction module is used to calculate the predicted value of arsenic content in centipede grass based on the predicted value of arsenic content in centipede grass based on the LIBS technology and the predicted value of arsenic content in centipede grass based on the XRF technology, using the accurate detection model of arsenic content in centipede grass.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for quantitatively detecting the arsenic content in centipede grass according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for quantitatively detecting the arsenic content in centipede grass according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for quantitatively detecting the arsenic content in centipede grass according to any one of claims 1 to 6 is implemented.

Images