Volatile organic compounds (VOCs) pollution prevention is an urgent demand for national environmental protection and public health. Adsorption is one of the most effective methods to control VOC pollution. Current understanding of the adsorption mechanism is limited, which hinders its industrial application. Identifying the microscopic adsorption mechanism is a key issue that needs to be urgently addressed for the efficient treatment of VOCs. Existing conventional methods to characterize microscopic adsorption cannot directly monitor ultrafast adsorption in real-time. New theories and techniques need to be constantly developed. In recent years, terahertz waves have shown great potential in characterizing adsorption processes. Weak interactions in the adsorption process include van der Waals forces and hydrogen bonds, whose vibrational modes are in the THz band. The surface interface mobility carrier changes due to bond breaking and formation have different THz response characteristics for different concentrations. Optical parameters of THz spectra, such as absorption peaks, amplitudes, and waveforms, have been used to characterize adsorption properties. The breaking and formation of chemical bonds in the adsorption process is on the order of picoseconds. Effective information on the femtosecond timescale is not directly available due to the limitation of the THz time resolution. Continued breakthroughs in terahertz time-resolved spectroscopy in ultrafast process monitoring satisfy the need to explore ultrafast adsorption processes on time scales and lay the foundation for uncovering microscopic adsorption mechanisms. This paper reviewed adsorption law studied by terahertz waves, and ultrafast processes were researched using terahertz time-resolved spectroscopy in recent years. Finally, the development directions and major challenges of terahertz time-resolved spectroscopy in ultrafast adsorption are proposed.

Elemental content detection is necessary for efficiently utilizing scrap steel, an important raw material for electric furnace steelmaking.In this study, a new method combining the sparrow search algorithm optimized kernel extreme learning machine (SSA-KELM) and laser-induced breakdown spectroscopy (LIBS) was proposed to analyze and model the element contents of 12 groups of steel samples, including medium-low alloy steel and low alloy steel. First, the portable LIBS spectrometer was used to collect laser-induced breakdown spectroscopy data of 12 different steel scrap samples in the range of 170~400 nm, and 28 different locations on the surface of each sample were selected for detection to reduce experimental fluctuations. The k-value check was used to eliminate gross errors, and the remaining data was averaged to obtain 336 groups of average spectrum data from 12 sample groups. Then, the obtained spectral data was subjected to baseline correction and normalization to reduce the baseline fluctuation. Multiple related spectral lines of the target elements were selected as the input features of the model, and the spectral data was divided into training and testing sets. A random sample from each steel type was selected as the models testing set, and the remaining data was used as the models training set. The sparrow search algorithm was used to optimize the parameters of the kernel extreme learning machine (KELM), and the model was established for the related elements. The final model for C, Cu, Mn, Cr, Ni, Si, V, Al, and Ti elements had an average correlation coefficient (R2) and root mean square error (RMSE) of 0.996 and 0.016, respectively, on the validation set. The quantitative analysis performance of the single variable calibration model and the genetic algorithm optimized KELM (GA-KELM) multivariate calibration model were compared, and the results showed that the SSA-KELM model had significant improvements in all indicators compared to the single variable calibration model and GA-KELM model. The combination of KELM and Sparrow search algorithm as a multivariate model can effectively reduce the interference of multiple factors on the target elements and enhance the performance of the quantitative analysis. It can rapidly and accurately detect various element contents in steel scrap on-site by combining it with the portable LIBS system.

In recent years, seawater pollution caused by microplastic waste has caught more and more attention. Raman spectroscopy, a non-destructive detection technique, has representative spectral characteristic peaks, making it easier to identify unknown samples. It has always been one of the popular detection methods in biochemical analysis. Deep learning has made remarkable achievements in feature extraction, target detection, and other fields. The feasibility of Raman spectroscopy based on the Markov transition field (MTF) combined with a convolution neural network (CNN)was explored for the accurate and efficient detection of microplastics. The Raman spectra of eleven types of microplastic samples were collected, and 100 spectra were collected for each sample; then, the spectral dataset was expanded through data augmentation. The one-dimensional Raman spectral data was converted into two-dimensional images using a Markov transition field to obtain a two-dimensional image spectral dataset. A small-sized multiple-kernel-based convolutional neural network (SSMKB-CNN) model with continuous small-scale convolutional kernels is designed, including one input layer, six convolutional layers, two pooling layers, one flattened layer, two fully-connected layers, and one output layer. The classification performance of the model is compared with the classification results of AlexNet, VGG16, and ResNet50 deep convolution neural network models based on a two-dimensional MTF image spectral dataset, along withthree classical machine learning classifiers based on a one-dimensional spectral dataset, including K-nearest neighbors (KNN), random forest (RF) and support vector machine (SVM)with three kernel functions (rbf, Linear and Poly). It could be seen from the training curves and the classification confusion matrix that the loss and accuracy curves of the four CNN models are stable and can achieve a good training effect. The accuracy rate of the proposed SSMKB-CNN model reaches 97.04%, and the macroprecision rate, recall rate, and F1-score are 97.05%, 95.06%, and 97.02%, respectively, which is superior to the other three CNN models used for comparison and the three machine learning classifiers. Each training epochconsumes 9 seconds, less than the three CNN models. Overall, the proposed SSMKB-CNN model has the best classification performance. The experimental results show that the Raman spectrum and SSMKB-CNN model combined with MTF can accurately and efficiently extract spectral features and make precise predictions, and the qualitative identification of microplastic samples using the Raman spectrum is realized. It can provide a method reference for the recognition technology of actual microplastic contaminants in seawater.

Milk is favored due to its high nutritional value and consumption rate. Authenticity is a common concern for value assessment. Recently, non-invasive and rapid identification methods have been preferred for the dairy industry. This work proposed a quick method using synchronous fluorescence (SF) spectroscopy and a support vector machine (SVM) for the identification of raw milk. With this aim, SF spectra of milk were recorded between 220 and 600 nm excitation range with Δλ of 10 to 180 nm, in steps of 10 nm. All the milk showed the same fluorescence excitation at band position 280 nm, which corresponded to tryptophan. However, the fluorescence intensity of pure milk at this location was significantly higher than that of the two types of milk powder, and skimmed milk powder was stronger than whole milk powder. It indicated that the same main components were in milk. However, there were differences in their concentrations by different treatment methods. Two types of reconstituted formula milk were differentiated based on intensity variations at wavelengths 350~400 and 450~500 nm. The excitation at these wavelength positions corresponds to vitamin A and carotenoids. At these bands, the skimmed milk powder had a stronger fluorescence intensity in the corresponding region than whole milk powder, mainly due to the scattering of fatty substances, which enhanced the fluorescence intensity. Parallel factor analysis (PARAFAC) was found to reduce three-dimensional SF spectroscopy to two-dimensional data, resulting in a better understanding of the characteristics of dairy products. When the suitable components were6, the maximum load value was at Δλ with 40 nm, where the difference in sample information was more significant. Then, the Δλ with 40 nm and the value of contaminated dairy products as input data were used to classify and identify adulterated milk for the support vector machine (SVM) classifier. The three SVM methods were the genetic algorithm for support vector machine (Ga-SVM), particle swarm optimization support vector machine (Pso-SVM), and grid search algorithm (Grid-SVM). The results showed that the optimal classification accuracy for the Grid-SVM mode training set, test set, and cross-validation (CV) accuracy were 100.00%, 100%, and 98.91%, respectively, with a model running time of only 6.724 seconds. The study demonstrated that SF spectroscopy with PARAFAC and SVM methods is a promising tool and can potentially become a rapid and nondestructive analytical technique for identification of adulteration milk.

Raman spectra of pyromellitic dianhydride crystal have been measured from ambient to 24 GPa, and the Raman vibrational modes of pyromellitic dianhydride have been assigned. Based on the pressure dependence of Raman mode frequencies, it can be observed that a phase transition of pyromellitic dianhydride occurred from Phase Ⅰ to Phase Ⅱ at 2 GPa. In comparing the slopes of the pressure dependence of the Raman vibrational modes between Phase Ⅰ and Phase Ⅱ, it can be found that most of the slopes of Phase Ⅰ were smaller than Phase Ⅱ. The results indicate that the sensitivity of functional groups of pyromellitic dianhydride to pressure decreases with increasing pressure, and the molecular crystal of pyromellitic dianhydride has a more compact stacking structure in phaseⅡ. Besides, we also calculated the Raman intensities ratio of the overtone of C—O stretching vibration mode (located at 1 837 cm-1) and the fundamental of CO stretching vibration mode (located at 1 865 cm-1) and the discontinuities of the Raman intensities ratio versus pressure was observed. This behavior confirms that the pyromellitic dianhydride crystal has undergone a phase transition at 2 GPa, as mentioned above.

The identification of wetland vegetation using hyperspectral data has traditionally been one of the focuses of vegetation remote sensing research. Hyperspectral remote sensing data contains more detailed spectral features of vegetation, providing a powerful means for identifying hyperspectral vegetation. In this paper, the hyperspectral data of three typical wetland vegetation species, Zizania latifolia; Phragmites australis Salvinia natans; They were measured as target samples in the study area of the east coast of Erhai Lake. The original spectra were transformed by first-order differentiation and envelope removal and analysed for their spectral features. The feature wavelengths in the original spectra and their transformed spectra were selected using two feature variable selection algorithms, namely, successive projection (SPA) and competitive adaptive reweighted sampling (CARS), and the support vector machine (SVM) and random forest (RF) were finally established based on the full-wavelength data as well as the feature wavelengths after the selection-Recognition models. The results show that both SPA and CARS algorithms have a good dimensionality reduction effect on hyperspectral data, and the number of selected feature wavelengths is between 5 and 18. Comparing the combination of different spectral transform processing and feature wavelength extraction methods for modelling experiments, the envelope removal-SPA-SVM model performs the best in identifying the three types of target samples, with a recognition accuracy of 0.937 5. At this time, the number of feature wavelengths selected for input modelling is only 10, which accounts for 4.7% of the full wavelength range, which greatly reduces the models computation time and of the selected feature wavelengths, 70% of the selected characteristic wavelengths are located in the characteristic absorption bands. Their distribution can better reflect the spectral absorption characteristic law caused by the differences in the chemical composition of vegetation. The experimental results show that the hyperspectral vegetation identification modelled by spectral transformation and feature selection is feasible and can provide a reference for other wetland vegetation identification methods.

The accurate measurement of dissolved oxygen concentration plays a crucial role in medical applications, marine monitoring, industrial and agricultural production, and other fields. A ratiometric optical fiber dissolved oxygen sensor is proposed. Organic modified silicates (ORMOSILs) using tetraethoxysilane (TEOS) and triethoxyoctylsilane (Octyl-triEOS) as precursors as carrier matrices, tris(4,7-diphenyl-1,10-phenanthrolin) ruthenium(II) dichloride complex (Ru(DPP)2+3) as the oxygen-sensitive dye, 7-amino-4-(trifluoromethyl) coumarin (AFC) as the reference dye. The absorption spectrum indicates that the oxygen-sensitivedye and reference dye can be excited by a light source with a central wavelength of 405nm. The emission spectrum indicates no spectral overlap between the emission wavelengths of the oxygen-sensitivedye and the reference dye so that the ratio method can measure the dissolved oxygen concentration. ORMOSILs prepared by -the sol-gel process fix oxygen-sensitive dyes and reference dyes on the end of plastic optical fiber to form composite oxygen-sensitive films. The thickness and hydrophobicity of the sensing film were characterized, with a thickness of 569 μm. The water contact angle is 81°. The sensor was tested in an aqueous solution. Under the excitation of a light source, there were obvious emission peaks at 605 and 409 nm for the oxygen-sensitivedye and reference dye. Oxygen has a quenching effect on the fluorescence of oxygen-sensitive dyes. As the concentration of dissolved oxygen increases, the fluorescence intensity of oxygen-sensitive dye gradually decreases. The fluorescence intensity of the reference dye remains stable at a certain value, and the purpose of detecting oxygen concentration is achieved by measuring the ratio of oxygen-sensitive dyes to the reference dye.The linear relationship between the fluorescence intensity of the ruthenium complex and the concentration of dissolved oxygen is described by the Stern-Volmer equation. The calibration curve of the sensor was 98.22% in the range of 0~20.05 mg·L-1. The sensitivity can reach 0.433 4/unit ［O2］, and the response time from saturated oxygen solution to saturated nitrogen solution is 144s, and from saturated nitrogen solution to saturated oxygen solution is 12 s, introducing the asymmetric factor ASY to indicate the asymmetry of the sensing film. The optical stability and repeatability of the sensor were characterized. The photostability and repeatability of the sensor were tested, and the ratio type fiber optic sensor can overcome light source fluctuations and has stronger stability compared to single fluorescence intensity sensing.

Fiber identification is an important work in preserving ancient paper. Exploring a non-destructive fiber identification method of Chinese traditional handmade paper is significant to the research and preservation of ancient Chinese books, archives, and paper cultural relics. In this study, Attenuated Total Reflection Fourier Transform Infrared Spectroscopy (ATR-FTIR) was used to analyze 64 standard Chinese traditional handmade paper samples of 17 categories whose fiber compositions were known.First, the main components in the paper and the attribution of all the infrared peaks were confirmed by referring to the infrared peaks of cellulose and lignin, as well as the X-ray diffraction (XRD) analysis results of some paper samples. Then, the comparative analysis of the spectra with very high similarity was carried out to summarize the spectral characteristics of various kinds of paper in the four bands of 4 000~1 800, 1 800~1 500, 1 500~1 200 and 1 200~600 cm-1, respectively. At the same time, by second derivative processing, the second derivative spectrum characteristics of each kind of paper were summarized in the bands of 1 500~1 200 and 1 200~900 cm-1, respectively. Finally, through the numerical calculation results, such as the infrared cry stallinity index and the ratios of the peak heights and the peak areas, a more detailed distinction among different kinds of paper was achieved. The above method was used for a blind test of 16 unknown samples, and the results of the infrared analysis were consistent with the results of the microscopic fiber analysis, which preliminarily proved the feasibility and effectiveness of this method. All the experimental results show that the non-destructive analysis method of ATR-FTIR can be used to make fast and accurate identification of the fiber type of hemp paper, mulberry paper, Wingceltis paper, Xuan paper, Thymelaeaceae paper and bamboo paper, but this method still has some limitations for the identification of the plant materials With closer genetic relationships,such as mulberry and paper mulberry, which are also difficult to distinguish through microscopic analysis.

Rare earth up-conversion nanoparticles (UCNPs) have been used to construct fluorescent nanosensors because of their low toxicity, good chemical stability, and low background fluorescence. Core-shell UCNPs were prepared using the solvothermal method, and water-soluble core-shell UCNPs (Cit-CS-UCNPs) were obtained by surface ligand exchange with sodium citrate. The Cit-CS-UCNPs were used as the energy donor of the fluorescence sensor, and manganese dioxide (MnO2) nanosheets were used as the energy receptor of the fluorescence sensor. Based on the fluorescence resonance energy transfer (FRET) mechanism, a fluorescence nanosensor (Cit-CS-UCNPs-MnO2) was constructed for the detection of hydrogen dioxide (H2O2) and tert-butylhydroquinone (TBHQ) as food additives. The prepared nanomaterials morphology, structure, and properties were characterized by scanning electron microscopy (SEM), fluorescence spectrum, and ultraviolet spectrum (UV-vis). The effects of quenching agent concentration, incubation temperature, and incubation time on the detection performance of the fluorescence sensing system were investigated. According to the fluorescence spectra and UV-Vis experimental results, the maximum emission peak of Cit-CS-UCNPs is 654 nm. After the combination of Cit-CS-UCNPs and MnO2, core-shell UCNPs undergo fluorescence quenching. When H2O2 is present, the fluorescence of Cit-CS-UCNPs recovers. The results indicate that H2O2 and MnO2 nanosheets undergo redox reaction at this band, and MnO2 nanosheets are reduced to Mn2+, which gradually dissociates from the surface of Cit-CS-UCNPs. In the presence of TBHQ, the peak of the Cit-CS-UCNPs-MnO2 and TBHQ system was shifted to 253 nm, indicating that the redox reaction occurred between TBHQ and MnO2 nanosheets, and the FRET effect was reduced in the Cit-CS-UCNPs-MnO2 system, and the fluorescence intensity increased. It can be seen from the SEM results that MnO2 nanosheets are uniformly coated around Cit-CS-UCNPs, and maintain good dispersion in water, indicating that MnO2 nanosheets are modified on the surface of Cit-CS-UCNPs. The concentration of the quencher potassium permanganate (KMnO4) was optimized, and the results showed that when the concentration of KMnO4 is 10 mol·L-1, the quenching efficiency can reach 90%. The detection conditions were optimized, and the results showed that when the incubation time of H2O2 was 25 min, the redox reaction between MnO2 and H2O2 was complete. The fluorescence recovery value of Cit-CS-UCNPs-MnO2 was the maximum. The incubation time of TBHQ was 30 min. Under the optimal experimental conditions, the fluorescence intensity of Cit-CS-UCNPs-MnO2 has a good linear relationship with the concentration of H2O2 (0~1 000 μmol·L-1) and TBHQ (0~0.6 mmol·L-1). The optimal experimental conditions were maintained, and representative metal ions (such as K+, Na+, Ca2+ and Mg2+) and common food additives (BA, Glu, PS, Suc, Nat and Ino) in food were selected as research objects. The results showed that compared with H2O2, Cit-CS-UCNPs-MnO2 did not react strongly to other added substances, and the overall fluorescence signal of the sensor did not fluctuate much. Therefore, it can be seen that Cit-CS-UCNPs-MnO2 can detect the specificity of H2O2 and TBHQ.

Ethyl carbamate，A group 2A carcinogen, is commonly found in fermented foods such as grape, soy sauce, and pickles. Due to its toxicity and carcinogenicity, it can accumulate in organisms and produce a variety of lesions and has become a potential threat to the modern food processing industry. The traditional detection method of ethyl carbamate has the disadvantages of a complicated sample pretreatment process, long analysis time, expensive equipment, high requirements for testers, and the inability to detect large samples quickly. Therefore, there is an urgent need to develop a rapid, accurate, and efficient method for urethane detection. Surface-enhanced Raman spectroscopy (SERS) technology has the characteristics of high sensitivity, high selectivity, resistant photo bleaching, short response time, and non-destructive. SERS overcomes the problems of Raman spectroscopy, such as weak scattering signals and large background interference, and can realize the rapid qualitative and quantitative detection and analysis of trace substances, showing good application prospects in the territory of food safety detection. In this study, surface-enhanced Raman spectroscopy was used to detect the content of ethyl carbamate in grape spirit rapidly. Silver nanosphere silicon cone array and gold-silver alloy nanosphere silicon cone array were used as SERS substrates. The effects of SERS substrate structure, mixing time, storage time, and excitation wavelength on Raman scattering signals were investigated, and the optimal SERS substrate was determined. The results show that good Raman signals can be obtained when the excitation wavelength is 785 nm, the mixing time is 60 min, and the thickness of silver nanoparticles is 10 nm using Ag-Au/SiNCA as the SERS-enhanced substrate. The characteristic peak position of 1 442 cm-1for quantitative analysis of ethyl carbamate in grape spirit was determined using density functional theory, and a linear equation was established between the peak intensity of the characteristic peak at 1 442 cm-1and the EC concentration. The equation has a good linear relationship in the concentration range of 1×10-3～1×10-8 mol·L-1, and the coefficient of determination R2=0.821 3. The recovery rate of the method is between 83.06% and 110.00%, and the minimum detection limit can reach 3.28×10-8 mol·L-1. This detection method is very simple and quick to operate, and analyze. If this method applies EC in grape spirit, an effective and rapid detection method can be provided with ideas and references.

Laser-micro confocal Raman spectroscopy is widely used to study molecular structure, bonding effects, loading, and internal stress. The signal intensity and quality of Raman spectra are particularly important for the analysis. However, Raman laser irradiation on the surface of sensitive samples produces thermal effects and, thus, structural damage. The damaged parts of the sample collapse and shine. To obtain high-quality Raman spectra and realize non-destructive characterization, this paper innovatively proposes adding KBr as a heat sink to eliminate the thermal effect of laser. A series of carbon materials that are easily damaged under laser radiation were selected as the research objects, and the relationships between the carbon content of the samples, the addition of KBr, and the Raman spectral parameters and signal intensity were explored. The experimental results show that adding KBr can effectively inhibit the sensitive samples from being damaged by the Raman laser. When the addition ratio of KBr is 1∶2, accurate and nondestructive Raman spectroscopic detection of this material can be realized. In addition, the graphitization R values of carbon materials detected in different microregions are very close to each other; with the increase of carbon content, the position of the D peak is shifted to the low-wave region by ～3 cm-1, and the position of the G peak is shifted to the high-wave region by ～2 cm-1, and the difference between the peak positions of the D peak and the G peak is increased, with a decrease in the intensity ratio, and an increase in the ordering of the samples.

Surface-enhanced Raman spectroscopy (SERS) derived from surface plasmon resonance (SPR) has become an effective tool for surface analysis because of its high sensitivity and surface specificity. Moreover, SPR can induce and stimulate the catalytic reaction on the surface of noble metal nanostructures. The combination of SERS and SPR is beneficial to in-situ monitoring of SPR catalytic reaction at the surface interface of noble metals. This paper used p-nitroiodobenzene (PNIB) as a probe molecule to demonstrate SERSs capability in characterizing catalytic coupling reactions. The effects of laser power, wavelength, and substrate on the catalytic coupling were investigated, and the mechanism of the coupling reaction was proposed preliminary. The results revealed that azo compound 1,2-bis(4-iodophenyl)diazene was produced by SPR-catalyzed coupling reaction of PNIB on the surface of gold nanoparticle monolayer film (Au MLF). The equilibrium of the reaction was reached with short duration by increasing laser power, i.e., from 200 s at 3 mW to about 50 s at 15 mW. The coupling efficiency was increased to about doubled. On the Au@Ag nanoparticle film, the coupling efficiency is increased by about 10 times, and the coupling efficiency is significantly dependent on the excitation wavelength and SERS substrates. The reaction activity order of different substrate surfaces is Au@Ag (532 nm) > Au@Ag (638 nm) > Au (638 nm). It demonstrated that the hole-trapping agent sodium sulfite increase the coupling efficiency of the reaction by about an order of magnitude. It indicated that the reaction of PNIB coupling to 1,2-bis(4-iodophenyl) diazene catalyzed by SPR is a reduction reaction induced by hot electrons.

Zeolites are a class of solid acid catalysts with a wide range of applications in the current catalytic field due to their regular pore structure, large specific surface area, and efficient acidic properties. The catalytic reactivity of zeolites is closely related to their acidity. Accurately characterizing the acidic properties of zeolites is of great significance for establishing the structure-activity relationship between zeolite structure, acidity, and catalytic reaction performance. One of the most effective methods to characterize the type of acid centers and acid strength of zeolites is in situ infrared spectroscopy using pyridine as a probe molecule.First, this paper describes the experimental principles of in situ FTIR-pyridine adsorption method for characterizing the acidic properties of solid acid catalysts. Then, using ZSM-5 as a model, the in situ FTIR-pyridine adsorption was used to optimize the testing conditions for identifying the surface acidity of the zeolite and the effects of experimental conditions such as activation temperature, activation time, adsorption time of pyridine, desorption temperature and desorption time on the relevant FTIR characteristic peaks were investigated.The results showed that the optimal experimental conditions for the interaction between pyridine and acidic sites on zeolites were: activation temperature of 400 ℃, activation time of 60 min, adsorption time of 10 min at room temperature, desorption temperature of 150 ℃, and desorption time of 30 min. Under these experimental conditions,the pyridine was efficiently adsorbed with the acidic sites on the zeolites, and the intensities of the FTIR characteristic peaks corresponding to Brnsted and Lewis acids were saturated. Simultaneously, the interference of physical adsorption, hydrogen-bonded adsorption, and contaminants adhered to the sample surface on the adsorbed pyridine was effectively eliminated, and the optimal FTIR spectra were obtained with high repeatability. Finally, three modified zeolites, including Fe-ZSM-5, HZSM-5, and Na-HZSM-5, were characterized by optimized experimental methods, all of which obtained FTIR spectra with excellent quality, and the acid amount ratios of Brnsted acid to Lewis acid in agreement with the reports. The method improves the test efficiency and success rate. It excludes the relevant interference so that the measured information on the zeolites acidity category, strength, and relative content is more accurate and reliable. Meanwhile, the optimized experimental method provides a reference for characterizing the acidic properties of other solid acid catalysts, which is of great significance in guiding the preparation and mechanism research of solid acid catalysts.

This paper proposes a simultaneous quantitative analysis method for Pb and As in soil based on the spectral peak characteristics of Lα and Lβ of the Pb element and Kα and Kβ of the As element, aiming at the difficulty of accurately quantifying Pb and As in soil simultaneously by energy-dispersive XRF spectrometry. The method constructs the relationship between the characteristic peak intensities of Lα and Lβ of Pb element at different concentrations, accurately analyzes the characteristic peak intensities of Lα of Pb element and Kα of As element according to the overlapping peak intensity information of Lα of Pb element and Kα of As element in the measured sample, and realizes the accurate back calculation of Pb and As in the soil sample based on the established quantitative analysis curve. Through quantitative detection of Pb and As coexisting soil samples with different concentrations, compared with the results measured by ICP-MS, the relative error of this method for detecting 20.25～844.84 mg·kg-1 heavy metal Pb is between 0.86% and 11.09%, with an average relative error of 6.17%; the relative error for detecting 26.99～825.93 mg·kg-1 heavy metal As is between 0.10% and 14.72%, with an average relative error of 8.11%. This method achieves the simultaneous accurate quantitative analysis of Pb and As contents in soil coexisting with Pb and As, and provides a method for developing rapid XRF precision detection equipment for soil heavy metals on site.

The identification of the sequence of seal stamps and ink has always been a hot and difficult issue in the research and practice of forensic document examination, which could prove the authenticity of the documents and infer whether the documents are consistent with the facts of the case, thus providing a legal basis for the judicial department to investigate and judge the case. The common identification methods include optical microscopes, fluorescence, document inspection, and computer software-assisted methods. The identification conclusion is easily influenced by subjective experience and is questioned due to its lack of objectivity. In this study, based on the preparation of 54 handwriting samples of 3 types of signature pens and 9 types of seal stamp combinations, 18 printed samples of laser printed, and 9 types of seal stamp combinations, the samples were tested by Confocal Microscope Raman Spectrometer. The optimal condition was obtained from the handwriting samples (10# Pilotg-1) and seal stamps (1# Print show). The 72 prepared samples were placed in a dark storage box for one year. Under this optimal condition, the Raman spectrum of four measuring sites in the sample, followed by ink before seal stamps, seal stamps before ink, seal stamps, and ink, were measured, and each sample was tested three times. The experimental results were unaffected by storage environment or paper background factors. Whether handwriting or printed samples, the intersection of ink before seal stamps displayed some of the characteristics of the seal stamps. At the same time, the intersection of seal stamps before ink displayed some of the characteristics of the ink. The Raman spectra of the intersection showed most of the Raman shifts of the corresponding seal stamps or ink on the upper layers, consistent with the observation of μ View under a 20× objective lens. Raman shift of the sample was compared to identify the crossing sequence of seal stamps and ink of handwriting/printed. The results show that the novel method is convenient, quick, non-destructive, highly sensitive, and accurate and has practical guiding significance for identifying the crossing sequence of seal stamps and ink.

Soil is the material basis of human survival; its characteristics closely relate to peoples production and life. Traditional soil heavy metal detection methods such as atomic absorption spectroscopy and inductively coupled plasma mass spectrometry analysis are weak and expensive, so the development of low-cost operating soil elements quantitative analysis method at the same time. Laser-induced breakdown spectroscopy (LIBS) technology has been widely used because of its rapid and multi-element simultaneous analysis. However, because it is not easy to carry, a split-type field LIBS detector was developed to meet the field testing needs. Its design is to divide the instrument into two parts, probe head, and chassis, and connect it through a plastic pipe. Using a miniature diode pump laser, the pulse energy is up to 100 mJ, with a wavelength of 1 064 nm. The repetition frequency is 1~10 Hz. In addition, using a multichannel high-resolution spectrometer improves LIBSs analytical performance. FPGA is used to realize the us-level delay time function to reduce radiation background interference. To obtain spectral data in 11 soils, The pulse energy was 100 mJ, The delay time was set to 1us, Integration time of 2 ms, Spectra from 10 different positions were collected for each sample, Each position was measured 20 times, A total of 200 spectral data were collected, To reduce the noise interference, The spectral data for each sample were mean-preprocessed after the Beads algorithm baseline correction, The three principal component components with the largest contribution rate were obtained using PCA principal component analysis, In the clustering analysis of 11 different types of soils in different regions by the Kmeans++ algorithm, of the same category of soil into the partial least squares (PLSR) algorithm, Each element selects two characteristic lines and 10 points to enhance the spectral information, One sample was selected as a prediction for quantitative analysis of five soil heavy metal elements, Cu, Cr, Ni, Co, and Cd. the results show that, In contrast to that where no cluster analysis was performed, This method can significantly improve the fitting correlation coefficient of the elements, The correlation coefficients of the five heavy metal elements increased from 0.953, 0.992, 0.989, 0.982, 0.99 to 0.999, 0.998, 0.999 5, 0.996 5, 0.993, respectively, The correlation coefficient of 0.99 and above all meet the requirements of LIBS linear analysis, The average relative error between the prediction results and the actual content increased from 83.45%, 16.03%, 22.94%, 43.91%, 125.768% to 1.14%, 0.99%, 0.895%, 1.879%, 1.862%, respectively, It can be found that after the cluster analysis, Its prediction error is greatly reduced, All were within 5%, With a relatively good analytical performance, The correlation coefficient and prediction error of the five elements are improved compared with the direct PLSR method. Combining PCA and Kmeans++ can be more accurate clustering after dimension reduction, reduce the influence of noise and redundant information, speed up the calculation, reduce the influence of abnormal points on the clustering effect, and improve the robustness.

Qinghai-Tibet salt lakes are famous for enriching boron and lithium resources. Nevertheless, the chemical species of borate in the brine varies with the chemical type of salt lake. Among them, the existing borate forms in sulfate-type salt lake brine are the most complicated. Generally, the borates do not crystallize out from the brine during the whole evaporation process of brine but accumulate in the bischofite-saturated brine in different kinds of boron species, which are supersaturated with magnesium borates. This phenomenon may significantly impact the subsequent separation and extraction of lithium and magnesium salts. Therefore, the deep research on the chemical forms, species distribution, and their interactions in the salt lake brine is of great significance for the highly efficient development of salt lake resources. Compared with the classical Raman spectroscopy, the simplified Raman integrating sphere, designed based on the Raman scattering principle, can improve the exciting lights efficiency and the Raman scattering signal. It is characteristic of a strong Raman scattering signal, low detection limit, and high signal-to-noise ratio for the characterization of the borate structures, which favors the quantitative analysis of the chemical forms of the borate in the complicated brine system. Based on the above, this study aimed to investigate the chemical forms of borate in salt lake brine using the Raman integrating spheres. It also elucidated the changes of polyborate ions during the brine evaporation process. Secondly, the response surface method was used to explore the effects of the coexisting salts on the determination of B(OH)3 in salt lake brine. The results showed that borates in the salt lake brine could be polymerized to form poly borate ions such as B3O3(OH)-4 and B6O7(OH)2-7 during the brine evaporation process, which agreed well with the borate changes in the alkaline-earth metal solution system of MgCl2-MgO-2B2O3-H2O, but differed greatly with that changes in alkaline metal solutions. The relative error of the B(OH)3 determination in brine was less than 5% after being corrected by the response surface interference model. Therefore, the distribution of B(OH)3 in the brine was also studied during the evaporation process, which helped explain the polymerization mechanism among borate ions in brine from a quantitative perspective. In sum, this research could provide new ideas and methods for further study of borate speciation and their interaction mechanism in complicated brine systems.

Novelty luminous carbonized polymer dots (CPDs) were synthesized by rapid and facile hydrothermal carbonization of sodium alginate (SA) and o-phenylenediamine (PDA). The SA-oPDA CPDs reveal a large Stokes shift up to 110 nm under the excitation wavelength at 352 nm and emission wavelength at 462 nm. The synthetic CPDs exhibited bright blue fluorescence with a fluorescence quantum yield of 16.9% and excellent photostability and excitation wavelength independence. Just for the characteristics of the alginate polymerization chain, the SA-oPDA CPDs also demonstrated the practical feasibility of establishing the fluorescent films by self-assembled chitosan polyelectrolyte molecular with electrostatic adsorption. Thus, it has great potential application in light-emitting devices. Meanwhile, the SA-oPDA CPDs were found to have extraordinary amphoteric charge adjustability by pH value. With the change of solution environment from acidic to alkaline, the position of the fluorescence emission peak changed from 465 to 425 nm, and the surface charge of the carbon point changed from positive charge to negative charge. The structure and morphology characterization showed that during the high-temperature carbonization process, the polymer is carbonized to form a luminescent carbon core and contains hydroxyl, amine, and carboxyl on the surface of CPDs. Due to the active protonation/deprotonation of amino or carboxyl groups, the CPD solution exhibits luminescence tunability with pH alterations. The SA-oPDA CPDs exhibited an amphoteric nature with the isoelectric point between pH of 6 and 7, respectively. After the CPDs were combined with lead ions in the solution environment at pH 6.80, the fluorescence emitted by the solution changed from blue to blue-green, which could realize an obvious visual response to a certain concentration of lead ions. The experimental results also showed that the reduced drug, especially quercetin(Que), had a significant quenching effect on the fluorescence signal of CPDs. The degree of fluorescence quenching had a good linear relationship with the concentration of quercetin in the range of 7.9×10-6~7.7×10-5 mol·L-1. The equation was I0/I=0.878 8+1.689 7×10-6cQue, correlation coefficient r2=0.974 8, and the LOD of detection was 1.7×10-6 mol·L-1.

The radial velocity method is quite effective in discovering and characterizing exoplanets based on the radial velocity change of the target stellar. It plays an important role in the detection of exoplanets. Owing to stellar activities, the differences between spectra and template, and the noise in spectra for other reasons, radial velocities calculated by the cross-correlation function algorithm may have some errors. This paper proposes a method of measuring stellar radial velocities, and an attention mechanism is integrated. Observed spectra are processed to remove noise from the spectra, and radial velocity is calculated based on the periodicity of stellar radial velocity. First, Gaussian Process Regression is used to establish spectral models, which helps reduce the influence of noise and get more precise spectra. A subset of data is used to reduce the computation cost. Then, based on the idea of attention mechanism, different weight is assigned for the absorption lines to figure out differential radial velocities between spectra. Finally, radial velocities of stellar are obtained using the relationship of differential radial velocities. This paper analyzes the effect of the signal-to-noise ratio and the number of spectra on the mean error of radial velocities. The experimental results show that compared to the cross-correlation function algorithm, the stellar radial velocity calculation method combined with the attention mechanism can effectively reduce the error of radial velocities when the signal-to-noise ratio is low. Increasing the number of spectra helps improve the accuracy of radial velocities to some degree. Finally, The spectra of HD85512 are analyzed. Compared with other algorithms, the algorithm proposed in this paper reduces the error of radial velocities and greatly improves the accuracy.

Color is one of the important optical properties of gemstones, and it is also an important parameter affecting their quality and price. As a common jade variety, the complex structure, color ions, valence states and coordination of yellow-green prehnite have led to great controversy about its color genesis. In this paper, the species, valence, and coordination state of the main chromogenic ions of two typical yellow-green prehnite samples (light yellow-green sample LYG and dark yellow-green sample DYG) were investigated using modern analytical methods such as Fourier infrared absorption spectroscopy (FTIR), ultraviolet-visible spectroscopy (UV-Vis), laser-exfoliation-plasma mass spectrometry (LA-ICP-MS) and Mssbauer spectroscopy, which provides a scientific explanation for the color genesis of yellow-green prehnite. UV-Vis analysis indicates that both sample LYG and sample DYG showed obvious absorption bands at ~425 and ~585 nm, which were related to the electron leap of Fe3+oct and the charge transfer of Fe2+ch-Fe3+oct, respectively. Ultrafine parameter analysis of Mssbauer shows that the IS of Fe3+ in both sample LYG and sample DYG is 0.34 mm·s-1, and the QS values are 0.22 and 0.35 mm·s-1, respectively, both consistent with the characterization of Fe3+ in the octahedral ligand of prehnite. Sample LYG (IS=1.08 mm·s-1) and sample DYG (IS=1.07 mm·s-1) have similar IS values for Fe2+, both of which are consistent with the characterization of Fe2+ in octahedral coordination. However, the QS value of Fe3+ in sample DYG (QS=2.78 mm·s-1) is significantly higher than that of the light yellow-green sample LYG (QS=1.12 mm·s-1), suggesting that Fe2+ is in the distorted octahedron in the former structure. The content of Fe and the proportion of Fe2+ in sample DYG (4.04 wt%; 11.88%) are higher than those of the light yellow-green sample LYG (3.55 wt%; 5.27%), suggesting that the yellow-green color of prehnite may be related to the content of Fe and the proportion of Fe2+. The contents of V in sample LYG and the dark yellow-green sample DYG are 633 and 1 810 μg·g-1, respectively, indicating that V may contribute to the yellow-green color of prehnite. Still, its content is significantly lower than Fe, so the contribution to the color is much lower than Fe3+ and Fe2+. The present study not only identified the main chromogenic ion species (Fe) of yellow-green prehnite, but also determined the valence states (Fe3+ and Fe2+) and coordination states (octahedral coordination) of the chromogenic ions, which provides a solid theoretical basis for the quantitative study of the color of yellow-green prehnite.

Sphene is known as titanite, which is a transparent neosilicate mineral that can be used as a gem. Other elements can substitute the cations Ca2+, Ti4+, and Si4+ in sphene. Yellow sphene was selected for heat treatment to explore the heat treatment process of sphene. The results showed that the yellow sphene samples were heated in an oxidizing atmosphere at 700 ℃ and turned brown-red above, whereas, the reducing atmosphere heated again to turn yellow again. The composition analyzed by the electron microprobe analysis (EMPA) shows that the samples are consistent with the standard sphene composition, and also contain a small amount of FeO (average 0.601 wt%) and Al2O3 (average 0.210 wt%). The mid-infrared reflectance spectra 1 600～400 cm-1are consistent with standard titanite, showing absorptions attributed to O—Si—O and Si—O vibrations. In the mid-infrared transmission spectra 4 000～2 000 cm-1, the absorption is caused by O—H stretching vibration and accompanied by 3 450 cm-1 centered absorptions broadband. In the near-infrared transmission spectra 10 000～4 000 cm-1, the absorption is caused by O—H stretching vibration. The UV-visible absorption spectrum illustrates that the yellow color of the untreated titanite sample is due to the substitution of Fe2+as an isomorphic counterpart for Ti4+ in the octahedral site, leading to the charge transfer forms of Fe2+ to Ti4+ (IVCT), generating an absorption band at 420 nm. The charge transfer between O2-→Fe2+ and O2-→Fe3+ causes the strong absorption towards the ultraviolet at 450 nm. The brownish-red color after oxidation treatment is attributed to the partial conversion of Fe2+ to Fe3+, resulting in the overlapping of the strong absorption edge at 540 and 450 nm towards the ultraviolet respectively, caused by the charge transfer forms of Fe2+ to Fe3+ (IVCT).

Serpentine is one of the earliest jade used in China. Identifying the origin of serpentine jade is of great significance to understanding the development of Chinese ancient jade culture and rebuilding the ancient jade trade route. However, due to the large number of sources of serpentine jade, there is still no proven technology to identify the geographic determination of serpentine jade. In this paper, serpentine jade from Hanzhong, Shaanxi; Dunhuang, Gansu; Luanchuan, Henan; Xiuyan, Liaoning; Taian, Shandong, and Wushan, Gansu were studied. A Linear Discriminant Analysis (LDA) model was established based on Principle Component Analysis（PCA） of 200 high-quality Raman data collected from 66 samples. The results show that the mineral compositions of serpentine jade from the six regions differ. The main mineral compositions of serpentine jade in Hanzhong are chrysotile and lizardite. Dunhuang serpentine jade is a homogeneous mixture of chrysotile and serpentite. The main mineral composition of Taian serpentine jade includes lizardite (Mojade) and antigorite (Bijade and Cuibanjade). The main mineral composition of serpentine jade in Luanchuan, Xiuyan, and Wushan areall antigorite. Under the premise of strictly controlling the experimental conditions, the Raman spectrum data combined with PCA+LDA analysis can distinguish serpentine jade from different origins. The established LDA models correct rate of origin discrimination is 96.25% and 92.50% for training and test data, respectively. There sult shows the potential value of tracing the origin of serpentine jade by nondestructive Raman spectroscopy. Combining Raman spectroscopy data with statistics or machine learning methods to build a discriminant model may be a new technical path to solve the serpentine jade origin traceability bottleneck.

Spectral noise removal is a necessary process for remote sensing regional applications, and the noise removal effect can directly affect the monitoring accuracy of regional surface information. To study and analyze the decomposition mechanism of the discrete wavelet algorithm on spectral data and explore the spectral noise information removal and spectral processing method based on the discrete wavelet algorithm, this study takes the winter wheat canopy spectra and leaf water content as the data source, and then denoises the spectral data using the discrete wavelet algorithm with the wavelet base of Meyer; and then separates the information of the denoised spectral data by using the wavelet bases of Meyer, Sym2, and Coif2, and constructs the spectral data by combining the correlation analysis algorithm and partial least squares algorithm. Then, we separated the information of the denoised spectral data with Meyer, Sym2, and Coif2 as the wavelet bases and constructed a model for estimating the water content of winter wheat leaves by combining the correlation analysis algorithm and partial least squares algorithm. The studys conclusions are as follows: (1) Under the discrete wavelet algorithm, with the increasing number of merging scales of the merging spectral curves, the original spectral curves local large, medium, and small features were highlighted in order. The correction amplitude of the merging curves was also gradually reduced with the joining of the decomposition scales of H10—H1. With the sequential addition of H10—H1 decomposition scales, the magnitude of the correction of the decomposition information to the merged curves is also gradually weakened, in which the merged spectral curves are almost unchanged after the sequential merging of H3—H1. (2) The denoising method proposed in this paper can change the sensitivity of the spectra to the water content of winter wheat leaves and the band positions of the sensitive bands to a certain extent: in the 1~3 scale, the sensitivity of the spectra to the water content of winter wheat leaves is reduced, and the distribution of the band positions of the sensitive bands is changed. Within 4~10 scales, it can significantly enhance the sensitivity of the spectrum to the water content of winter wheat leaves (Coif2); the denoising method proposed in the study can enhance the sensitivity of the local bands to the water content of winter wheat leaves (Sym2). (3) The denoising method proposed in this study can significantly improve the stability of the spectrum to the model. It can improve the accuracy and stability of the optimal model within the Sym2 and Coif2 wavelet bases, in which the validation accuracy is improved by 8.6% (Sym2) and 34.1% (Coif2), which indicates that the denoising treatment proposed in this study is effective.

Soil texture affects many physical, chemical, biological, and hydrological characteristics and processes, such as vegetation distribution, soil and water conservation capacity, and microbial activity. Accurate acquisition of soil texture is of great significance for wetland ecological restoration and protection. Based on 57 measured surface soil texture and visible-near-infrared hyperspectral data in Tianjin coastal wetland, the soil samples were smoothed by S-G and transformed by first derivative (FD), reciprocal transformation (RT), reciprocal first derivative (RTFD), square root (SR), square root first derivative (SRFD), logarithm of reciprocal (LR) and logarithm of reciprocal first derivative (LRFD),the characteristics and correlations of spectral curves of different soil texture categories were analyzed. A competitive adaptive reweighting algorithm (CARS) was used to select the characteristic bands, and partial least square regression (PLSR), random forest regression (RFR), and support vector machineregression (SVR) algorithms were combined to compare the modeling effects of different spectral transformations. The results show that: (1) The texture categories of wetland soil are mainly silty loam and silt. The spectral reflectance of silt is the highest in the 400~2 400 nm band, and the spectral reflectance of sandy soil is the lowest in the 400~2 000 nm band. The correlation between the spectral reflectance of FD, RTFD, and SRFD and the soil particle size content has significantly increased. The absolute value of the maximum correlation coefficient is above 0.58, and the highest is 0.70. (2) The feature band number of eight spectral transforms screened by the CARS algorithm is 1.05%~6.15% of the total band number, effectively reducing the information redundancy of spectral data. (3) Compared with the three estimation models for particle size content, the SVR model of SRFD and RTFD spectral transformation had the best accuracy and was superior to the other two models, the clay (SRFD) test set (R2=0.72, RMSE=1.86%, nRMSE=11.33%), the silt (SRFD) test set (R2=0.72, RMSE=2.82%, nRMSE=7.30%) and the sand (RTFD) test set (R2=0.71, RMSE=5.75%, nRMSE=5.91%). The results of this study can provide a basis and technical support for the accurate monitoring of soil texture in coastal wetland areas with hyperspectral data.

The development of high-value utilization paths for agricultural and forestry waste was highly consistent with the major strategic demand to “further promote green and low-carbon energy transformation”. In this paper, the objective was to investigate the pathway of propanediol (PG) and diethylene glycol (DEG) liquefaction catalyzed by phosphoric acid of cellulose from corn stalk at atmosphere pressure, aiming at understanding the mechanism of lignocellulosic biomass liquefaction reaction under the action of acid-catalyzed polyhydroxy alcohols. The chemical groups, molecular weight and distribution, molecular structure, and pyrolyzation of cellulosebiofuels were analyzed by Fourier transform infrared spectroscopy (FTIR), nuclear magnetic resonance spectra (NMR), gelpermeation chromatography (GPC), and thermo gravimetric analysis (TGA). FTIR showed that the biofuels had similar FTIR characteristics. At the early liquefaction stage, cellulose degradation produced more hydrocarbons, ethers, and carbonyl compounds. At the later stage of liquefaction, the carbohydrate degradation products, hydroxyl or olefin in cellulose, reacted with PG/DEG to generate organic matter insoluble in 1, 4-dioxane. GPC explained that with the progress of the reaction, the breakage degree of the cellulose molecular chain would be aggravated, and more and more low molecular weight (LMW) substances were generated by degradation. However, when the reaction time reached a certain value, the degradation products reacted with PG/DEG to produce larger molecular weight substances, resulting in the molecular weight of biofuel no longer being reduced. Results from 1H- and 13C-NMR presented that cellulose was degraded under liquefaction, and the molecular chain was broken, but part of the glucose structure was still preserved. With the reaction, these structural units were transformed again to produce LMW compounds. When the reaction continues, polymerization reactions could occur between these products or with PG/DEG, forming new substances with consistent structures and gradually stable properties. The results of TGA showed that the biofuel contained 70%~85% compounds with carbon number less than 25 and 5%~10% compounds with carbon number more than 25. In conclusion, this paper revealed the liquefaction reaction process of cellulose by studying the structural changes in the liquefaction process of polyhydric alcohols, which laid a theoretical foundation for exploring the liquefaction mechanism of corn stalk whole components.

In order to realize the real-time prediction of nitrogen content in the rice leaf population by using the rice leaf spectral index, the spectral reflectance of the top three fully expanded leaves (upper 1, upper 2 and upper 3 leaves were recorded as L1, L2 and L3, respectively) at the main growth stages of rice in cold region (T1 mid-spike differentiation stage, T2 jointing stage, T3 booting stage, T4 full heading stage and T5 wax ripening stage) under different nitrogen and variety differences in different years were collected. The change rule and the relationship between spectral index and leaf nitrogen content were explored. P-k, Root Mean Square Error (RMSE), Symmetric Mean Absolute Percentage Error (SMAPE), Root Mean Square Error of Calibration (RMSEC), Root Mean Square Error of Interactive Verification (RMSECV) and Residual Prediction Deviation (RPD) were used to verify the accuracy of the model. The results showed that with the increase of nitrogen fertilizer input, the leaf reflectance decreased in the visible region, while the leaf reflectance increased in the near-infrared platform. With the advance of the growth period, in the visible light region, the reflectance of L1 leaves of different varieties decreased first and then increased, and the reflectance of L2 and L3 leaves increased all the time. The sensitive bands of leaf nitrogen percentage were 500~550 and 650~700 nm. The correlation analysis of the spectral index and leaf nitrogen percentage content showed that the correlation coefficient of the spectral index of the following leaves was high in the early stage of growth, but it was the opposite in the later stage of growth. The L2 leaf index FD-NDNI in the T1 period, L2 leaf index GM2 in the T2 period, L2 leaf index Lic2 in the T3 period, L1 leaf index MRESRI in the T4 period, and L1 leaf index Ctr1 in the T5 period were selected as the best indexes to predict leaf nitrogen content in different periods. The regression equations R2 for predicting leaf nitrogen content were 0.54**, 0.60**, 0.66**, 0.62**, and 0.51**, respectively, which reached extremely significant levels. The P-k values of the validation indexes were 0.00, 0.04, 0.06, 0.01 and 0.04, respectively. RMSE were 0.39, 0.58, 0.22, 0.54, 2.56, SMAPE were 1.11, 1.41, 1.03, 1.64, 3.89, RMSEC were 0.17, 0.15, 0.13, 0.13, 0.13, RMSECV were 0.18, 0.14, 0.12, 0.12, 0.14, the RPD were 2.46, 2.19, 3.15, 1.74 and 3.01, respectively. Among them, the prediction effect of the L2 leaf index Lic2 at the T3 stage was the best. In summary, with the help of the selected spectral indicators, the nitrogen nutrition status of rice at different growth stages can be predicted quickly, non-destructively, and in real-time, and the sustainable development of high-yield and high-quality cold rice can be promoted.

The middle Echelle grating spectrometer has the advantages of wide spectral range and high Spectral resolution. To separate the spectra that are aliased due to the high diffraction order of the middle Echelle grating, the dispersion system of the middle Echelle grating spectrometer uses two elements that conduct cross dispersion in mutually perpendicular directions, and finally forms a two-dimensional spectral image on the detector. To obtain the intuitive spectral information of the substances detected by the spectrometer, it is necessary to convert the two-dimensional spectral image into the one-dimensional spectral image corresponding to the light intensity and wavelength through the spectral image reduction algorithm. The quality of the spectral image reduction algorithm directly determines the material analysis accuracy and efficiency of the middle Echelle grating spectrometer. An improved piecewise polynomial fitting two-dimensional spectrogram restoration algorithm is proposed based on existing spectrogram restoration algorithms. The middle Echelle grating spectrometer based on the algorithm in this paper uses a transmission prism, which is different from the traditional method of ray tracing to obtain fitting data points. The pixel coordinate data corresponding to the wavelength is obtained through the dispersion law of the middle Echelle grating and prism. Then the detector image surface is divided into multiple areas according to pixel coordinates on the horizontal axis; Due to the certain correlation between the X coordinate and the diffraction level m, the principle of the algorithm is to establish a fitting relationship between the X coordinate and m, which can be used to obtain the diffraction level of the light ray from the X coordinate of the detectors image plane. By combining the relationship between the Y coordinate of the image plane and the wavelength, the relationship between the coordinate and the wavelength can be obtained. Perform polynomial fitting on each region to obtain the relationship between each regions wavelength and pixel coordinates and establish a model for restoring the entire region spectrum. Experimental verification shows that the calculation error of the established spectrogram reduction model is 0.000 1 nm. In summary, the algorithm designed in the paper has simple operations, can quickly achieve spectral restoration, and improves the accuracy of spectral restoration.

The monitoring of aircraft paint cleaning based on Laser-induced breakdown spectroscopy (LIBS) technology requires limiting the peak power density range to ensure the stability of plasma excitation and paint cleaning. However, for the widely used high-frequency (kHz-level) pulsed laser paint removal technique, the peak power density is relatively low, which limits the plasma excitation during the paint removal process, and the strong continuous background spectra generated by the high-frequency laser ablation of the material interferes with the plasma spectral acquisition. Based on the demand for controllable cleaning of the functional paint layer of the skin, the thesis designs a LIBS monitoring platform for high-frequency laser paint removal based on the writing of the control software of LabVIEW embedded development system and the integration of laser cleaning, spectral acquisition, control and display modules. The 2024-T3 aluminum alloy double-paint layer specimen was selected as the research object, and the spectra of the paint layer/substrate system with wavelengths in the range of 360~700 nm were collected (top paint layer: TC; bottom paint layer: PR; substrate: AS). The original spectra were preprocessed by smoothing filter, baseline correction, and normalization, and 12 characteristic spectral lines were selected for principal component analysis (PCA), and their dimensionality reduction data were used as the input variables for linear discriminant analysis (LDA), to establish the PCA-LDA discriminant model. Finally, the model was imported into the LIBS monitoring platform, and the classification accuracy of the high-frequency laser paint removal LIBS monitoring platform was verified through experiments. The results show that: only the cumulative variance explanation rate is greater than 85% as the principle of principal component selection, which can not meet the classification needs of LDA in the paint removal process; by optimizing the number of principal components of LDA, and ultimately selecting the first 9 principal components as the input of LDA, the detection accuracy of the LIBS platform is significantly improved. At this time, the classification accuracy of the PCA-LDA model based on LIBS spectra reaches 92.5%. It can be seen that the designed high-frequency laser paint removal LIBS monitoring platform can complete the material identification of different structural layers of the paint layer/substrate system, thus realizing the effective monitoring of high-frequency pulsed laser controllable paint removal.

In recent years, deep learning technology has been applied more and more in the quantitative analysis of near-infrared spectroscopy. However, the traditional convolutional neural network is applied to the spectral analysis due to the problems of a small amount of spectral data and insufficient data quality in near-infrared spectral data. Overfitting problems will occur in quantitative analysis. To improve the ability of convolutional neural networks to extract spectral information and enhance the ge-neralization of the network, this paper proposes a multi-feature fusion convolutional neural network model (MWA-CNN) based on wavelength attention to quantitative analyze the dry matter content in mango by near-infrared spectroscopy. MWA-CNN adds an attention mechanism and a multi-feature fusion mechanism based on the traditional convolutional neural network. The network can learn different spectral feature maps and weight information of different wave bands during the training process, thereby extracting high-quality spectral information to alleviate the overfitting problem in traditional convolutional neural networks and improve the accuracy of regression analysis.In the study, the near-infrared spectrum data of 11 691 mango samples were used, 80% of the samples were used as the training set, 20% of the samples were used as the test set by random method, and the test set root mean square error (RMSEP) and the training set root mean square error were passed. (RMSEC), coefficient of determination (R2), and mean absolute error (MAE) for model evaluation. In this paper, we first standardize the spectral data for pre-processing and then compare the prediction results with four traditional models of partial least squares regression (PLS), extreme learning machine regression (ELM), support vector machine regression (SVR), and traditional convolutional neural net-work (CNN) under the original spectral conditions.The prediction results show that the MWA-CNN network performs the best among the five methods, and the RMSE of MWA-CNN in the test set is 0.669 9. The traditional CNN effect is second only to MWA-CNN with an RMSE of 0.740 8, and the degree of over fitting of MWA-CNN decreases significantly compared to the traditional CNN. The RMSE of the test set in MWA-CNN compared to the training set increased by 15.69%, while the RMSE of the test set in the CNN compared to the training set increased by 151.45%. By adding noise with different signal-to-noise ratios to the spectra and then predicting the spectra with five models respectively after adding noise, the experimental results show that the MWA-CNN model can achieve the best results among the five models under various signal-to-noise conditions. It can be seen from the experimental results that the MWA-CNN has high prediction accuracy and generalization ability in NIR spectral quantile regression and a certain noise immunity capability.

Plastrum Testudinis is a popular traditional Chinese medicine (TCM) with abundant medicinal and edible value, and it is widely applied to clinical medical treatment and medicinal slice preparation. Studies show that the contents of trace elements in Plastrum Testudinis are directly proportional to its growth years. However, due to inexperience and nonstandard breeding, adulterated Plastrum Testudinis medicines are on the market. Because of the limitation of empirical and chemical-based methods, a heterogeneous ensemble learning (HEL) method based on a hyperspectral imaging technique is proposed to identify the growth years of Plastrum Testudinis. First, the Plastrum Testudinis samples with different growth years are taken as research objects. The original hyperspectral images of visible near-infrared ray (VNIR) and short-wave infrared ray (SWIR) lenses are captured on the hyperspectral imaging system. Then, the heterogenous ensemble learning (HEL) model is constructed based on support vector machine (SVM), logistic regression (LR), and K-nearest neighbors (KNN). Results show the fused hyperspectral images of VNIR and SWIR include more abundant spectral information. The HEL model can achieve satisfactory prediction ability by identifying the different growth years of Plastrum Testudinis samples. In addition, considering the detection efficiency, an unsupervised band selection is employed to reduce the complexity, eliminate the redundant bands in hyperspectral images, and improve the classification performance further. When the number of selected spectral bands is 32, the classification accuracy reaches 96.35%. Experimental results demonstrate that the HEL model based on hyperspectral imaging can accurately and rapidly identify the different growth years of Plastrum Testudinis samples and provide a novel technique reference for the attributes identification of TCM.

Iron ore resources are the foundation of our economic development and social progress. In the process of iron ore mining, the rapid and accurate determination of iron ore grade has an important influence on the mining decision and economic benefit. Hyperspectral imaging technology has the advantages of wide image coverage and high accuracy and has been widely used in ore classification and composition inversion. However, the band range of existing hyperspectral imaging sensors mainly includes visible and shortwave infrared (Vis-SWIR) and near-infrared (NIR). The two data types are mostly acquired independently, lacking continuity, and the accuracy of the model built with single data is often low. Therefore, the fusion of spectral data obtained by multiple sensors can effectively solve the problems of the small band range of a single sensor and few bands containing target characteristics and improve the accuracy of iron ore grade inversion based on hyperspectral imaging technology. In this study, Pika L and Pika NIR-320 hyperspectral imagers were used to collect imaging spectral data of Anshan iron ore in Vis-SWIR and NIR bands, respectively, and a spectral series fusion method based on mutual information (MI) was proposed. Firstly, the two groups of spectral data were preprocessed. Then, mutual information is calculated on the processed data to conduct a series fusion of spectral data. Finally, Vis-SWIR, NIR, and spectral data based on the series fusion of different bands were used as data sources to establish RBF neural network grade inversion models, and the accuracy and precision of the models based on spectral data before and after fusion were used as the discrimination index of the effectiveness of the fusion algorithm. The results show that the accuracy and precision of the model built after series fusion of spectral data is higher than that built using Vis-SWIR and NIR spectral data alone. Compared with the spectral data based on series fusion of other bands, the accuracy and precision of the model established based on the mutual information calculation of series fusion spectral data at 959.89 nm are the highest, R2 0.88, RPD 2.97, RMSE 4.464, MAE 3.32. This study proposes a new idea for multi-sensor spectral fusion, which has practical significance for the application of imaging spectrum technology in the inversion of iron ore grade.

Hyperspectral images have continuous spectral information of features and have great potential for shadow detection, but high band redundancy requires band preference. Normalized Shaded Vegetation Index (NSVI) can expand the spectral difference, and the application of NSVI in hyperspectral images will identify shadows more effectively. ZY1-02D satellite is the first hyperspectral operational satellite independently developed and successfully operated in China, with a large data signal-to-noise ratio and strong coverage capability, and it is important to perform accurate shadow detection on this hyperspectral image. In this paper, ZY1-02D AHSI images were used as experimental data to extract and analyze the spectral reflectance of vegetation in bright areas, vegetation in shaded areas and water bodies, and Combining Competitive Adaptive Reweighted Sampling (CARS) and Successive Projection Algorithm (SPA) to filter the main wavebands that can effectively distinguish typical features, the characteristics of the algorithms are considered to select the characteristic wavebands further to construct NSVI. The optimal threshold value is determined by the step method to classify the images, and the best band for constructing NSVI is compared in terms of image element value distribution, classification accuracy and spectral enhancement effect. A comprehensive evaluation is made by combining different shadow indices, bands and images to verify the significance and universality of the method in this paper. The results show that band 32 and band 73 are the best bands for NSVI construction, corresponding to the Red band and NIR band, respectively; the classification accuracy of NSVI constructed by different bands is generally higher than 90%, and the classification accuracy of NSVI constructed by the best band is 94.33% with a Kappa coefficient of 0.832 8, which is the best classification effect; NSVI can enhance the spectral difference between typical features and alleviate the “easy saturation” phenomenon of Normalized Difference Vegetation Index, and the small peaks generated by the accumulation of water bodies in this image is helpful to extract water bodies; The classification of NSVI in ZY1-02D AHSI image is better than Normalized Different Umbra Index and Shadow Index, and the classification accuracy in another scene image also reaches 93.55% with a kappa coefficient of 0.816 7; Therefore, the wavebands filtered by the algorithm are representative, and the NSVI constructed by the best waveband has better shadow detection ability in ZY1-02D AHSI images, which has a certain reference and significance for hyperspectral image shadow detection and construction of vegetation index.

The spectral curve is an important representation of vegetation physiological traits. The study of vegetation spectra is significant for monitoring vegetation growth and health status. Current research on vegetation spectra mainly analyzes different vegetation within the same vegetation type, and there are few reports on the systematic integration of spectral characteristics of different vegetation subformations. The ecological pattern of the Tibetan Plateau has changed significantly in recent years. To conduct research on vegetation spectral characteristics on a large scale in the Qinghai-Tibet Plateau region and to explore the spectral separability among different vegetation subformations based on high-resolution Sentinel-2A/B remote sensing images, the spectral curves of 19 vegetation subformations were analyzed by using the first derivative of the spectrum and vegetation indices. The results showed that: (1) The spectral curves of vegetation sub-formations exhibit significant differences in the near-infrared band, and their reflectance value can be used to distinguish different vegetation sub-formations; (2) First-order derivative processing can elucidate the three-edge features characteristics of vegetation, the first-order derivative spectra of vegetation are of certain regularity, with significant differences in the red edge; (3) Vegetation indices of different vegetation subformations are consistent with changes of the red edge characteristics parameter, and the vegetation indices are discriminative, although their values are relatively small;(4) By combining near-infrared peaks, three-edge characteristics parameters, and vegetation indices, different vegetation subformations can be distinguished. This study analyzed the spectral curve characteristics of different vegetation at the vegetation subformations scale. The experimental results can provide data support for improving the vegetation subformations spectral database and have practical significance for the fine identification of vegetation subtypes on the Qinghai-Tibet Plateau.

Gancao is a kind of Chinese herbal medicine with the functions of tonifying the spleen and benefiting the qi, clearing away heat and detoxifying toxins, expelling phlegm, relieving cough, etc. Due to the influence of genetics and geography, there is a great difference in the quality and price of different varieties and places of origin, and there are often unscrupulous merchants who use the second-rate to make the best to gain benefits. Aiming at the structural properties of glycyrrhizic acid and glycyrrhizin, which are closely related to the quality of Gancao, this paper constructs a spectral sensing method for the rapid identification of Gancao varieties and origins through six organic dyes and metal ions that can react with them optically in a specific manner. In this paper, 175 samples from five main Gancao production areas, Xinjiang, Inner Mongolia, Ningxia, Beijing, and Uzbekistan, were collected to measure the UV-visible spectral data of Gancao extracts and the addition of different metal ions-organic dyes, respectively. By comparing the differences in UV-visible spectrograms, we have screened out six metal ions and organic dyes that can react with Gancao extracts with specific optical reactions and used them to construct UV-visible sensing points. All the collected samples were then detected and analyzed by the constructed UV-visible spectral sensing method and combined with partial least squares discriminant analysis (PLS-DA) and random forest (RF) algorithms to identify the varieties and origins of Glycyrrhiza glabra. The results showed that the spectral peaks of the original Gancao species and origin overlapped in several places, and the Q2 in the PLS-DA classification results was only 0.75. That is., the pure spectra provided preliminary identification of Gancao but with low predictive accuracy. To further improve its identification accuracy, metal ions, and organic dyes were added to the Gancao extracts to produce spectral peak variations, which were more pronounced for both Gancao species and origin compared to the original UV-Vis spectra of Gancao. Using vector-coded partial least squares discriminant analysis (Dummy codes-PLSDA) and random forest (RF) algorithms to establish identification models of Gancao species and origin, it was found that the accuracies of the fusion spectra of the six sensing materials reached 100%, which significantly improved the recognition ability of the metal ion-organic dye sensing array for Gancao, and the accuracy of the RF model was significantly higher than that of the PLS-DA model, in which the identification accuracies of RF and PLSDA models with a single sensing site were greater than 98.65% and 91.30%, respectively, and much higher than 65.22% of the original spectra. Therefore, in this paper, an ultraviolet-visible spectral sensing method was constructed to accurately identify Gancao species and their origin using metal ions-organic dyes-. The method provides a new idea for rapidly identifying other Chinese herbal medicines, protecting consumers health and interests and promoting sustainable development of Chinese medicine resources.

The purpose of this study is to study the effect of supercritical carbon dioxide (ScCO2) on the chemical structure of coal after solvent pretreatment. The lean coal samples collected from Changzhi Huoerxinhe were adopted. Coal samples were pretreated with tetrahydrofuran (THF), hydrochloric acid (HCI), and hydrofluoric acid (HF), respectively. The Fourier transform infrared spectroscopy (FTIR), solid-state nuclear magnetic resonance (13C-NMR), and X-ray diffraction (XRD) were used to study the chemical composition and structure of coal samples. Results show that: ①The FITR peak-fitting spectra were basically consistent with the experimental curves, but the positions and intensities of the absorption peaks of each functional group still showed some deviations, and the aliphatic peaks disappeared in part of the bands after the pretreatment with acids (HCI and HF). The two types of acid pretreatment order are different, and the effect on the oxygenated functional group, fat structure, and aromatic structure of coal is not the same. The intensity of aromatic CC structure and oxygenated functional group peaks were enhanced after HF-HCI treatment. In contrast, the intensity of the aromatic CC structure was weakened after HCI-HF treatment, and the intensity of the structure of the oxygenated functional group was insignificant. The intensity of the aromatic CC structure of the coal samples was enhanced, and the intensity of the structure of the C〖ZJLX,Y〗C oxygenated functional group was reduced after THF pretreatment. After THF pretreatment, the intensity of the aromatic CC structure was enhanced, and the intensity of the CC oxygenated functional group was reduced. Overall, the intensity of aromatic CC structure is much larger than that of aliphatic structure and oxygenated functional group structure. ②In the 13C-NMR spectrum, the chemical shift of the main functional group peaks exhibits a certain degree of deviation. Aromatic carbon faB chemical shifts are overall shifted in the direction of increasing. The proportion of aromatic carbon is much greater than that of aliphatic carbon, indicating that aromatic carbon forms the bulk of the coal macromolecular structure. ③In the XRD pattern, the diffraction intensity of the 002 and 101 peaksk increases significantly, and the microcrystal structure parameter d002 shows an increasing trend. It can be seen that THF, acidic treatment, and ScCO2 all have a certain impact on the macromolecular structure of coal, causing the coal macromolecular network structure to become looser, which in turn increases the overall distance d002 of the microcrystalline layer network. Research shows that changes in the spectral characteristics of the coal after solvent action were not only related to the nature of the solven, but also to the order of inorganic acid treatment, and consequently, the functional group composition and macromolecular structure were altered to varying degrees, which in turn affected the extraction effect of ScCO2 on the pretreated coal.

To rapidly detect Perilla aldehyde (PAE) content in Perilla Frutescens, hyperspectral imaging technology was employed, and the hyperspectral images of Perilla Frutescens were acquired from four distinct producing regions. Based on obtaining effective wavelength images of Perilla Frutescens, its texture features and energy values obtained by wavelet transform were fused with spectral values in various ways to create different characterization vectors. These vectors were then employed to construct corresponding rapid detection models for PAE content. The models prediction capabilities were thoroughly compared and analyzed to determine the optimal prediction strategy for PAE content. The specific methods are as follows. (1) Four methods were employed to preprocess the raw spectral values. After evaluating the predictive performance of the constructed models, it was determined that Local Weighted Scatterplot Smoothing (LOWESS) emerged as the optimal preprocessing method. (2) The Competitive Adaptive Reweighted Sampling (CARS) and Successive Projections Algorithm (SPA) were employed to extract the characteristic wavelengths of the preprocessed spectral information, and the corresponding spectral values were then computed to facilitate their integration with other mentioned features in the paper. (3) Principal Component Analysis (PCA) was utilized to get effective wavelength images from the hyperspectral images. The grayscale co-occurrence matrix (GLCM) was then applied to the effective wavelength images to extract four texture features: Energy (ASM), Contrast (CON), Correlation (COR), and Entropy (ENT); simultaneously, the Daubechies wavelet was employed to conduct three-level decomposition of the effective wavelength image, and the energy of the low-frequency component derived from the decomposition was also considered as a characterization feature of the effective wavelength image. (4) The extracted features of wavelength spectral values, wavelet energy values, and texture features were utilized to construct feature input vectors in different ways, and based on the mentioned vectors, four detection models were then constructed: Partial Least Squares Regression (PLSR), Backpropagation Neural Network (BPNN), Random Forest (RF), and Extreme Gradient Boosting (XGBoost); and then these models were evaluated and compared according to their prediction capabilities to identify the optimal input vectors and prediction models. The research results indicate that the prediction capabilities of the four prediction models using single-class feature input vectors are all inferior to that of the input vectors fused with multi-class features; the optimal input vector is the feature fusion input vector, which incorporates the texture feature values and wavelet energy values from effective wavelength images as well as the spectral values corresponding to the feature wavelengths selected by CARS after LOWESS preprocessing. Among these models, the XGBoost model demonstrates the strongest prediction capabilities. The R2c and RMSEC of the training set are respectively 0.998 08 and 0.022 49, and the R2p and RMSEP of the testing set are 0.989 44 and 0.036 40, respectively. This research finding introduces a novel approach for the rapid detection of PAE content in Perilla Frutescens, and it also serves as a valuable reference for developing detection strategies for other components.

Inedible shelled Torreya grandis bad seeds will be produced during post-ripening treatment and frying, which cannot be accurately recognized and rejected manually without destroying the shells, affecting the overall quality of shelled Torreya grandis seeds. This study used two near-infrared spectrometers to collect spectral data of shelled normal and bad Torreya grandis seeds and eight spectral pre-processing methods was studied and compared. Then, a single wavelength selection method (Uninformative Variables Elimination, Competitive Adaptive Reweighted Sampling, Successive Projections Algorithm, and Subwindow Permutation Analysis) and a joint wavelength selection method were adopted to select characteristic wavelength, and Linear Discriminant Analysis (LDA) and Support Vector Machine (SVM) methods were applied to establish the identification model of Torreya grandis bad seeds. Also, the models performance was compared to determine the better wavelength selection method for different spectrometers. The results show that for spectrometer 1, preprocessing can not improve the model performance effectively. The Successive Projections Algorithm is the optimal wavelength selection method. The sensitivity, specificity, and accuracy of the LDA and SVM models in the prediction set are 97.10%, 95.00%, 96.00% and 97.10%, 97.50%, and 97.30%, respectively, superior to the full-wavelength model. The number of modeled wavelength variables was reduced from 661 to 9, only 1.36% of the original number of wavelength variables. For spectrometer 2, baseline correction is the optimal preprocessing method, and Subwindow Permutation Analysis is the optimal feature wavelength selection method. The sensitivity, specificity, and accuracy of the prediction sets of the developed LDA and SVM models are 100.00%, 92.50%, 96.00% and 100.00%, 95.00%, and 97.30%, which are consistent with full-band model performance. The number of modeled wavelength variables was reduced from 155 to 55, which is 35.48% of the original number of wavelength variables. It can be seen that near-infrared spectroscopy can better identify the shelled bad Torreya grandis seeds, and the appropriate wavelength selection method can effectively screen the characteristic wavelengths, simplify the model, and improve the accuracy and stability of the model. It is also found that the wavelength range of 1 000~1 300 nm is related to the starch, fat, and protein content of Torreya grandis seeds, making it more suitable for identifying bad Torreya grandis seeds. This study provides a reference for the rapid and nondestructive identification of shelled Torreya grandis bad seeds.

The application of organic fertilizer is an effective way to improve the quality and efficiency of crop production. Studying the distribution of dissolved organic matter (DOM) in soil profiles under the application of organic fertilizer from different sources is helpful to optimize the utilization of organic fertilizer and further understand the environmental behavior of DOM. This paper used 2D correlation spectroscopy (2D-COS) to study the effects of different organic fertilizers (pig manure, chicken manure, sheep manure, cow manure, and biogas residue) on DOM distribution in the soil profile. The results showed that organic fertilizer return to the field mainly affected the soils relative content and composition of DOM. After applying different organic fertilizers, the relative content of DOM in soil was always higher than that in control (CK) treatment, and the relative content of DOM in the 0～10 cm soil layer had the greatest effect, with an average increase of 14.67 g·kg-1. Different types of organic fertilizer had a certain difference in the increase of profile DOM, among which chicken manure organic fertilizer had the greatest effect on the increase of profile DOM, increasing by 21.42%. The application of organic fertilizer mainly increased the relative content of tyrosine-like components (C4), and had the greatest impact on the relative content of C4 in the 10～20 cm soil layer, with an average increase of 4.12%, in which sheep manure and biogas residue had the greatest impact, with an increase of 7.80% and 7.89%, respectively. 2D-COS analysis showed that applying organic fertilizer from different sources greatly impacted dissolved microbial metabolites and protein-like components (295, 315 nm); CK and chicken manure treatment mainly affected dissolved microbial metabolites, and other treatments first responded to protein-like components. The humification index HIX, Fn(330) and Fn(280) decreased with the increase of soil depth. Still, the changes of HIX, Fn(330), and Fn(280) were different with different types of organic fertilizer, among which cow manure had the greatest effect on HIX and Fn(330). Sheep manure had the greatest effect on Fn(280). The biological index BIX and freshness index β∶α increased with the increase of soil depth, and the influence of biogas residue was the most significant. Still, the fluorescence index FI had no significant change.

The rapid and precise detection of heavy metals in soil is the key to the efficacious prevention and remediation of soil contamination. Employing a portable X-ray fluorescence spectrometer facilitates the in-situ, non-destructive, and rapid detection of typical heavy metals. This advanced analytical technique also obviates the need for elaborate sample digestion procedures. However, the accuracy of the XRF-based heavy metal detection technique is significantly influenced by the soil matrix effects, which considerably limits the accuracy of such measurements. Calibration against standard soil with a similar matrix is imperative. As a result, this study combined pattern recognition and the standard curve method to achieve a precise analysis of typical heavy metals in various soils. The dataset comprises the X-ray fluorescence spectra and heavy metal contents across six characteristic soil types collected within China: humid-thermo ferritic, paddy soils, black soils, flavor-aquic soils, yellow-brown earth, and yellow-red earth. The spectral data is refined using a five-point, three-times window movement smoothing algorithm and a min-max normalization approach, followed by principal component analysis (PCA). Post-PCA dimensionality reductions first five principal components are employed as input feature variables, with soil types serving as labels. A predictive model based on a Radial Basis Function (RBF) kernel for Support Vector Machine (SVM) is constructed to categorize soils by matrix similarity. The models hyperparameters are optimized using the Horned lizard optimizer algorithm, yielding an optimized kernel function (g) of 0.038 1 and a penalty factor (c) of 7.852 9, with a correct classification rate of 100% under a five-fold cross-validation. The quantitative analysis utilizes the standard curve method. For the six soil types, the correlation coefficients for Chromium (Cr) ranged from 0.994 7 to 0.999 3, for Nickel (Ni) from 0.986 8 to 0.999 4, for Copper (Cu) from 0.992 9 to 0.999 9, for Zinc (Zn) from 0.984 1 to 0.999 8, and for Lead (Pb) from 0.987 7 to 0.999 6. Furthermore, the correlation coefficients of Arsenic and Lead (As & Pb) ranged from 0.961 3 to 0.999 5. The above results indicate a favorable linearity for heavy metals within the same matrix. Subsequently, the established RBF-SVM model and standard curves are applied to a prediction set of 24 samples. The predictive outcome corroborates a 100% classification accuracy for the six soil types. Upon classification, corresponding standard curves are utilized for quantitative analysis. The results show that the average relative prediction errors for Cr, Ni, Cu, Zn, Pb, and As are 2.24%, 3.66%, 2.72%, 2.15%, 2.13%, and 5.55%, respectively, below 6%. These findings prove the excellent applicability of the RBF-SVM model in combination with the standard curve method for the rapid detection of typical heavy metals in soil. This algorithm will facilitate the rapid quantitative detection of typical heavy metals in natural soil.