Remote spectral detection is an important way to explore large-scale space and matter,which plays critical roles in astronomy,meteorology,and the deep sea. However,detecting remote and often weak spectra poses high requirements for the performance of spectrometers. According to the spectral performance evaluation standard proposed by Jacquinot,compared with other spectral systems (grating spectrometer,prism spectrometer,etc.),the Fabry Perot (F-P) interferometer has a large aperture and high spectral resolution,which possesses the intrinsic advantage in the field of remote weak light detection. Since Fabry first used the F-P interferometer for astronomical observation in 1914,the F-P interferometer has been widely used in remote spectral measurement,and various improved F-P interferometers have been developed in recent years. Traditional F-P interferometers mainly face three problems when used in remote spectral detection:narrow free spectral range,difficult installation and adjustment of large aperture F-P interferometers,and unconcentrated spectral energy induced by the circular interference structures. This article introduces three typical improved F-P interferometers,including the cascaded F-P interferometer that greatly extends the free spectral range,the rotating scanning F-P interferometer that eases the adjustment and is suitable for extreme environments，and the circle-to-line interferometer optical system(CLIO) that converts interference rings to interference lines with improved energy concentration.This article provides a systematic summary of the important applications of the F-P interferometer in meteorology,astronomy,and the deep sea. In meteorology,a large-aperture F-P interferometer for measuring wind speeds in the atmospheres mesosphere and thermosphere was introduced,and a German Heisenberg high-precision F-P interferometer for measuring trace gases and their isotopes in the atmosphere was also presented. In astronomy,a cascaded F-P interferometer designed by the University of Wisconsin in the United States for studying interstellar material emission lines was introduced. In the field of oceanography,the main application examples of domestic F-P interferometers for measuring Brillouin scattering were introduced,including the underwater Brillouin scattering system designed by Beijing Normal University in 2004 and the spaceborne Brillouin scattering system designed by Shanghai Jiao Tong University in 2021. Finally,this article proposes that spectral recognition accuracy and thermal stability are difficulties that need to be solved in the future applications of F-P interferometers in remote spectral measurement.

With the increasing demand for nuclear energy,uranium exploration has become a key link in the supply of nuclear energy. Uranium exploration methods mainly include radioactive geophysical surveys,geochemical surveys,and other traditional methods,most of which have the shortcomings of inaccurate detection data and low efficiency.This study used near-infrared spectroscopy and stoichiometry to explore the feasibility of screening and identifying uranium super enriched plants. Through the investigation of the growth and characteristics of plants in the uranium mining area,the ultra-enriched plants were selected,the leaf comprehensive spectra in different regions were obtained by a near-infrared spectroscopy analyzer,and the spectral response relationship was compared and analyzed. It was found that the absorption peaks of the two hyper-enrichedplants were located in two bands:650~700 and 950~1 050 nm. The absorption peak of chlorophyll in the former band was mainly generated by the combined frequency of C—O and C—H bond stretching vibration. In the latter band,the absorption peak of water is mainly caused by the 5-order frequency doubling of O—H bond bending vibration. The feature variables were selected by principal component analysis (PCA) and successive projections algorithm (SPA),and the two samples were randomly divided into training and prediction parts according to the ratio of 3∶1,respectively. The detection model of uranium enrichment in super-enriched plants was constructed by combining the two methods of partial least squares(PLS) and least square support vector machine (LSSVM),and the prediction effect was compared. Obtained the detection model of Setaria uranium enrichment based on PLS had the best effect,with a discrimination accuracy of up to 100%,RMSEP of 0.115,and R2 of 0.946. The Setaria detection model is superior to the coverage in the two modeling methods. It may be that the enrichment coefficient of setaria morifera is higher than that of ciderage.The results show that the detection model of uranium enrichment in super-enriched plants established by near-infrared spectroscopy combined with the partial least squares method has the best effect,and it is feasible to screen and identify uranium super-enriched plants. This method provides an important reference for the ecological restoration of the spent uranium ore area and a new idea for the use of specific and indicative plants to search for uranium ore.

Graphite crystals with excellent chemical inertness,high melting point,strong elasticity,and unique electronic properties are widely used in many fields. Pyrolytic graphite is increasingly used in XRF technology due to its excellent high integral reflectivity for X-rays. Pyrolytic graphite is a synthetic graphite material with a layered structure. Annealing close to 3 000 ℃ can adjust the material structure. The structural changes are enhanced,while the annealing is accompanied by deformation. Different combinations of annealing temperatures and types of deformation result in the production of pyrolytic graphite with different structures; there are mainly two types of pyrolytic graphite:highly oriented pyrolytic graphite (HOPG) and highly annealed pyrolytic graphite (HAPG),respectively. HOPG crystals,typically having a mosaic spread in the range of 0.3 to 3°,offer a higher integral reflectivity among 2 keV to tens of keV,which is at least an order of magnitude higher than the reflectivity of all other crystals for X-rays,but moderate resolution. The HOPG energy resolution is mainly affected by mosaic spread and intrinsic widths of the Bragg reflections; larger mosaic spread results in significantly lower energy resolution. HAPG crystals are a modified material similar to HOPG crystals,with a lower mosaic spread of 0.05 to 0.1°and a higher resolution but a lower integral reflectivity. Its mosaic spread is about 1/5 compared with HOPG crystals of the same thickness. Monolayer HAPG crystals can be deposited on the substrate and bent into small radii and almost any shape without significantly reducing the energy resolution. Graphite crystals are mainly used as monochromators in the field of X-ray fluorescence,which can reduce high background caused by white light,fluorescence,and stray light in Von Hamos spectrometers and EDXRF instruments,the Von Hamos spectrometer with the focusing of pyrolytic graphite crystals can obtain a better integral reflectivity and energy resolution; which improved the sensitivity and detection limit of the analysis. It also performs well in studying fine structures,measurements of radioactive elements,and mammography. This paper mainly reviewed the research and applications of pyrolytic graphite in X-ray fluorescence spectrometers in the last two decades.

As a fast and nondestructive detection method,spectral detection is gradually applied in more and more fields. With the continuous improvement of information accuracy requirements,obtaining spectral information with high resolution and low wavelength error is a primary challenge that researchers must face when designing spectral detection systems. Studying a new type of dispersive element becomes a feasible scheme. Traditional dispersive elements,such as diffraction gratings,have an extensive wavelength response range but usually can provide a resolution of the order of nm and cannot meet the requirements of the current hyperspectral resolution. Virtual Image Phased Array (VIPA) has the characteristics of large angular dispersion and high spectral resolution,which can effectively improve the spectral resolution of the system. Therefore,it has been continuously used in the research of fine spectral detection in recent years. However,the corresponding relationship between the wavelength and the dispersion angle is nonlinear,and the same wavelength is prone to generate multiple diffraction orders,resulting in additional light energy loss. Therefore,based on the cross-dispersion optical path of VIPA,this paper conducts theoretical analysis and experimental verification on how to reduce the dispersion nonuniformity of VIPA. Firstly,based on the dispersion law of paraxial theory,the influence of the incident waist size,incident angle and other parameters on the nonlinearity of VIPA dispersion and the number of diffraction orders is analyzed theoretically.The number of diffraction orders of each wavelength and the calibration value of the spot centroid corresponding to each wavelength of the broad spectrum light source at different VIPA tilt angles are recorded through simulation. The linearity under different angles is compared by calculating the average distance difference between the centroid coordinate and the fitting curve in each case. Finally,the experiment is carried out according to the theoretical analysis and the simulation results. The number of diffraction orders of the cross-dispersion optical path of VIPA under different VIPA tilt conditions was recorded in the experiment,and the centroid coordinates of each diffraction order spot of different wavelength light sources under the experimental condition of VIPA tilt of 2° are obtained by using a variable wavelength laser. Similarly,the linearity of different diffraction orders is analyzed by calculating the distance average between the centroid of different diffraction orders and the fitting curve. The results show that the nonuniformity of VIPA dispersion can be reduced to some extent by controlling the parameters of the incident beam.The larger the tilt angle of VIPA and the closer to the center of the detector,the better the diffraction order linearity.

As one of the four emerging pollutants, the harm caused by “microplastics” has become increasingly prominent. The detection and identification of microplastics are the keys to pollution assessment and risk management prevention and control. This paper uses microplastics (including PA, PE, PET, PP, PS, and PVC) in fishmeal as the research objects. The XGBoost algorithm studies and constructs the qualitative recognition models of near-infrared and infrared spectroscopy. The XGBoost algorithm studies and constructs the qualitative recognition models of near-infrared and infrared spectroscopy. Optimising the main hyperparameters of the XGBoost model using the GridSearchCV toolkit. The hyperparameter optimization results of the near-infrared spectroscopy model were n~~estimators: 300, learning~~rate: 0.08, gamma:0, max~~depth: 4, min~~child~~weight: 1. The hyperparameter optimization results of infrared spectroscopy are n~~estimators: 100, learning~~rate: 0.02, gamma: 0.20, max~~depth: 4, and min~~child~~weight:1. The average Precision of the NIR qualitative recognition model constructed based on the optimized hyperparameters was 0.985, the average Recall was 0.977, and the average F1 score was 0.978, which improved by 40.17%, 51.00%, and 50.00% compared with the model before optimization. The average precision,average recall,and average F1 scores of the infrared qualitative recognition model were all 1.000,and the optimized model effect improved by 20.67%, 27.50%, and 26.33%, respectively. Further comparative analysis with the PLS-DA model shows that the XGBoost model of the infrared spectrum is the same as that of the PLS-DA model, and the effect of each parameter (Accuracy, Precision, Recall, F1 score) of the XGBoost model of the near-infrared spectrum is better than that of PLS-DA model to varying degrees. In summary, the XGBoost algorithm can effectively identify different types of microplastics in fishmeal. This study provides theoretical and technical support for rapidly detecting and identifying microplastics in fishmeal.

Surface Enhanced Raman Scattering (SERS) technology has great potential in detecting pesticide residues,but there are still limitations in trace and quantitative analysis. This paper proposes a pesticide detection strategy based on nano scale convex polyhedrons Au@4-ATP@Au (NCPs-Au@4-ATP@Au). XRD results showed that because of the selectivity and inductance of probe molecules to gold precursors,the crystal surface structure information of NCPs-Au@4-ATP@Au nanoparticles and spherical gold nanoparticles is significantly different,which is reflected in the strong reflection peak at (200). Combined with SEM and absorption spectrum,it can be determined that NCPs-Au@4-ATP@Au has both spherical and polyhedral structure characteristics. Compared with spherical gold nanoparticles,the absorption peak was significantly red shifted and closer to the wavelength of excited light,which was theoretically more conducive to enhancing the SERS signal. Experiments showed with NCPs-Au@4-ATP@Au,which was coated with high index crystal surface and embedded with probe molecule 4-ATP as the enhanced substrate,the Limits of detection (LODs) of Carbendazim (CBZ) reached 0.66 nmol·L-1. According to the Raman and SERS spectral shift CBZ molecules,it can be preliminarily confirmed that CBZ molecules are adsorbed to gold nanoparticles through the NH bond andCObond. Au@4-ATP@Au improved the sensitivity because of the multi-convex structure. Meanwhile,with 4-ATP as the calibration signal,the spectral stability and timeliness were also improved. After normalization,spectral stabilitys relative standard deviation (RSD) was as low as 7.03%,the signal intensity decreased by 5.87%,and RSD was 2.94% in half a month. The results showed that NCPs-Au@4-ATP@Au improved the trace and quantitative detection ability of SERS in detecting pesticides,and the substrate is expected to promote the practical application of SERS.

X-ray fluorescence analysis (XRF) is a remarkably effective analytical technique for quantitatively studying heavy metal elements in soils. Due to matrix effects and elemental interferences, existing machine-learning methods suffer from inadequate performance and instability in predicting lead (Pb) and cadmium (Cd) concentrations using soil XRF spectra. Therefore, this paper proposes a PCA-BLS method for the XRF quantitative analysis of heavy metals in soil based on principal component analysis (PCA) combined with the broad learning system (BLS). It can accurately, efficiently, and stably determine concentrations of Pb and Cd in soil. First, the 56 standard soil data are feature-reduced using PCA. The first three principal components of Pb and Cd are selected as features. Then, the optimal principal component features are fed into the width learning system for calibration and testing. Using the grid search determine the optimal network structure. The three optimum parameters for the BLS corresponding to the Pb and Cd elements are 2,11,11 and 3,19,15, respectively. Using support vector regression (SVR), BP neural network, and the original BLS compared with the PCA-BLS. PCA-BLS achieved performances of 0.954, 1.433, and 1.014 in the R2, RMSE, and MAPE corresponding to Pb. In the quantitative Cd, PCA-BLS obtains the R2 of 0.982, RMSE of 1.215, and MAPE of 1.059. Grid search visualization demonstrates the stable performance of PCA-BLS in predicting two heavy metal elements. The experimental results show that PCA-BLS can effectively correct for matrix effects and interferences in soil XRF. The PCA-BLS is a promising method for quantitative XRF spectroscopy that accurately predicts Pb and Cd elemental concentrations while maintaining model stability.

Terahertz technology has a broad application prospect in communication systems. Digital codingmetasurface can control electromagnetic wave propagation and has been widely concerned. Most of the reported codingmetasurface achieve full 2π phase coverage of reflected or transmitted LP (linearly polarized) and CP (circularly polarized) waves based on PB (geometric phase) or transmission phase,limiting the flexibility of terahertz wave regulation. This paper proposes a metasurface element structure consisting of three layers,from top to bottom:the metal pattern structure layer,intermediate medium layer,and metal base layer. In this paper,a metasurface element is proposed by combining the propagation phase and geometric phase. The metasurface is designed to generate vortex beams based on the transmission phase mechanism under the incident of online polarized terahertz waves,and multi-vortex beams are generated by convolution operation. According to the geometric phase principle,phase coverage in the 2π range can be achieved by rotating the metal pattern structure layer under the circularly polarized terahertz wave incident. The metasurface is designed to generate vortex beams based on the transmission phase mechanism under the incident of online polarized terahertz waves, and multi-vortex beams are generated by convolution operation. According to the geometric phase principle, phase coverage in the 2π range can be achieved by rotating the metal pattern structure layer under the circular polarized terahertz wave incident. The metasurface can be properly arranged by using phase coding. The designed metasurface can deflect the reflected beam, showing good flexibility. Secondly, the designed metasurface generates vortex beams at multiple frequency points. At 0.9 THz, vortex beams with topological charges l=1 and l=2 are generated with mode purity of 68.9% and 69.5%. At 1.2 THz, vortex beams with topological charges l=1 and l=2 are generated with mode purity of 91.03% and 87.2%. In addition, the deflection vortex beam is generated by convolution, and the focusing coding metasurface is designed to realize the focusing function. In this paper, we propose a terahertz multi-dimensional phase-controlled reflection metasurface, which combines the propagation phase with the geometric phase to realize the regulation of two independent polarized channels. The results show that the designed metasurface provides a multi-degree-of-freedom method for polarization and phase manipulation of terahertz waves, which has broad application potential in terahertz systems.

Open set recognition (OSR) has been investigated for approximately 10 years. It can recognize samples from the known classes in the training dataset,whereas it rejects samples from the unknown classes not included in the training dataset. The current OSR schemes are mainly based on Support Vector Machine (SVM) and deep learning neural networks. These OSR schemes are mainly used in natural scenery images and are rarely used in spectral analysis. In this paper,the classical fuzzy reasoning classifier in the closed set is improved with application to tree class spectral classification in the open set. First, a Flame-NIR spectrometer picks up the wood near-infrared (NIR) spectral curve. After metric learning processing,the spectral 4-dimensional (4D) feature vector is used as a classification feature. Second,the fuzzy reasoning classifier is improved for its use in an open set scenario. A new generalized basic probability assignment (GBPA) is used based on the confidence value of a fuzzy rule and the product of membership degree probability in each dimension. The comparison experimental results on wood NIR datasets with different “Openness” values indicate that our proposed scheme (Fuzzy Reasoning Classifier in an Open Set,FRCOS) outperforms the state of the art OSR schemes based on machine learning and deep learning with good performance evaluation measures.

Blood is a regulated exceptional genetic biological resource. In response to the issue of easy oxidation and deterioration in traditional blood spectral detection, dynamic confocal Raman fluorescence spectroscopy technology based on biomimetic blood vessels was used to conduct blood species identification research on six types of poultry and livestock, including pigs, horses, pigeons, chickens, ducks, and geese. The preprocessing process of the original spectrum includes baseline removal, smoothing, and normalization. Linear discriminant analysis is used to reduce the dimensionality of spectral data, and then support vector machines are used to establish recognition models. Gaussian kernel functions are selected, and the parameters C and γ Make their classification accuracy the highest, the optimal C and γ 0.2 and 0.134, respectively. The recognition accuracy of the artificial fish school support vector machine model reaches 97.2%. The dynamic confocal Raman fluorescence spectrum based on biomimetic blood vessels used in this article can meet the requirements of blood safety and efficiency detection, and the algorithm model optimized by the artificial fish school algorithm for support vector machine parameters shows good classification performance.

Hydrothermal method was used to synthesize nanosized Fe3+, Al3+, Co2+, and La3+ co-doped Ce1-4x(FeAlCoLa)xO2-δ(x=0.00~0.05) solid solutions. The solid solutions microstructure,morphology,spectral characteristics,and redox activities were systematically characterized and analyzed by XRD, TEM, SEM, UV, PL, Raman, and temperature-programmed reduction (TPR) with H2. XRD results showed the Ce1-4x(FeAlCoLa)xO2-δ solid solutionsexhibited the CeO2 cubic fluorite structure. A tiny diffraction peak corresponding to the Co3O4 impurity phase at 36.6° was observed when the doped content reached x=0.04, indicating that x=0.04 was the solid solubility of doped ions in the CeO2 lattice. The positions of the (111) diffraction peaks were shifted towards a higher angle, which proved the doped ionsinduced the distortion of the lattice. The TEM and SEM images showed the samples were spherical with high crystallinity, and doping caused lattice contraction. The UV absorption spectra revealed that the doped samples absorption edges were gradually red-shifted compared to pure CeO2. Extra absorption peaks corresponding to the doped ions were found in the region of 560~780 nm. The band gap energies decreased from 2.84 eV (pure CeO2) to 2.1 eV (x=0.05). The reasoncould be that the doped ions formed new impurity energy levels between the valence and conduction bands, which allowed the electrons to transition from the valence band to the lower impurity energy levels and then lowered the band gap energies. In addition, the distortion of the lattice and increased concentration of oxygen vacancies prevented the electrons from transferring to higher energies, which can also result in the reduction of band gap energies. PL spectra showed that doping significantly reduced the emission peak intensities. Raman spectra demonstrated that the dopingresulted in the shift of the F2g peak, the decrease of peak intensities, and the widening of peaks. Meanwhile, the relative intensities of the peak corresponding to the oxygen vacancies were also observed to be enhanced. Thus, both the PL and Raman spectra proved that doping increased the degree of lattice distortion and the concentration of oxygen vacancies. The H2-TPR test results showed that doping can effectively reduce the redox reaction temperatures and improve the redox activities. The sample doped with x=0.03 possess the lowest surface reduction temperature and the largest peak areas, which meansthis sample exhibited the best redox activities. It can be concluded that the redox performances of the samples were closely related to the grain sizes, lattice defects, and oxygen vacancy concentrations. This study showed that the four ions co-doped with CeO2 could effectively modify the microstructure and improve the samples catalytic activities at a low doping concentration.

SanHuang Temple is located in Xian City,Shaanxi Province,and houses three ancient emperors:Fuxi,Shennong,and Huangdi. The walls were originally painted with exquisite murals. It is the product of integrating ancient Chinese traditional medicine and Confucianism,which has great art and academic research value. However,SanHuang Temple has experienced wind and rain in natural preservation,and its status is precarious. To identify and analyze the materials for the construction of murals in the SanHuang Temple and to provide a scientific basis for the acquisition of the original information of the murals and subsequent protection and restoration,the tiny pigment samples taken from the SanHuang Temple were analyzed by comprehensive analysis methods such as the ultra-depth-of-field microscope,scanning electron microscope-energy dispersive spectrometer (SEM-EDS),X-ray diffraction analyzer (XRD),micro-laser Raman. The results show that the red pigments include lead,cinnabar and hematite,and the main colour components of black stains are carbon black mixed with Prussian blue. The blue tint is synthetic ultramarine blue,the green pigment is a mixture of Paris green and titanium dioxide,and the main component of the pigment base layer is dihydrate gypsum. The scientific and technological analysis of mural production materials shows that the SanHuang Temples murals have been historically restored. The restoration age is relatively late,and there was the use of mixed pigments in the restoration,among which the mixed use of carbon black and Prussian blue is the first time to be found in the mural,which provides a basis for the research on the production materials and repair history of the SanHuang Temple murals in Shaanxi Province,and provides scientific and technological support for the later mural protection,restoration and restoration display,which has critical cultural relics protection significance.

As a kind of nutrient-rich nut,it is of great economic value and practical significance to test the quality of almonds. Because of the almond hard shell,it is difficult for traditional detection methods to realize internal detection.In this paper,the emerging terahertz transmission imaging detection technology is used to study almond plumpness detection. Firstly,the terahertz spectral images of almonds with different fullness are acquired. Secondly,the terahertz spectra of sample free region,empty shell region and full almond region are extracted,respectively. To improve the accuracy of the model and reduce the computational effort,Competitive Adaptive Reweighting Sampling (CARS),Uninformative Variable Elimination (UVE),Successive Projections Algorithm (SPA),Monte Carlo Uninformative Variable Elimination (MCUVE) and Genetic Algorithm (GA) for feature extraction of terahertz spectral information. The corresponding Least squares support vector machine (LS-SVM),Random forest (RF) and K-nearest neighbor (KNN) qualitative discriminant models are established to detect and identify the full and empty regions of almonds.In addition,the terahertz feature image was to jpg format and then to RGB format,the shell image and kernel image were separated by G-channel extraction and image binarization,and the ratio of shell kernel pixels in the terahertz feature image was detected. The image of shell and kernel were separated by contour extraction and image binarization. The actual plumpness was the ratio of shell kernel pixels in the original image. The terahertz transmission imaging techniques feasibility for detecting the almonds plumpness was proved by calculating the error between the detection plumpness and the actual plumpness. The established KS-GA-RF model had the best identification effect,with an accuracy of 98.21%. According to the ratio of shell and kernel pixels,the corresponding detection and actual fullness were calculated,respectively,with an error of 16%. This study verified that combining terahertz graph and spectrum could well realize the visual detection of inner kernel plumpness of P. chinensis,providing a new idea for the accurate classification of almonds. It also delivers a theoretical reference for terahertz imaging to detect the plumpness of other nuts and has significant application value.

Soluble solid contents (SSC) are an important indicator of apple quality and maturation and can be used for quality analysis and ripeness prediction. In this paper, 552 samples of Sugar Core Red Fuji apples from Aksu of Xinjiang Province were picked at equal intervals of three days from the fruit swelling and setting stage to the complete mature stage, the visible near-infrared spectroscopy (vis-NIRS) of the samples at 380 to 1 110 nm were collected respectively, and the SSC were measured. Then, the collected discrete data of vis-NIRS were transformed into spectral curves using the basis function smoothing method, i. e., function-type data, and respectively, the functional linear regression model was established with Vis-NIRS curves, first-order derivative curves, and second-order derivative curves as functional explanatory variables and SSC as scalar response variables. To confirm and analyze the performance of the model, partial least squares regression (PLSR), kernel support vector machine (KSVM), random forest (RF), gradient boosting tree (GBM) and deep neural network (DeepNN) were established by the original spectral discrete data after moving smooth, first-order derivative and second-order derivative pre-processing. The results show that among the 18 models, for the training set, the PLSR-dNIRmodel, KSVM-dNIR model, RF-dNIR model, GBM-dNIR model, and Deep NN-d2NIR model were outperformed the FunLR-NIR model, FunLR-dNIR model and FunLR-d2NIR model, and the Deep NN-d2NIR model was optimal (rc=0.999 6, R2c=0.998 6, RMSEC=0.074 0, RPDC=27.436 6). For the test set, the FunLR-NIR model, FunLR-dNIR model, and FunLR-d2NIR model outperformed all other models, and the FunLR-NIR model was optimal (rv=0.953 4, R2v=0.907, RMSEV=0.585 6, RPDV=3.301 7). The results of the training sets and test sets show that the kernel support vector machine model, random forest model, gradient boosting tree model, and deep neural network model are prone to overfitting. In contrast, the functional linear regression model has better generalizability. Besides, the prediction results of the three functional linear regression models (FunLR-NIR model, FunLR-dNIR model, and FunLR-d2NIR model) showed that all the models have good robustness and high prediction accuracy. The experimental results showed that the functional linear regression models combined with vis-NIR spectroscopy and functional data analysis could successfully and effectively achieve the prediction of soluble solid contents of apples at the ripening stage.

Hg2+ is one of the most toxic heavy metal ions,which can cause air,soil,and water pollution,seriously damaging human health. Therefore,developing effective analytical methods to detect Hg2+ in environmental systems is particularly important. Fluorescent probes have been widely used to detect Hg2+ due to their advantages,such as high sensitivity,good selectivity,fast response time,and real-time online detection. In this paper,a novel “turn-on” fluorescent probe (2-(pyren-1-yl)-1,3-oxathiolane,POX) with 1,3-oxathiolane as receptor was designed and synthesized based on Hg2+-promoted deprotection reaction of thioacetal,and1H NMR,13C NMR,and HRMS characterized its structure. The selectivity,competitiveness,concentration titration,pH titration,time dependence,the limit of detection,and recognition mechanism of POX for the detection of Hg2+ in CH3CH2OH/H2O solution were investigated. The results showed that POX could quickly recognize Hg2+ in a wide pH range and exhibited high selectivity and sensitivity. Adding Hg2+ to the solution of POX resulted in a clear fluorescence emission peak at 386 nm,indicating that POX showed a remarkable “turn-on” fluorescence for Hg2+,and its recognition process was almost unaffected by other metal ions. Fluorescence titration experiments indicated that POX had a good linear response (R2=0.999 4) in the range of Hg2+ from 0~ 6.5 μmol·L-1,with a detection limit of 0.168 μmol·L-1. The RSD of POX for detecting Hg2+ in actual water samples was less than 2.92%. The simple synthesis,easy availability of raw materials,and wide pH applicability of POX suggested that itcould be used as a potential tool for the qualitative and quantitative detection of Hg2+ in the environment.

Dendrobium nobile leaf blades nitrogen content is a decisive factor for precise fertilization. Traditional nitrogen content detection methods are time-consuming and can completely deplete samples. Efficiently detecting the nitrogen content of D.nobile leaf blades has become a growing concern for herbal medicine cultivation enterprises. To quickly and non-destructively obtain the nitrogen content of D.nobile leaf blades,this study used fresh D.nobile leaf blades as experimental samples. After obtaining their hyperspectral images in the range of 402.6~1 005.5 nm and nitrogen chemical detection values,the images underwent the extraction of regions of interest(ROI),followed by preprocessing of the spectral information within those regions learner algorithms including Partial Least-Squares Regression (PLSR),Kernel Ridge Regression (KRR),and Support Vector Regression (SVR),as well as ensemble learning algorithms including Random Forest (RF),Bagging,and Adaboost,were utilized to model the nitrogen content of D.nobile. Regression prediction models were constructed based on the full-band spectral information of fresh D. nobile leaf blades and feature bands of spectral information extracted through CARS,and the prediction accuracy was compared. The results showed that when constructing the monitoring model based on the full-band spectral information,the RF model built with spectral data preprocessed by the Savitzky-Golay filtering （SG） method had the best prediction result (R2CV=0.961 4, RMSECV=0.081 8, R2P=0.972 6, RMSEP=0.063 3),and all models achieved R2 values over 0.90. When constructing the regression prediction model based on feature bands extracted through CARS,the Bagging model had the highest accuracy and stability,with the best prediction result observed in the SG-CARS-Bagging model (R2CV=0.938 7, RMSECV=0.100 0, R2P=0.953 5, RMSEP=0.082 6),while the accuracy of the individual learner models KRR and SVR was significantly lower. The CARS algorithm feature extraction removed some important bands,improving modeling efficiency but reducing model accuracy. Therefore,when optimizing the feature parameters of the regression model,it is necessary always to consider the balance between accuracy and efficiency. The research results indicate that ensemble algorithms such as RF,Bagging,and Adaboost have higher stability and prediction accuracy than individual learner algorithms such as PLSR,KRR,and SVR. They are more suitable for analyzing and processing hyperspectral data and have obvious advantages in the nitrogen nutrition monitoring of D.nobile.

In this study,larch wood,the main timber species in Northeast China,was improved by vacuum heat treatment to explore its cellulose morphology and structural thermal response mechanism. The wet chemical method was used to extract the cellulose of untreated and heat-treated larch wood under different vacuum heat treatment conditions (180~220 ℃,6 hours). The cellulose morphology was observed by scanning electron microscope,and its chemical structure and crystallinity were analyzed by X-ray photoelectron spectroscopy and X-ray diffraction. The results showed that the cellulose extracted from untreated wood was rod-shaped and had an ordered parallel and rigid structure. After heat treatment,the wood cellulose curved,and its surface twisted. With the increase inheat treatment temperature,the surface distortion of cellulose became more significant. The X-ray photoelectron spectroscopy analysis showed that the increase in heat treatment intensity decreased the cellulose O/C ratio from 65.5% to 59.5%. The C1 content increased,and C2 and C3 decreased,respectively,indicating that the cellulose carbon chain extracted from heat-treated wood was shortened. Hemicellulose and lignin-carbohydrate formed by the condensation reaction of lignin and carbohydrate during heat treatment might remain on its surface. In addition,with the increase inheat treatment temperature,wood celluloses crystallinity increased,the proportion of cellulose Ⅱ increased,and cellulose Ⅰ decreased. Furthermore,the morphology of cellulose extracted from wood before and after heat treatment presented different structures due to the differences in the cell structure and molecular group conformation between cellulose Ⅰ and Ⅱ. This research will help to deeply understand the thermal response characteristics of wood chemical composition and provide a theoretical basis for the optimization of wood vacuum heat treatment technology.

Zn3As2 and Zn3P2 have the same pseudo-cubic lattice structure and present a wide range of application prospects in the field of optoelectronic devices because of their high electron mobility,narrow direct band gaps,and good air stability. At present,there is relatively little research on the nanostructure of Zn3As2-Zn3P2 solid solution,and Zn3(As1-xPx)2 (x=0, 0.05, 0.1) master alloys were obtained by high-pressure sintering technology,and then a variety of Zn3(As1-xPx)2 nanostructures are synthesized by chemical vapor deposition,including macro-sized nanoribbons (Length 3~10 mm; Width 1~4 mm; Thickness ~20 μm),nano sails,nanorods and nano silver hairpins. The effect of P doping on phase composition,element content,microstructure,and spectral characteristics was systematically investigated. XRD results showed that the main phase of Zn3(As1-xPx)2 macroscopic nanoribbon samples was α′ phase. With the increase of P doping contents,the (224) diffraction peak shifted to the right,indicating a decrease in the lattice constant. Electron spectroscopy analysis showed that the actual content of P in these nanoribbons corresponding to x=0.05 and x=0.1 Zn3(As1-xPx)2 master alloys was x=0.026 and x=0.062,respectively. The microstructure analysis showed that the growth mode of Zn3As2 macroscopic nanoribbons was along the 〈221〉 crystal face rhombus-shaped layer-like growth and that P doping led to a reduction in the macroscopic size of the nanoribbons,accompanied by growth mode change from rhombus-shaped layered growth to nanoparticle stacked layered growth. Raman spectra of the nanoribbon samples showed characteristic peaks at 79, 97, 198, 320, 428 and 1 107 cm-1. P doping led to a blue shift of 1 107 cm-1 characteristic peaks in Raman spectra,and 1 101 and 1 599 cm-1 characteristic peaks in Fourier infrared spectroscopy (FTIR),and 300,422,and 635 nm characteristic peaks in PL spectra were also blue-shifted. The linear relationship between photocurrent and voltage of Zn3As2 and Zn3(As0.974P0.026)2 nanoribbons indicate good ohmic characteristics,and the photoresponse of Zn3(As0.974P0.026)2 nanoribbons after P doping shows the highest sensitivity under 900 nm conditions.

Laser-induced breakdown spectroscopy (LIBS) is an emerging atomic spectroscopy technique that has the advantages of low sample pre-treatment and rapid, in situ, and simultaneous multi-element measurements. LIBS demonstrates good prospects in the field of coal analysis. In recent years, chemometric and machine learning models have been widely used to improve the quantitative accuracy of LIBS in coal analysis. Generally, these models rely on a certain number of training samples to ensure the reliability of the prediction results. However, obtaining the certified content (label information) of coal samples used for model training requires traditional chemical analysis, which is complex and time-consuming. This may lead to insufficient training samples and poor model performance. To tackle the small sample problem in LIBS-based coal analysis, this work proposes a semi-supervised learning method based on the ensemble of multiple models. 5 baseline models are first established based on the initial training set, including multiple linear regression (MLR), partial least squares regression (PLSR), locally weighted partial least squares regression (LW-PLSR), support vector regression (SVR), and kernel extreme learning machine (K-ELM). The unlabelled data are processed using the 5 models, and 5 prediction values are obtained. For each unlabelled sample, the standard deviation of the 5 prediction values is calculated, and the unlabelled sample corresponding to the smallest standard deviation is added to the training set. Its pseudo label is the average of the 5 prediction values. As the training set is iteratively expanded, its corresponding training model is updated. The final training model is optimized and used to analyse the test samples. The proposed method is tested on a coal dataset containing 20 training samples, 39 test samples and 280 unlabelled samples. The results show that the proposed method improves the coefficient of determination (R2) for content prediction of fixed carbon, ash，and volatile by 0.033,0.102 and 0.118, respectively. Therefore, if the number of training samples is insufficient, semi-supervised learning can effectively improve the accuracy and reliability of LIBS quantification.

Animal glue,eggs,and milk were often mixed in ancient China as cementitious materials in painted cultural relics. To explore the interaction of mixed binders and the stability under light aging conditions,to understand the deterioration mechanism of polychrome relics,the stability and protein secondary structure of egg-animal glue,milk-animal glue,and milk-egg mixed binders were characterized by Scanning electron microscopy,FTIR and TG analysis. The results showed that after mixing,the above binders thermal stability and light stability were significantly increased. The infrared spectrum results show that spectral peak broadening after aging of the sample,and the amide Ⅰ band of the protein characteristic spectrum with a wavenumber range between 1 600~1 700 cm-1 shifts towards the wavenumber direction. The Gaussian fitting results of the amide Ⅰ band show that with the change of mixing ratio,the content of ordered structure in the mixed binders shows a trend of first increasing and then decreasing. In the samples with the highest ordered structure content,the random curl structure content also reaches a lower level. At the same time,when the mixing ratio of egg and animal glue is between 1∶1 and 1∶2,the content of the ordered structure is higher than any single binders,with a maximum content of 64.87%,indicating its strongest stability. The content of the secondary structure of protein in the aged binders has changed. The α-helix structure shows a downward trend,while the random curl structure shows a significant increase. The α-helix structure in several mixed binders is still higher than that of a single binder material,indicating that the mixed binders have a certain degree of stability. The results of Thermogravimetric analysis also showed that the thermal stability of the mixed binders before and after aging was higher than that of the single binders,presumably because the protein in the mixed system combined with the lipid in eggs and milk and the interaction between protein and lipid further promoted the stretching and rearrangement of protein molecular structure,which improved the stability. From the stability perspective,the reason for the mixed-use of adhesive materials in painted cultural relics was explained,providing a reference for studying ancient adhesive materials aging mechanism and protection methods.

Infrared thermography is a widely used nondestructive testing method in cultural relic protection. In current works,active infrared thermography is often used to evaluate cultural relics disease conditions and structural defects. In contrast,passive infrared thermography is not paid enough attention,and the dynamic relationship between cultural relics and their environment is rarely directly investigated. Taking the crack disease of Yuan Dynasty architectural murals in Fengguo Temple,Yixian County,Liaoning Province,as an example,this paper explored the feasibility of passive infrared thermography to study the cross-coupling effect between cultural relics and their existing environment. Based on obtaining more comprehensive information on the eave wall structure by taking thermal infrared images,continuous infrared thermography monitoring was carried out on the longitudinal long cracks in the middle of the east and west walls as well as the whole N1 mural of the north wall,and temperature measurement was carried out on the cracks,the ordinary wall surface and the wall surface where the wood keels located. Then,the cumulative anomaly and wavelet analysis methods were used to analyze the changing trend and periodic temperature fluctuation at these characteristic positions. The monitoring and analysis results were qualitatively discussed from the perspective of heat transfer,and the influence modes at different positions of the mural were distinguished. Combined with the characteristics of the cracks,the formation process was preliminarily judged. The results show that the mural has an uneven multi-layer structure,and there are differences in the heat transfer pathway and structural strength. The temperature trends of the east and west walls are similar,but there are obvious differences in fluctuation signal and energy. The overall fluctuation energy of cracks is greater than that of walls and is positively correlated with the degree of development of the cracks. The fluctuation energy of the first main cycle comes from the diurnal variation of temperature. In contrast,the secondary main cycles may be related to random disturbances in the heat transfer process. The energy fluctuation during the daytime and after sunset plays a key role in the heat transfer effect. The heat transfer process at the wood keels and the cracks is faster,so the high-temperature centre and the transverse heat transfer trend are formed over time. The occurrence of crack disease results from the mechanical strain differences of material under the influence of temperature change. There are two heat transfer pathways:one is the heat conduction with the eave wall as the medium,and the other is the convective heat transfer through the air channel at the eave column; the former mainly induces the formation of several parallel transverse short crack groups in the units,while the latter promotes the formation of longitudinal long cracks at the eave columns. This paper can provide a reference for evaluating the heat transfer process of other cultural relics in complex environments.

Identifying celestial spectra is essential for making new astronomical discoveries and conducting detailed studies of celestial objects. The LAMOST DR8 v1.0 release of low-resolution spectral data contains approximately 530 000 spectra named “Unknown”. The reason is that they have no category labels. And 88.56% of these spectra have signal-to-noise ratios between 0 and 10. Therefore,the effective output of LAMOST will increase if we analyze these spectra. In this paper,we propose an ODS-YOLOv7 model to deal with the problem of the “Unknown” spectral classification. It is an end-to-end category prediction model and is suitable for one-dimensional spectra. We also add a one-dimensional convolutional attention module to improve the accuracy of spectra recognition. After training on a set of known category spectra with signal-to-noise ratios between 0 and 10, the ODS-YOLOv7 model can learn the effective features of the low signal-to-noise spectra. Thus,it can enable us to predict “Unknown” spectra. Experiments show that the model has an F1-score of 0.98,0.95,and 0.95 for the spectral identification of low signal-to-noise stars,galaxies,and quasars spectra with known labels. In the meantime,ODS-YOLOv7 obtains the best results in comparison experiments with traditional algorithms KNN,RF,DT,SVM,and deep learning algorithms 1D CNN,1DSSCNN,ResNet,DenseNet,and VIT. The experimental results also give confidence in the predictions of the ODS-YOLOv7 model for the “Unknown” spectra in DR8 v1.0,with 92% of the confidence levels above 60%. To ensure the quality of the model output,only spectral categories with a prediction confidence level greater than 99% are selected as output in this paper. Ultimately,37.19% and 47.03% of the “Unknown” spectra released in DR8 v1.0 and DR9 v0,respectively,are predicted by this model.In addition,the paper tests the accuracy of the models predictions using manual authentication. To improve the interpretability of the model,the paper takes the Grad-CAM method for two-dimensional image visualisation. It improves it into an algorithm suitable for visualising one-dimensional spectral data to predict output features. Experiments show that the model focuses on different features in the visualisation of different classes of astronomical features and that the model is good at predicting low signal-to-noise “unknown” spectral classes.

Traditional one-dimensional spectra perform poorly when classifying celestial objects with a low signal-to-noise ratio (SNR). Therefore,the paper uses two-dimensional spectra and proposes a feature fusion model called TDSC-Net (Two-Dimensional Spectra Classification Network),incorporating an attention mechanism for stellar classification. TDSC-Net employs identical feature extraction layers to get features from the two-dimensional spectra of stars,specifically from the blue and red ends. The extracted features are fused and employed for the classification task. The stellar spectral data in this experiment is selected from the LAMOST (the Large Sky Area Multi-Object Fiber Spectroscopic Telescope) database. They are using Z-score normalization on the spectra to reduce convergence difficulties caused by significant variations in spectral flux values and evaluating the model performance by four metrics:Precision,Recall,F1-score,and Accuracy. The experiments consist of two parts:In the first part,TDSC-Net is employed to classify A,F,G,K,and M-type stars to validate the reliability of using two-dimensional spectra for multi-class stellar classification. In the second part,the two-dimensional spectra are classified based on different SNRs to investigate the impact of SNRs on classification accuracy. The first parts results show that the five-class classification accuracy reaches 84.3%. The classification accuracies of A,F,G,K,and M types are 87.0%,84.6%,81.2%,87.4%,and 89.7%,respectively. These accuracies are higher than the results obtained from one-dimensional spectra classification after spectra extraction. The results of the second part indicate that even in the low SNR (SNR<30),the accuracy of two-dimensional spectra classification can still reach 78.9%. Once the SNR surpasses 30,the impact of SNR on spectra classification becomes less significant. These provide evidence for the importance of using two-dimensional spectra classification in low SNR and demonstrate the effectiveness of TDSC-Net in stellar spectra classification.

We apply the Micro-XRF,which can be used to in situ non-destructively study the element distribution of the skeleton and surrounding matrix in vertebrate fossils to scan the holotype and paratype of the Middle Triassicmarine reptile Mixosauruspanxianensis (~244 Ma),visualizing the overall element distribution of the specimens. Additionally,the paratypes regions of interest are tested using a handheld X-ray fluorescence spectrometer as an adjunct. The research results show that the bone and matrix elements present a different distribution pattern. The skeleton clearly controls Ca,P,Sr and Y. The matrix where the fossil is preserved is rich in Ca,K,Fe,and Mn. In addition,Zn is variouslydistributed in different fossil bone parts of the paratype specimen,where the Zn content is higher in the trunk region than in the skull. In terms of fossil morphology,the maps clearly resolve the fossil morphology. The right forelimb and gastralium of the paratype specimen,which are invisible in regular light,are particularly well-resolved by the elemental maps. At the same time,the calcareous matrix and bone can be distinguished better. In taphonomy,comparing the elemental features of fossilized marine and terrestrial fossils demonstrates how the burial environment affects the distribution characteristics of certain elements. Th,Ce,Cu and Sr responded to the burial environment,while some elements related to bone,such as Ca,P,and Y,were less affected by the burial environment. The distribution of Zn in different bone regions was altered by bone development,and the paratype specimens centrums and ribs,where Zn is elevated,are likely to be in the stage of rapid ossification,indicating that the paratype specimen was subadult.

Peanut is a high-quality plant protein resource. The content of peanut protein components and subunits significantly affects its functional characteristics and determines its application range in the food field. Peanut proteins mainly include arachin and arachin. Among them, arachin contains four subunits (40.5, 37.5, 35.5, 23.5 kDa), arachin I contains three subunits (15.5, 17, 18 kDa), and arachin II contains only one 61 kDa subunit. To realize the rapid, non-destructive, high-throughput and high-sensitivity detection of peanut proteins main components and subunits, 145 high-quality peanut samples in China were used as the research object. Firstly, the portable near-infrared peanut quality rapid tester was used to collect the spectra of different peanut samples in the wavelength range of 900~1 700 nm. Then,the peanut protein components and subunits were determined by polyacrylamide gel electrophoresis. The subunit content of arachin was between 44.3% and 67.3%. The content of arachin subunits was between 35.2% and 55.7%. The 61 kDa subunit content ranged from 13.5% to 25.3%. The content of 40.5 kDa subunit was between 6.8% and 16.0%. The content of 37.5 kDa subunit was between 6.9% and 17.4%. The content of 35.5 kDa subunit was between 5.7% and 19.2%. The 23.5 kDa subunit content was between 18.7% and 27.4%. The content of the 18 kDa subunit was between 5.9% and 11.7%. The content of the 17 kDa subunit was between 6.9% and 13.6%. The content of the 15.5 kDa subunit was 4.5%~11.9%. The near-infrared spectral models of two protein components and two subunits (arachin, arachin, 37.5, 23.5 kDa) were optimized by comparing seven spectral pretreatment methods, including normalization, first derivative (FD) and second derivative (SD), baseline calibration, detrend, multiple scattering corrections (MSC) and data element resolution (Deresolve), combined with principal component analysis (PCA) and partial least squares (PLSR). The near-infrared spectroscopy models of six subunits (61, 40.5, 35.5, 18, 17, 15.5 kDa) were constructed, and the simultaneous detection of the above 10 indicators was realized. The results showed that the calibration sets correlation coefficient (Rcal) 0.90～0.96, and the corrected root mean square error (SEC) was 0.25%～1.27%. The prediction sets correlation coefficient (Rcp) was 0.76～0.96, and the root mean square error of prediction (SEP) was 0.50%～1.81%. It has good predictive ability and can be used for rapid detection of protein components and subunit content in peanut varieties, which provides a new method for fast evaluation of peanut protein quality.

Due to the enormous market value of geographically iconic,adulteration or fraud often occurs. Therefore,to ensure the brand benefits from geographically iconic rice and consumer rights,it is important to identify polished rice varieties accurately. Near-infrared spectroscopy is a common method to distinguish polished rice varieties. The varieties can be classified by extracting the different features of different types in near-infrared spectroscopy. However,there are some problems in the existing studies,such as insufficient characteristic wavelength selection performance and insufficient discrimination accuracy for specific varieties,which limit the improvement of the discrimination accuracy of polished rice varieties based on the near-infrared spectroscopy method. In response to the above problems,this paper studies the optimisation of milled rice variety identification based on near-infrared spectroscopy from the two aspects of characteristic wavelength selection and variety identification strategies for four types of rice,Wuchang,Xiangshui,Koshihikari,and Yinshui in Northeast China. First,permutation entropy (PE) and adaptive sliding window (ASW) were combined to improve the feature wavelength selection performance. An adaptive sliding permutation entropy (ASW-PE) based method for selecting the characteristic wavelength of polished rice spectrum was proposed and compared with the traditional algorithm. Secondly,a discriminant strategy based on the discriminant objective was proposed to improve the discriminant accuracy of different specified varieties. By studying the matching optimisation of the spectral preprocessing algorithm and classification modelling algorithm,a discriminant process of “specified cultivation-selected model-selected algorithm” was established. Experimental results show that using the adaptive sliding permutation entropy algorithm proposed in this article to select characteristic wavelengths can reduce the milled rice variety discrimination error by at least 50% compared with the traditional algorithm; using the milled rice variety judge strategy based on the discrimination target proposed in this article. Compared with the conventional judge strategy based on fixed models,the discrimination accuracy can be improved by at least 2.5%.

Donkey meat has excellent flavor and rich nutrition and is in high price and low supply. The problem of cooked donkey meat adulterated with other meat,such as horse and mule meat,needs to be solved urgently. To realize the qualitative and quantitative analysis of cooked donkey meat samples of different adulteration ratios,horse and mule meat samples were used to degrade donkey meat. The gradient was 10%,and the donkey meat contents were 0%~100%. Spectra of samples were collected in the range of 4 000~12 500 cm-1. The methods of linear discriminant analysis,support vector machine,and generalized regression neural network combined with smoothing algorithm (5 points,15 points,25 points),multiplicative scattering correction (MSC),standard normal variable (SNV),Baseline correction,normalization,and Detrend were used to establish the NIR discriminant models of adulterated cooked donkey meat samples. Partial least squares regression (PLSR) and backpropagation (BP) were used to establish quantitative models to determine the content of donkey meat in adulterated samples. For minced after cooked meat samples,the results of SNV pretreatment combined with a support vector machine were optimal,and the discriminant accuracy of the calibration set and prediction set was 98.70% and 94.78%. The results of Detrend pretreatment combined with linear discriminant analysis were optimal for minced before cooked meat samples. The discriminant accuracy of the calibration and prediction sets reached 98.47% and 96.23%,respectively. Compared with the PLSR model,the BP model obtained better results,with a higher coefficient of determination (R2),relative percent deviation (RPD),and lower root mean square error (RMSE). For the adulterated samples of minced after cooked meat samples,the BP model of the donkey and mule adulterated samples was better after Detrend pretreatment. R2,RMSE,and RPD of the cross-validation set and prediction set were 0.971,0.067,5.844,0.980,0.086,6.984,respectively. After normalized treatment,the results of BP model of donkey and horse adulterated samples were optimal,and the parameters were 0.997,0.032,18.026,0.982,0.089,7.454,respectively. For the adulterated samples of minced before cooked meat samples,the results of the BP model with Detrend pretreatment were better,and the optimal quantitative model parameters of donkey and mule adulterated samples were 0.982,0.041,7.470,0.986,0.103,8.452,respectively. The best model parameters of donkey and horse adulteration were 0.986,0.036,8.348,0.961,0.101,and 5.044,respectively. The results show that the NIR spectroscopy combined with different modeling algorithms can realize the rapid,nondestructive detection of different donkey meat contents. The methodology can be used for future qualitative and quantitative analysis of cooked donkey meat adulteration.

Arc additive manufacturing has the advantages of high deposition efficiency, low cost, and unrestricted deposition shape and size. However, the dimensional accuracy of the formed parts by arc additive manufacturing is still difficult to guarantee accurately. The size of the deposited layer is one of the criteria for evaluating the quality of component formation, and it is crucial for judging processing quality and defect compensation. Therefore, real-time monitoring of the variation of the deposited layer size during the arc additive manufacturing process is of great significance for optimizing process parameters and ensuring the formation quality of additive manufacturing components. Arc spectral information can reflect the arc state, closely related to the forming quality. Therefore, studying the relationship between arc spectrum and deposited layer size is very important. This study used titanium alloy (TC4) material as the substrate and welding wire, and the arc plasma spectral signals were studied to investigate the correlation between GTAW additive arc spectral characteristics and deposited layer size. Firstly, a spectral acquisition system was constructed to collect arc spectral signals at different positions above the molten pool, around the molten pool, and below the tungsten electrode. Secondly, based on the principle of high spectral line separation, the wavelengths of 404.20 nm TiⅠ spectral line, 416.36 nm TiⅡ spectral line, 420.20, 434.81, 480.50, and 487.98 nm ArⅡ spectral lines, and 696.54 and 794.82 nm ArⅠ spectral lines were selected. The peak intensity features of these spectral lines were extracted, and the ReliefF algorithm was used to explore the correlations between different spectral line intensity features and deposited layer size. The results showed that among all the spectral lines, the peak intensity features of the 404.03 nm TiⅠ element spectral line, 416.36 nm TiⅡ element spectral line, and 794.82 nm ArⅠ element spectral line above the molten pool had a strong correlation with the deposited layer size. At the same time, combining the ReliefF algorithm, the correlations between peak intensity features of different positions and deposited layer size were studied. The results showed that the spectral line with the strongest correlation between the molten pool above and the deposited layer size was the 696.54 nm ArⅠ spectral line, and the spectral line with the strongest correlation between the molten pool around and below the tungsten electrode and the deposited layer size was the 794.82 nm ArⅠ spectral line. Furthermore, to reduce random errors, the PCA algorithm was used to fuse the intensity features corresponding to the three spectral lines with the highest correlation with the deposited layer size, and a new fusion feature was obtained. Then, combining the K-nearest neighbors algorithm, a deposited layer size prediction model was established. The fused feature and the feature values of samples with the highest correlation with the deposited layer size at different positions were extracted, and the accuracy of predicting the samples category based on these four features was calculated. The results showed higher accuracy for predicting the deposited layer size based on the fused feature. Finally, based on this new feature, combined with the threshold segmentation method, dynamic monitoring of the variation of the deposited layer size was achieved.

The Roland circle optical system has many advantages. Still,due to using a planoconvex lens in the system,the imaging points that have undergone dispersion also generate a certain amount of offset distance near the Roland circle,resulting in the imaging points of each characteristic wavelength arranged sequentially on a curved surface. Due to the current use of new collection devices such as linear array CMOS or CCD,where the photosensitive area of these sensors is a flat surface,serious chromatic aberration is caused in the results. This article proposes a chromatic aberration correction method using the weighted least squares method to fit the image plane and calculate the two correction parameters of the center offset distance and offset tilt angle of the system image plane. Based on the original Roland circle optical system,these two correction parameters are introduced for chromatic aberration correction,thereby determining the optimal position of the linear array photoelectric sensor image plane and achieving high resolution. In addition,elements such as C,P,and S are very important in the VUV spectrum. The chromatic aberration correction method based on weighted least squares proposed in this article can achieve optimal resolution by increasing the weight proportion of characteristic wavelengths of such elements. Finally,optical simulation was conducted using Zemax software to complete the simulation and optimization of the Roland circle optical system. The imaging width simulation of spot diagrams in the 170~410 nm wavelength range and the spectral images of each elements characteristic wavelengths collected on CMOS sensors were provided. The resolution in practical applications was calculated,and the illuminance spectral images showed that their resolution could be maintained at 0.02 nm,close to the optimal theoretical resolution of the system. The research results of this article indicate that using the weighted least squares method to fit the image plane can achieve chromatic aberration correction for Roland circle optical systems and achieve high resolution.

Oil slick pollution seriously influences the ecological environment. Many countries have invested a lot of money and workforce in investigating oil slick on water. Thickness measurement of oil film on water with high precision is important for the prevention and treatment of oil slicks. Here,diesel was taken as the research object. Fourier Transform Infrared Spectrometer obtained theabsorption spectra of diesel at different temperatures (293/298/303/308/313/318 K) and water at room temperature (298 K). It was found that diesel and water were absorbed in the near-infrared region (6 000.0~11 000.0 cm-1),and the absorption spectra of diesel did not shift with temperatures. The wavenumberν1 (8 381.6 cm-1) with the maximum absorption coefficient of diesel was selected to establish the inversion model of oil film thickness on water based on a single-wavelength. ν1 and the wavenumber ν2 (8 918.7 cm-1) with an absorption coefficient around zero were selected to establish an inversion model of oil film thickness on water based on dual-wavelength. Furthermore,the absorption coefficients of water at these two wavenumbers were very small and had little influence on the measurement results. A calibration tool with known oil film thicknesses (0~1 000 μm) was employed to validate the measurement precision of the inversion models. It was found that the average relative deviations between the film thicknesses obtained by the two models and the known values were 36.4% and 2.5%,respectively. The maximum relative deviations were 44.7% and 3.7%,respectively. The maximum standard deviations were 7.0 and 5.6 μm,respectively. It revealed that the dual-wavelength model was superior to the single-wavelength model. On this basis,a novel film thickness measurement system for oil on water with dual-wavelength absorption spectroscopy was developed. The temporal resolution of the developed system was 0.03 s. The variations of oil film thicknesses after dropping a certain volume (1 mL) of diesel on the water surface were investigated,and the ultrasonic pulse echo-method was simultaneously employed to compare with the developed system. Each drop of diesel was measured 10 times by the two methods. A total of 20 groups of oil film thickness data (16~35 mL) were measured,and the average values of the results measured by the two methods were compared. It showed that the average relative deviation of the oil film thicknesses measured by the two methods was 2.5%. Moreover,the maximum relative deviation was 3.7%. The maximum standard deviation of oil film thicknesses obtained by 10 repeated measurements of dual-wavelength absorption spectroscopy was 6.4 μm. In summary,the measurements of the thickness of oil film on water could be achieved by the developed high-precision system. It had the advantages of interference-free,fast-response,compact-structures,etc.,and could be applied to different types of oil slick measurement. The developed system was expected to provide new ideas and scientific guidance for the monitoring,preventing and treating oil spills.

The heat treatment of ruby belongs to the optimization treatment,and borax is commonly used as the flux. In the market,heat-treated rubies are sold naturally and have a high value,but the amount of borax in the crack will affect the price. Therefore,in this paper,the borax in this kind of ruby was quantitatively studied,and NanoVoxel-4000X X-ray computed tomography was used to scan and analyze the sample,and 2D images of borax filling in fractures were obtained. The grayscale values of light and shade were used to characterize the different densities of the material in the crack,and different colors characterized the 3D images of the borax filling distribution. By comparing the color scale of thickness,it can be seen that the overall content of borax is small. At the same time,the histogram of the borax thickness distribution of samples shows that the thickness of borax in all fractures is less than 130 μm,and the thickness-to-volume ratio of all borax is less than 10%,which also indicates that the filling amount is small. Finally,Avizo 9.0 software was used to segment the ruby sample and borax content to obtain the actual volume of the two. Then,the percentage of borax content in the total volume of the ruby could be calculated. The proportion of borax in the sample is 10.69%,which is not only an important index in the quantitative detection of borax in heat-treated rubies but also provides data support for the study of the quantitative classification standard of borax in heat-treated rubies.

An in-depth study of a stellars spectrum provides insight into its chemical composition and physical properties. Stellar spectrum classification is an important direction in stellar spectrum research. With the emergence of massive stellar spectrum data,artificial classification cannot meet scientific research needs. Based on this,this paper constructs the SATS algorithm,which realizes the automatic classification of F,G,and K-type stellar spectra. Firstly,the SATS algorithm uses singular value decomposition (SVD) to denoise the normalized stellar spectra. Then,the SATS algorithm performs feature extraction on the stellar spectrum. The feature extraction layer consists of six modules:Incremental principal component analysis (IPCA),nuclear principal component analysis (KPCA),sparse principal component analysis (SparsePCA),FactorAnalysis,independent component analysis (FastICA) and Transformer(the six modules are collectively referred to as Analysis module),to ensure that the variance contribution rate is above 0.95,IPCA,KPCA,SparsePCA,FactorAnalysis and FastICA extract the stellar spectral features into 300 dimensions. Finally,the stellar spectra are fed into the SoftMax layer for automatic classification. SATS algorithm combines multiple analysis modules to improve the accuracy of classification further using a single analysis module. Once again,the combination of Transformer modules and multiple Analysis modules improves classification accuracy. The most significant advantage of the SATS algorithm is that it performs multiple feature extraction on the stellar spectrum,which retains the stellar spectral information to the maximum extent and minimizes the information loss by different feature extraction methods. The final classification accuracy of the SATS algorithm is 0.93,which classification accuracy is also higher than that of the hybrid deep learning algorithms CNN+Bayes,CNN+Knn,CNN+SVM,CNN+Adaboost and CNN+Adaboost 0.86,0.88,0.89,0.87,0.89.

A new method for multi-parameter water quality detection with variable optical range has been proposed, addressing scientific and technical challenges in the current national standard analytical methods. This method utilized ultrasound and micro-nano bubble (US-MNB), continuous spectrum analysis, and sequential injection analysis (SIA). A water quality multi-parameter detection system was designed and validated for the analysis of total phosphorus (TP), chemical oxygen demand (COD), ammonia nitrogen (NH3-N), and hexavalent chromium (Cr6+). The feasibility of the new method was verified. The systems core consists of a digestion chamber integrating ultrasound and micro-nano bubbles and a detection chamber with an adjustable optical range. This structural design enabled rapid digestion and stable detection. The multi-parameter detection process was optimized based on national water quality testing standards. Spectrophotometry and sequential injection analysis techniques were employed for continuous spectral detection of the four water quality parameters. Firstly, the TP was digested by a US-MNB with strong oxidants at normal temperature and pressure. At the same time, the complexes were directly determined by spectral scanning after the color reaction of NH3-N in the detection chamber, and thenthe TP was determined after digesting. Similarly, the compounds in the detection chamber after the chromogenic reaction of the Cr6+ were directly determined by spectral scanning. At the same time, the COD was digested, and then the COD was measured after digestion. The time used for the whole detection process was greatly reduced, and the determination of multi-parameter of water quality can be completed automatically in a short time, significantly improving the efficiency of detection. In this system, four water quality parameters(TP, COD, NH3-N, and Cr6+)were measured, combined with the least squares method to establish a regression model, fit the regression equation and calculate the correlation coefficient, and plot the concentration-absorbance standard working curve of each parameter. The results showed that the TP standard working curve had a fitting coefficient ≥0.984 5, with a positive correlation between concentration and absorbance. The repeatability (RSD) ranged from 3.05% to 3.62%, and the spike recovery rate was 97.8% to 103.6%. The COD standard working curve had a fitting coefficient ≥0.998 7, with a negative correlation between concentration and absorbance. The repeatability (RSD) ranged from 2.12% to 2.74%, and the spike recovery rate was 98.7% to 104.7%. The NH3-N standard working curve had a fitting coefficient ≥0.995 3, with a positive correlation between concentration and absorbance. The repeatability (RSD) ranged from 3.41% to 3.59%, and the spike recovery rate was 99.2% to 102.4%. The Cr6+ standard working curve had a fitting coefficient ≥0.993 8, with a positive correlation between concentration and absorbance. The repeatability (RSD) ranged from 3.51% to 3.92%, and the spike recovery rate was 98.9% to 109.3%. The system accurately determined the content of TP, COD, NH3-N, and Cr6+ in water samples and demonstrated excellent stability and reliability. The studyon the multi-parameter detection method with variable optical range in water quality based on US-MNB is of great significance in broadening the application of spectroscopy in the field of rapid detection of water quality multi-parameters and improving detection efficiency.

The estimation and correction of long-term radiometric degradation in remote sensing instruments are crucial for improving the stability and accuracy of satellite remote sensing applications. This study used long-term observation data of FY-3D/MERSI-Ⅱ over stable targets in deserts and DCC (Deep Convective Clouds) to establish a function relationship model between instrument radiation response and time. We further obtained estimates of the best radiometric response degradation and radiometric calibration coefficient using the inverse variance weighted average method to integrate the results of two stable targets. The radiometric degradation tracking resultsshow that from the beginning of FY-3D/MERSI-Ⅱs in-orbit to September 2022, there is significant degradation in three blue light bands (band 1, bands 8—9 with wavelengths less than 500 nm) and four near-infrared-shortwave infrared bands (bands 5—7, band 19 with wavelengths greater than 1 000 nm), with annual degradation rates ranging from 1.47% to 4.32%. Precision testing of L1 products reveals that the operational calibration bias of seven significantly degraded Bands exceeded ±5%, with bands 5 and 8 having the most significant bias, about -20%. After applying the radiometric calibration coefficient sequence obtained in this study, the radiometric calibration precision remains stable over time, with calibration bias maintained within ±3%. The accuracy validation results of L2 products show that the use of newly obtained calibration coefficients has improved product accuracy compared to operational products. These results indicate the effectiveness of the radiometric calibration method established in this study.

Restricted by spatial resolution and detector level,traditional hyperspectral image target detection algorithms focus more on quantitative processing based on spectral analysis. In recent years,with the development of ground and near-ground imaging platforms and spectral imaging technology,land-based hyperspectral images have realized the unification of high spatial and spectral resolutions. Compared with hyperspectral remote sensing images,land-based hyperspectral images have a higher spatial resolution,and their targets are characterized by rich details and large scales so that the geometric shape information and fine spectral information of the targets can be utilized in the target detection task at the same time. Constrained energy minimization (CEM) is a classical target detection algorithm for hyperspectral images,which is suitable for the case that specific components account for a small proportion of the total variance of the image,highlighting certain target information to be detected and suppressing the background information,to achieve the effect of separating the target to be detected from the image. However,CEM is sensitive to the targets scale,and the algorithms detection effect decreases significantly as the number of target pixels increases. This problem is because CEM is based on the assumption that the target spectral information is excluded from the statistical background. Still,it is difficult to exclude the target spectral information in advance. Instead,it directly counts the spectra of each pixel of the full-domain image to approximate instead of the background spectra. To solve the problem that CEM is ineffective in the detection task of larger targets and to improve the algorithms target detection capability in land-based hyperspectral images,this paper proposes a CEM method based on spatial inspection guidance (Space inspection guidance CEM,SIG-CEM). The method first analyzes the acquired hyperspectral images to be measured by principal component analysis,feeds the first principal component image into the spatial target detection model,and frames the target using the coordinate information obtained from the detection results. Then,the image elements containing the target in the framed region are removed when the autocorrelation matrix in the CEM is obtained,thus effectively reducing the suppression of the target. The experiments using publicly available remotely sensed hyperspectral images and measured land-based hyperspectral images show that the SIG-CEM algorithm can avoid the influence of the traditional CEM algorithm in which the target signal participates in the operation as a background signal on the detection results. In the experiments on the public dataset,compared with other traditional target detection algorithms,the AUC value of the SIG-CEM algorithm reaches 0.973 7,which effectively improves the accuracy of target detection; in the experiments on the measured land-based hyperspectral image data,the AUC value of the SIG-CEM is improved by an average of 0.055 compared with that of the CEM. At the same time,the experiments,to a certain extent,verify that the SIG-CEM algorithm has strong robustness and applicability for different types of hyperspectral images. This study proposes a target detection method specifically for land-based hyperspectral images,which promotes the development and application of land-based hyperspectral images in target localization and identification in the future.

With the rapid development of industrialization and social economy,water pollution and deterioration of water sources are increasingly aggravated,and effective water quality monitoring is an important prerequisite for water source protection. Miyun Reservoir is an important surface water source in Beijing,which plays an important role in protecting water safety in the capital. In order to monitor the water quality parameters and pollution degree of Miyun Reservoir more accurately,this study used four phases of UAV hyperspectral remote sensing data to construct a water quality parameter retrieval model based on a deep neural network algorithm. Total nitrogen (TN) and total phosphorus (TP) water quality parameters in Miyun Reservoir were retrieved. Firstly,the hyperspectral image dimensionality reduction processing based on the recursive feature elimination method was used,and the spectral data and groundwater quality monitoring data were superimposed. The network structure parameters,such as the number of hidden layers and the number of ganglion points,were determined by minimizing the error in the training process. Then,the migration method gradually expanded the network from knowledge source domain to network,and the water quality parameters of TN and TP concentration in Miyun reservoir were trained and verified. Finally,the water quality parameters of Chaohe Dam and Baihe Dam in Miyun Reservoir were retrieved to reveal the spatio-temporal evolution of the main water quality parameters. The results show that ① the R2 of the TN and TP concentration retrieval models constructed in this study are 0.835 5 and 0.770 3,and the MSE is 0.015 3 and 0.000 8. The Ensemble Deep Belief Network (EDBN) model based on random subspace has a better retrieval effect on water quality parameters. ②TN concentration in Miyun Reservoir fluctuates with seasons,with a low concentration in summer and a relatively high concentration in autumn. The change in TP concentration is relatively stable,indicating that the control effect of phosphorus pollution in the surrounding area of Miyun Reservoir is good.③The water quality of the Baihe Dam was better than that of the Chaohe Dam. The seasons obviously affected the changes of the former,while the latter was significantly affected by human activities. The TN concentration of Miyun reservoir was in Class III,and the TP was generally in Class II. The water quality can meet the standards of drinking water sources,but it is still necessary to strengthen the supervision of nitrogen and phosphorus pollution. These results will provide an important scientific basis for efficiently monitoring water quality and water resources protection in the Miyun reservoir.

With the opening of the Arctic shipping route,the number of vessels traveling to and from polar ice regions has been increasing yearly,leading to an increased risk of oil spills in the ice zone. Difficulties in cleanup and long-lasting pollution characterize oil spills in icy areas. Therefore,the development of fast and accurate monitoring methods has become an important approach to improving cleanup efficiency and reducing pollution hazards. Remote sensing technology has been widely applied in monitoring oil spills in open waters,but there is relatively less research on monitoring oil spills in ice-covered seas. In particular,there are few reports on the reflection spectral characteristics of oil-contaminated sea ice and their variations with viewing angles.In this study,through simulated experiments of oil spills on sea ice,visible-near-infrared reflection spectra of oil-contaminated sea ice were measured at different observation zenith angles and relative azimuth angles. Measurements were taken at intervals of 10° for zenith angles ranging from -50° to 50°,and relative azimuth angles included 0°,90°,180°,and 270°. Analyzing the spectral standard deviation before and after ice pollution,the wavelength of 560 nm with the most significantdifference was selected as the characteristic wavelength for identifying oil-contaminated sea ice. This study constructed a kernel-driven model to explore the relationship between the reflectance difference of the characteristic wavelength and the geometric variations of observations. This Ross Thick-Roujean-r-RPV model considered the forward scattering characteristics of sea ice. The model was tested using measured data,and the fitting errors in the principal plane and vertical principal plane were 0.004 62 and 0.004 16,respectively,showing better fitting performance than commonly used kernel-driven models such as Ross Thick-Li sparse,Ross Thick-Li sparse R,Ross Thick (QU)-Roujean,and Ross Thick (QU)-Li sparse R-r-RPV. Using this model,the study further simulated the angular effects of the reflectance difference in the characteristic wavelength of sea ice before and after oil pollution under different observation geometries. The results showed that under the same observation geometry,there were differences in the reflectance spectra of sea ice before and after pollution,with polluted sea ice having lower reflectance than clean sea ice. Additionally,clean sea ice peaked in the wavelength range of 1 013~1 196 nm,disappearing after pollution. When the azimuth angles of observation were different,there were also differences in the reflectance of sea ice,characterized by an increase in the forward direction with increasing observation angles and a decrease in the backward direction with increasing observation angles. In the principal plane direction,the reflectance increased initially and then decreased with increasing observation angles. The largest spectral difference was observed at a zenith angle of 50° and a relative azimuth angle range of 250°~290°,which is most favorable for extracting oil spills in sea ice. The research findings of this study can provide a reference for the selection of bands and observation geometries for monitoring sensors of oil spills in icy regions for ships.

In natural environments,vegetation communities often demonstrate vertical structure due to competition and natural selection among plant groups and between populations and their environment. Different growth stages of the same plant also exhibit various structural and biochemical parameters in the vertical dimension. Detecting these three-dimensional spatial distribution features enables quantitative assessment of ecological environments in three dimensions,crucial for estimating forest carbon reserves and biodiversity conservation. Traditional passive hyperspectral remote sensing and lidar techniques face significant limitations in vertical vegetation profiling. However,the emergence of Hyperspectral LiDAR (HSL),a new type of sensing instrument,offers a fresh approach to studying the vertical distribution of physiological and biochemical parameters in vegetation. Yet,due to hardware constraints,the suitability of HSL under unmanned aerial platforms for complex three-dimensional forest scenes remains insufficiently explored. This paper begins with laboratory experiments using a prototype HSL to conduct small-scale indoor measurements with torch flower plants as the target to verify its integrated capability in extracting spatial and spectral information. Subsequently,the three-dimensional radiative transfer model LESS was used to simulate forest scenes with vertical heterogeneity. The model simulated airborne HSL devices for extracting hyperspectral three-dimensional point clouds of forests. Vegetation indices and a random forest model are utilized to invert chlorophyll and carotenoid concentrations in forest canopy layers. The results show that in indoor experiments,the HSL echo information effectively discriminates the spectral differences at different heights of plant structures. The NDVI values in the upper red leaf area and lower green leaf area of torch flowers are respectively less than and greater than 0.5. The LESS model successfully constructed high-resolution hyperspectral three-dimensional point clouds of forest scenes. Out of 24 groups of vegetation indices,17 groups exhibit good detection accuracy (MAPE<13%). The chlorophyll concentration inversion model demonstrates an R2 of 0.93 with MAE values of 6.26,3.40,and 2.81 for the upper,middle,and lower layers,respectively. The carotenoid concentration inversion model shows an R2 of 0.91 with MAE values of 1.59,2.58,and 0.39 for the upper,middle,and lower layers,respectively. This study indicates that HSL is an effective device for extracting vegetation spectral information in three dimensions and possesses immense potential for investigating the vertical distribution of biochemical components in complex vegetation scenes such as forests when deployed on aerial platforms.

When using UV-induced fluorescence technology to detect petroleum hydrocarbon pollutants in soil,soil with different particle sizes will cause large errors in the measurement. To eliminate the influence of Soil particle size (PS) on the measurement,Three common soil petroleum hydrocarbon pollutants (soil crude oil,soil diesel oil,and soil gasoline) with different particle sizes of soil samples were prepared. Through the establishment of a UV-induced fluorescence detection system,the fluorescence characteristics of various soil petroleum hydrocarbon pollutants under different particle sizes were studied,and the correction method of soil petroleum hydrocarbon particle sizes was established. The results showed that for soil crude oil,soil diesel oil,and soil gasoline samples when the sample concentration was 4, 4 and 10 g·kg-1,the fluorescence signal of each type of soil oil sample had a good linear correlation with particle size. The correlation coefficients R2 reached 0.998 9,0.968 6,and 0.904 5,respectively. The experimental results were interpreted and analyzed through the adsorption model of soil particulate matter. The soil particle size correction method was established to correct the fluorescence signals of petroleum hydrocarbon samples of different soil types under different particle sizes. Before and after particle size correction,the correlation coefficients R2 of fluorescence intensity and concentration of petroleum hydrocarbons in three types of soil increased from 0.326 5,0.004 7,and 0.329 8 before correction to 0.983 8,0.983 2,and 0.953 3 after correction,respectively. RE) were 11.02%,5.71%,and 10.19%,respectively. The proposed correction method can effectively reduce the influence of soil particle size on the fluorescence intensity of petroleum hydrocarbon pollutants in soil and provides the theoretical basis and technical support for the rapid,in-situ,and accurate detection of petroleum hydrocarbon pollutants in soil by UV-induced fluorescence technology.