With the end of the “2020 SO Identity Mystery”, the technique of space objects’ characterization and identification by spectra stands out again in Space Domain Awareness. The outstanding advantage of this technique is the ability to identify the material of space objects by the spectra reflected from their surface and then confirm the identity of space objects. We can still recognize the material by this method even when the image does not have a spatial resolution. In other words, The feasibility of characterizing space objects’ materials with a low-cost, small-aperture telescope has been demonstrated, which is unrealizable with traditional observation methods. Ph. D. Jorgensen’s thesis attracted much attention in this field in 2000, which started a research boom about the spectral characterization of space objects. However, the application of spectral characterization and identification techniques in space objects is still severely limited after development of over 20 years, which is inextricably linked to the approach of space objects’ characterization by spectra, as well as the complexity and unpredictability of the space environment. Researchers always characterize and identify the actual in-orbit objects based on measurements from the ground laboratory. However, there are some indescribable differences between them owing to the effects of the space environment. Spectral unmixing is a popular approach for determining the substance of space objects, whose principles and applications are thoroughly explained. This paper analyzes that the main factor of unsuccessful unmixing is the discrepancy between laboratory measurements and actual measurements, and the accuracy of unmixing largely depends on the perfection of the spectral library. So, the impact of the space environment and observation geometry on the spectral attributes of space objects must be considered before constructing the spectral library. Moreover, the use of Artificial Intelligence algorithms, on the other hand, can substantially improve the ability of space objects’ characterization and identification by spectra. This paper presents a detailed review and discussion of four aspects: space objects’ spectral attributes and classification, space object materials’ characterization and identification, the reddening of space objects’ spectra and the development of spectral databases for space objects. For relevant researchers’ convenience, we also analyze the difficult and key issues in this process and condense some constructive suggestions worthy of reference.

Surface-enhanced Raman Scattering (SERS) technology has the advantages of high sensitivity, high resolution, non-destructive detection and no pre-processing. It has become a powerful tool for qualitative and quantitative molecular detection. The trace detection technology, which amplifies the Raman signal of the target analyte, can even provide rich structural information on the molecular level. Although the SERS enhancement mechanism has always been disputed, currently acknowledged enhancement mechanisms include physical enhancement (electromagnetic field enhancement) and chemical enhancement (principally contribution of charge transfer). With the application of metal, non-metal and other materials in SERS in recent years, many scholars have inspired extensive interest in the factors affecting the Raman signal enhancement of SERS substrate. Hence, it is of great significance to research the mechanism of SERS enhancement. In this review, the SERS enhancement mechanism is primarily elaborated on three aspects of the SERS electromagnetic enhancement mechanism, chemical enhancement mechanism and synergic mechanism, to analyze which factors affect the substrate enhancement effect and provide some references for the analysis of the SERS enhancement mechanism. At the same time, the problems faced by different base structures in the process of enhancement mechanism analysis were proposed: (1)In the electromagnetic enhancement mechanism, the single noble metal base had a great influence on the SERS electromagnetic enhancement mechanism due to its uneven distribution of “hot spots” and uncontrollable factors, which led to poor SERS sensitivity and repeatability. (2) In the chemical enhancement mechanism, the single semiconductor material was widely used in the SERS substrate due to its advantages of affordable price, stable material performance and easy surface modification. However, the effect on the SERS chemical enhancement mechanism was not obvious due to its low enhancement ability. (3)The SERS substrate is not limited to single metal or non-metal materials but more the combination of metal and non-metal.

In this thesis, it took the non-destructive rapid testing of solid soluble content (SSC) in tomatoes as example. The near-infrared (NIR) spectra data of big and small tomatoes were collected by linear variable filter (LVF) NIR spectrometer and digital light processing (DLP) NIR spectrometer respectively. The average NIR spectra of big and small tomatoes and the difference spectra were calculated for LVF and DLP spectra respectively. The characteristics of the NIR spectra data of the two types of tomatoes collected by LVF and DLP spectrometer were compared respectively. Principal component analysis (PCA) was done on the LVF and DLP spectra respectively, and the distribution of the scores of the first 3 principal components were compared. The data were divided into calibration and external validation sets according to the SSC gradient. Partial least squares regression combined with a full cross-validation algorithm was applied to develop the quantitative calibration models of SSC in tomato for the spectra data collected by LVF and DLP spectrometer respectively. It is demonstrated by the result that: (1) The spectral characteristics of the average spectra and difference spectra of LVF-NIR spectra of big and small tomatoes are similar to those of DLP-NIR spectra, which indicates that it is feasible to carry out non-destructive and rapid testing of SSC in tomato by the LVF and DLP NIR spectrometers. (2) The separation trend of the score scatters of the first 3 principal components of LVF-NIR spectral data of big and small tomatoes was not obvious, while there is little separation trend for that of DLP-NIR spectral data. (3) The ratio performance deviation (RPD) values of the models developed by the LVF-NIR spectral data were no less than 2.11. Among them, the preprocessing of normalization acquired the optimized model, of which the number of factors (Nf), determination of calibration (R2C), root mean square error of calibration (RMSEC), determination of cross validation (R2CV), root mean square error of cross validation (RMSECV), RPD, correlation coefficient of prediction (rP) and root mean square error of prediction (RMSEP) were 8, 0.949 1, 0.27, 0.899 9, 0.38, 3.16, 0.882 6 and 0.63 respectively. The RPD values of the models developed by the DLP-NIR spectral data were no less than 1.60. Among them, the preprocessing of normalization acquired the optimized model, of which the Nf, R2C, RMSEC, R2CV, RMSECV, RPD, RP and RMSEP were 5, 0.823 5, 0.49, 0.728 6, 0.62, 1.94, 0.788 4 and 0.80 respectively. This thesis will, to some extent, provide reference to the non-destructive and rapid testing of SSC in tomatoes and the selection and evaluation of the non-destructive and rapid instrument for testing the quality of fruits and vegetables.

Fluorescent carbon quantum dots (CQDs) are new and promising photoluminescent nanomaterials. They have many advantages, such as diverse physicochemical properties, unique optical properties, low cost, eco-friendliness and abundant functional groups, which make them have broad application prospects in biological imaging, optoelectronic devices, photocatalysis, ion detection, targeted drug delivery and other fields. Light-emitting diode (LED) has been the focus of academic research for many years, widely used in liquid crystal displays, full-color displays and daily lighting equipment. Many kinds of fluorescent materials have been applied in LED research. Nano fluorescent CQDs have great application prospects in optoelectronic devices because of their advantages ofadjustable fluorescence emission wavelength, stable luminescence performance, environment friendliness, abundant raw materialsand low cost. However, the controllable preparation of CQDs is still a challenge. The fluorescence emission wavelengths of most CQDs are mainly concentrated in the blue and green wavelengths, and the quantum yield is low, which limits the development of CQDs in this field. Therefore, the synthesis of fluorescent CQDs covering thefull visible spectrum and the briefly analysis of their luminescence mechanism can greatly promote the application of CQDs in the field of white LED. In this paper, full-color fluorescent CQDs were successfully prepared by a one-step solvothermal method using triammonium citrate as the precursor and various low-toxic and inexpensive acid reagents as modifiers. The prepared CQDs were characterized by fluorescence spectrometer, transmission electron microscope, X-ray diffractometer, Raman spectrometer, X-ray photoelectron spectrometer, ultraviolet-visible spectrophotometer and Fourier transform infrared spectrometer. The results showed that the prepared CQDs had uniform size and good dispersion, the emitted fluorescence gradually changed from blue to red, the emissionpeak wavelength was adjustable between 450 nm and 650 nm, the fluorescence quantum yieldswere all above 30%, and the quantum yield of red CQDs was as high as 38.75%. The surface of the prepared CQDs was rich in carboxyl and hydroxyl groups, which gave them strong hydrophilicity. The luminescence mechanism of CQDs were explored by controlling the degree of graphitization and the number of carboxyl groups on the surface by different acid reagents. Full-color emission CQDs/epoxy composite films were prepared by adding one or more colors of CQDs to the epoxy resin, and three kinds of white LEDs with high color rendering index were successfully prepared. The CIE color coordinates of the warm white LED were (0.43, 0.39), the correlated color temperature was 3 913 K, and the color rendering index was 86. The CIE color coordinates of the prepared neutral white LED were (0.37, 0.37), the correlated color temperature was 4 170 K, and the color rendering index was 85.5. The CIE color coordinates of the prepared cold white LED were (0.30, 0.34), the correlated color temperature was 6 857 K, and the color rendering index was 80.4. This study provides a new idea for developing alternative phosphors for low-cost full-color fluorescent films and light-emitting devices.

Two-dimensional (2D) temperature and component concentration distribution based on tunable diode laser absorption spectroscopy is significant for combustion diagnosis, and the iterative algorithm plays an important role in the reconstruction of temperature and component concentration. It is found that the adaptive algebraic reconstruction technique and the least-square QR decomposition algorithm have good advantages in constructing 2D temperature and H2O concentration distribution. The simulated results show that four spectral absorption lines of H2O with wavelengths at 7 154.35, 7 153.75, 7 185.60 and 7 444.36 cm-1 are very suitable for measuring temperature and water vapor concentration distribution in high-temperature premix flame. Compared with the absorption lines at 7 444.36, 7 185.60, 7 154.35 and 7 153.75 cm-1, the absorption lines of CO2 and CH4 are very weak, the absorption lines of O2 and CO are almost no absorption in this band. Therefore, CO2, CH4, O2, CO and other gases in the combustion environment do not affect the absorption spectrum of H2O. By comparing the optimal relaxation factor, calculation time and reconstruction error of different algorithms, we find that the adaptive algebraic reconstruction technique has better reconstruction quality and shorter calculation time than the least square QR decomposition algorithm. Based on the AART algorithm, this paper further compares the 2D reconstruction effects of different absorption line pairs (7 444.36 and 7 185.60 cm-1, 7 154.35 and 7 153.75 cm-1) at 16, 32, 48 and 64 ray beams. The results show that the reconstruction results obtained from absorption lines 7 153.75 and 7 154.35 cm-1 are better than those obtained from absorption lines 7 185.60 and 7 444.36 cm-1. With the increased laser beams, the reconstructed results are closer to the assumed temperature and concentration distributions. The 32-beam arrangement is more suitable for the actual flame’s 2D temperature and concentration reconstruction. In order to analyze the ability of the AART algorithm to reconstruct different temperature and concentration distributions, this paper further simulates bimodal temperature and concentration distributions. The results show that with the increase of the ray beam, the temperature reconstruction error is always greater than the concentration reconstruction error, indicating that the ray number has a more obvious effect on temperature. In the bimodal distribution, the reconstruction error is the largest when the projected ray beam is 16, but the reconstruction results can also reflect temperature and concentration distribution trend.

Oocyte quality is the key factor to affect the ability of in-vitro fertilization, the potential of embryo development in vitro and the success of pregnancy of mammals, there by significantly affecting the assisted reproduction of human beings, the breeding of animals, the improvement of varieties and the preservation of endangered species. The research on oocyte quality has important application value. Currently, the common quality assessment methods are mainly performed from the morphology and biochemistry analysis perspective. However, the traditional method of evaluating the development quality of oocytes based on morphological selection has the characteristics of strong subjectivity, and the reliability of the results strongly depends on technicians’ experience. Although the new method of detecting biochemical indexes based on biological technology can make up for the shortcomings of morphological detection methods, there are still some shortcomings of invasive research, complicated biological operation steps, long experimental time-consuming and affecting the subsequent development of oocytes. As we all know, Abnormal changes in biological tissue are often accompanied by changes in its internal biochemical components, which often shows spectral differences. Thus, this paper used ultraviolet-visible (UV-Vis) spectrophotometer and self-modified multispectral imaging system to study the spectra of oocytes, and the spectral data were analyzed between immature and mature oocytes, fresh mature and postovulatory aged oocytes. It was found that (1) compared with the UV-Vis spectra (190~1 100 nm) of fresh matured oocytes, the spectra of immature oocytes and aged oocytes showed several peak changes. The UV-Vis spectra of immature oocytes missed two peaks of 205 and 579 nm and the band of 894~941 nm and added the peak of 593nm compared with that of fresh matured oocytes. For the UV-Vis spectra of the aged oocytes, three peaks of 205, 445 and 579 nm and the band of 846 to 941 nm disappeared and the peak at 593 nm also increased. (2) The visible-band multispectral imaging data showed that compared with fresh matured oocytes, the transmission spectrum intensity at various biological structures of immature oocytes decreased at 451 and 467 nm, and the corresponding data of aged oocytes decreased at 425, 431, 571 and 669 nm. In summary (specially, the spectral differences between fresh matured and aged oocytes in multispectral imaging)， it is thus clear that spectral detection shows great feasibility in identifying oocyte quality.

As a broad-spectrum insecticide, Cypermethrin is widely used in various agricultural products, such as fruits, vegetables and poultry and so on. Because of its large dosage and slow degradation rate, drug residues in fruits, vegetables, livestock and other agricultural products are harmful to human health. In order to avoid human intake, it is very important to detect cypermethrin residues in agricultural products. Among the current detection methods, the vibration spectrum technology has the advantages of being non-destructive and fast. Therefore, this paper uses the density functional theory method combined with the vibration spectrum technology to provide a theoretical basis for the vibration spectrum detection and identification of Cypermethrin, and provide a reference for the application field of pesticide residue detection. The specific research contents and results are as follows: the first step is to construct the molecular space configuration of Cypermethrin by using Gaussian view software. Based on the DFT/B3LYP method of density functional theory, the structure is roughly optimized with a 3-21G basis set and then reoptimized with 6-311++G basis set based on coarse structure to obtain the stable configuration and frontier orbital distribution of the molecule. Then, based on the optimized structure, the theoretical infrared and Raman spectra of Cypermethrin were calculated. The theoretical results show that Cypermethrin has obvious infrared activity in the range of 3 300~3 000 and 1 700~500 cm-1. The former is mainly the vibration of functional groups, and the latter is the vibration of the fingerprint region. It can also be seen from the calculation results that the stretching vibration and scissor vibration of methylene hydrocarbon on cyclopropyl at 3 044 and 1 459 cm-1, the wagging vibration of methyne on cyclopropyl at 1 196 cm-1and the rocking vibration of hydrocarbon in benzene ring at 1 153 cm-1in Raman spectrum have no activity in the infrared spectrum. The cyano group without infrared activity shows a strong band in the Raman spectrum. The benzene ring skeleton vibration is weakly absorbed in the infrared spectrum but shows a strong band in the Raman spectrum. These reflect the complementary advantages of infrared spectroscopy and Raman spectroscopy. The combination of the two spectra is more conducive to the identification and detection of compound structure. In the second step, the natural Raman spectrum of Cypermethrin powder was measured by experimental method. The theoretical calculation error was corrected by the frequency correction factor of 0.973. The experimental results were compared with the theoretical calculation results. The difference in the peak frequency wavenumber was mostly in the range of 4~10 cm-1, and the theoretical data were consistent with the experimental results. This study provides a theoretical basis for the vibration spectrum detection and structure identification of Cypermethrin, and provides a theoretical reference for its application in pesticide detection.

The excellent dielectric properties and metal substitutability of engineering plastics make them popular for 5G construction. The detection and characterization of several engineering plastics with similar appearances but different properties can help engineering plastics be better used in manufacturing 5G circuit boards and antenna modules. The terahertz time-domain spectroscopy detection technique (THz-TDS) was applied to spectroscopically detect several common engineering plastics, PEEK, PPS, and ABS, and the terahertz time-domain spectra of three engineering plastics were obtained. The terahertz frequency domain spectra of engineering plastics at 0.1~1.2 THz were obtained by fast Fourier transform of the terahertz time-domain spectra of the three engineering plastics. The related optical parameters were extracted to obtain the terahertz absorption spectra of engineering plastics. Analysis of the THz time-domain spectra shows that the THz time-domain spectra of different kinds of engineering plastics have differences in time delay lines and amplitudes, which can visually demonstrate the differences between various classes of plastics, indicating that THz-TDS is feasible for the classification and identification of engineering plastics. However, since the same engineering plastics exhibit similar peak positions and peaks in the terahertz band and no obvious THz characteristic absorption peaks for each material, they cannot be directly determined by fingerprint spectra. Based on this, the feasibility of applying a nonlinear instrumental convolutional neural network (CNN) to the study of engineering plastics without obvious feature absorption peaks is explored. An improved CNN classification model is proposed by optimizing the network structure and important weight parameters of CNN. The model uses the LeakyRelu activation function, adds BN layers, and utilizes the Adams gradient descent algorithm, which enables to ensure the robustness of the classifier, accelerates the network classification speed and improves the accuracy of THz spectrum recognition while solving the problem of easily falling into local optimum due to the insufficient amount of THz spectrum data, and compare this method with the traditional linear tool principal component analysis-support vector machine method (PCA-SVM). Validation experiments are conducted to verify the advantages and disadvantages of the two qualitative analysis models. The experiments show that the improved CNN classification model takes 0.15 ms to run on average, with 99.6% accuracy in the training set and 98.8% accuracy in the test set; compared with the traditional PCA-SVM classification model, its classification is significantly improved, and the classification accuracy is increased by 27.3% in the test set and 30.9% in the training set. The results show that the combination of THz-TDS and the improved CNN classification model can achieve accurate identification and classification recognition of the above three engineering plastics, which provides a new method for non-contact rapid non-destructive detection and identification of engineering plastics and can also be used as a reference for the identification and detection of other substances without THz characteristic peaks.

The Laser-induced breakdown spectroscopy technique has been used for the wall diagnosis in EAST Tokamak.Improving the accuracy of LIBS analysis in a vacuumis one of the bottlenecks for further development and application. In a vacuum, laser-induced plasma evolution highly depends on time and space. Studying the spatial-temporal evolution of plasma and understanding the species’ behavior is necessary to improve the LIBS accuracy further.Considering different elements related to the first wall and diverter used in Tokamak, the sample of ternary alloy-tungsten carbide copper ((WC)70Cu30) was used in this work. Multi-component plasma was produced by nanosecond laser ablation in a vacuum with a wavelength of 1 064 nm, the pulse width of 5 ns, and power density of 6.3 GW·cm-2. The temporal and spatial resolution measurement has been achieved using a linear fibre bundle.Six lines of C Ⅰ 833.51 nm, C Ⅱ 657.81 nm, Cu Ⅰ 515.32 nm, Cu Ⅱ 512.45 nm, W Ⅰ 429.46 nm and W Ⅱ 434.81 nm were selected to analyze the emission time scales. Moreover, the element separation in space and ion acceleration phenomena were also investigated. According to the spectrally-resolved time evolution results, it was found that continuous radiation mainly occurs in the early time of 80 ns, ionic emissions are 30~300 ns, and atomic emissions are 100～1 000 ns. The spatial distribution of atoms and ions corresponding to C, Cu and W elements are all different, indicating the separated species occurs during multi-component plasma expansion. The peak position and time for the six lines have been linearly fitted to obtain the corresponding species velocity, which ranges from 4.2 to 34.9 km·s-1. The results also show that the smaller the relative atomic mass, the faster the corresponding expansion velocity. (C Ⅰ>Cu Ⅰ>W Ⅰ, C Ⅱ>Cu Ⅱ>W Ⅱ); the ion velocity is greater than its atomic velocity (C Ⅱ>C Ⅰ, Cu Ⅱ>Cu Ⅰ, W Ⅱ>W Ⅰ). The element separation in space and ion acceleration is attributed to the element mass separation effect and transient sheath acceleration, which also reveals the spatial heterogeneity property of laser-produced plasma. The results provide important information for the LIBS theoretical model and a new idea for improving the accuracy of vacuum LIBS quantitative analysis.

Sichuan Longxi nephrite is important in ancient Shu (ancient Sichuan) jade culture. In order to get the basic mineral, chemical and spectroscopic characteristics of Sichuan Longxi nephrite with different colors, exploring its significant implications for identifying the jade color and alteration color of nephrite unearthed in Sichuan, lithologic thin section analysis, laser Raman spectroscopy, X-ray fluorescence spectroscopy and μ-XRF scanning were conducted on Longxi nephrite samples with different colors. The systematic field geological survey and sample collection shows that the color bands of Longxi nephrite outcrop have clear boundaries of grey black-green-white-green-gray black. The Longxi nephrite mainly comprises tremolite, containing a small amount of muscovite, calcite, apatite and other accessory minerals. The light bands contain calcite veins, and the dark bands contain relatively more accessory minerals. The Longxi nephrite samples in different colors have almost the same Raman spectra of tremolite with Raman peaks at 174, 228, 376, 392, 674, 935, 1 027, 1 061 cm-1 and the single peak of hydroxyl at 3 675 cm-1 in the range of 3 600～3 700 cm-1. None, characteristic Raman peaks of actinolite or graphite, indicate that the dark color of Longxi nephriteis not from dark-color minerals. The XRF analysis reveals that the main chemical compositions of Longxi nephrite samples are SiO2 (55.20%～57.94%), MgO (24.10%～25.00%), CaO (12.60%～13.80%), Al2O3 (0.39%～1.77%), and iron oxide (0.25%～0.42%). FeO is the main iron oxide tested by the Fe-VOL05 titration method, and there is no significant correlation between the Fe2+ content and the colors of Longxi nephrite. The μ-XRF scanning results indicate that Si and Mg are relatively high in black-deep gray Longxi nephrite, and the contents of V and Cr are relatively high in green-white Longxi nephrite. The contents of common coloring elements in jade,such as Fe, Mn and Cu, do not vary in different color bands of Longxi nephrite.It further proves that V and Cr lead to the green color in Longxi nephrite. The analysis of spectral characteristics and coloring mechanism based on nondestructive identification can be fully applied to the comparative raw material study of ancient jade arte facts, providing scientific data for provenance research on jade material and jade culture of ancient Shu culture.

In this paper, a new method for stable weighted mixture contraction of variables is proposed to address the problems of low prediction accuracy and poor interpretability of calibration models due to high spectral line dimensionality and many irrelevant variables when using attenuated total reflection-Fourier transform infrared (ATR-FTIR) spectrometers for measuring solution concentrations in crystallization processes. The method first proposes a stable weighted variable population analysis (SWVCPA) with a random binary sampling of the spectral variables and a weighted evaluation of the selected frequencies of the variables in the established superior sub-models and the stability indicators of the regression coefficients of the variables in all sub-models. By ranking the importance of variables and using an exponentially decreasing function to gradually force the filtering out of variables of low importance during the iterative process, an initial contraction of the spectral variable space is achieved, and the stability of the contraction is substantially improved. Then a new Dynamic Sparrow Search Algorithm (DSSA) is continued on the shrunken subspace to optimize the combination of variables further using the minimization of the root mean square error of training prediction (RMSEC) as the fitness function. This hybrid optimization approach combines the advantages of both types of variable selection algorithms, ensuring the stability of the prior variable contraction through a sub-model competition approach, preventing the algorithm from falling into a local optimum, and avoiding the traversal search for the remaining variable combinations through an intelligent optimization algorithm, allowing more variables to be retained for accurate selection. ATR-FTIR spectral data collected at six different concentrations during the cooling and crystallization of L-glutamic acid solutions were tested. The results showed that the new method reduced the number of spectral variables from 613 to 46 and that the root mean square error of prediction (RMSEP) was reduced from 1.727 9 to 0.165 4, and the coefficient of determination of prediction (R2) improved from 0.973 7 to 0.999 7 for the partial least squares (PLSR) model built using the selected variables compared to the original spectra. Genetic algorithm (GA) and variable population combination analysis (VCPA) for selecting variables, the solution concentration prediction model developed using the new method has higher accuracy and stability, indicating the practical application of the method to improve the accuracy and reliability of measuring solution concentration in cooling crystallization processes using ATR-FTIR spectroscopy.

Sugar degree, one of the important indicators, is evaluating apples’ internal quality. When establishing a parsimonious model for analyzing apple sugar degree, the quality of calibrated samples and wavelengths affect the model’s accuracy, later update and maintenance.In this paper, 90 apples were taken as objects, a total of 1 044 wavelength points in the 350~1 150 nm spectra bands were collected. This paper studied the efficiency and feasibility of the Lasso implemented Least Angle Regression (LASSOLars) on sample and wavelength optimization.A combination of Norris derivative filtering, first-derivation and Variable Sorting for Normalization was used to preprocess. Considering the concentration ranking, split 75% of the sample dataset into the original train dataset (68 apples) and 25% into the test dataset (22 apples), and obtained the optimal train subset by LASSOLars. Compared LASSOLars with other two variables selection methods such as Monte Carlo Uninformative Variable Elimination and Competitive Adaptive Reweight Sampling respectively. Analyzing the model results, samples and wavelength sizes & distributions. The result shows that the optimal train subset compressed 16% of the original train dataset. At the same time, not changing the average level of the original train dataset, and the distribution was closer to the test dataset, the model quality was not weakened after reducing calibrated samples.The RMSECV of the optimal train subset and original train dataset were 0.460 and 0.491, the R2CV were 0.913 and 0.916, the RMSEP were 0.462 and 0.471, R2P were 0.909 and 0.906. LASSOLars selected out 40 wavelength points, the least size with the best results and highest signal-to-noise ratio, RMSECV, R2CV, RMSEP, R2P and RPD were 0.933, 0.400, 0.944, 0.373, 2.838. Based on the samples and wavelengths optimization by LASSOLars, which expanded the application of LASSOLars in subset selection, and provides ideas for optimizing, updating and maintaining the model.

In using polymer products, people are most concerned about their failure conditions, and the important manifestation of failure is the yield of the material. So far, the dislocation theory has been widely used to explain the yielding phenomenon of polymer materials. This theory usually focuses on the crystal orientation and destruction while ignoring the crystal deformation and the stress acting on the crystal. The orientation and destruction of the crystal are only the results of the yielding, and the ability of the crystal to withstand stress is the direct cause of yielding. Therefore, this paper will start from the stress and inhomogeneous deformation of crystals to study the yielding behavior of polymer products, hoping to provide new ideas for understanding the failure behavior of polymer materials. Here, the widely used isotactic polypropylene (iPP) material is selected as the research object, and iPP samples with different lamellar thicknesses are prepared by isothermal crystallization of iPP melt at different temperatures. Two-dimensional wide-angle X-ray diffraction spectroscopy was used to monitor in situ the crystal destruction and crystal orientation processes of the iPP samples during stretching. The “covering method” was used to process the two-dimensional X-ray diffraction patterns for the first time, the change of the 2θ angle of the (110) crystal planes during the stretching process was observed in situ, and the deformation of the crystal in two directions (parallel to the stretching direction and perpendicular to the stretching direction) was distinguished. The results show that for iPP crystals with different lamellar thicknesses, the inhomogeneous deformation of the crystals during the uniaxial stretching process is a common phenomenon; the destruction and orientation of the crystals always occur at the same time, starting from the yield point, which is independent of the lamella thickness; the critical stress corresponding to the crystal destruction is related to the thickness of the crystal. The thicker the lamellae and the more stable the crystal, the greater the critical stress required. The above results show that in situ X-ray diffraction spectroscopy can observe the crystal structure changes during the stretching process in real time, thereby directly correlating the crystal structure evolution with the macroscopic mechanical properties.

The reduction of CO2 is still a challenge in energy sources and the environmental field. Diamine coordinated manganese tricarbonylcatalysts containing diamine ligands, as inexpensive molecule based inorganic materials, become a potentially interesting catalyst in the photoreduction of CO2. The ultraviolet-visible (UV-Vis) and infrared (IR) spectra are beneficial to the modulation of catalysts for the photoreduction of CO2. TheUV-Vis and IR spectra of a series of diamine coordinated manganese tricarbonyl catalysts, ［Mn(bpy)(CO)3Br］, (1), ［Mn(phen)(CO)3Br］(2), ［Mn(phen-dione)(CO)3Br］(3), ［Mn(phen-dione)(CO)3CH3CN］+(4) (bpy=2,2′-bipyridine, phen=1,10-phenan-throline, phen-dione=phenanthroline-5,6-dione) has been investigated using Density Functional Theory (DFT) and Time-Dependent Density Functional Theory (TD-DFT) calculations. The UV-Vis spectrawere obtained using several TD-DFT calculations. The simulated UV-Vis spectra of 1 and 2 all show two peaks centered at 371 nm (1), 408 nm (1) and 361 nm (2), 414 nm (2), respectively. Absorption peaks of 1 and 2 arise from a metal-to-ligand charge transfer (MLCT) transition. The simulated UV-Vis spectra of 3 and 4 show three absorption peaks centered around 290 nm (3), 337 nm (3), 431 nm (3) and 294 nm (4), 319 nm (4), 371 nm (4), respectively. Herein, except for the peak of 294 nm (4) generated by the ligand-to-ligand π—π* transition, the remain absorption peaks of 3 and 4 all arise from MLCT absorption. The increased electronegativity of diamine ligands is responsible for shifting absorption peak from the UV region to the visible region (red). The increased electronegativity of the Mn-centeredligand is responsible for shifting absorption peak from the visible region to the UV region (blue). Once an electron is transferred from the Mn-centered unit to the diamine ligands, the Mn center will become anelectron-deficient unit. Herein, the Mn-centeredunit contributes from the σ antibonding orbital between the Mn atom and ligand in the MLCT states. Thus, when the populated excited MLCT states are hit, the release of Mn-centered ligand (Br-/CH3CN) becomes favorable, and the system can form the active species. The simulated IR spectra show that two kinds of characteristic peaks are obtained, that is, (1) stretching vibration of Mn-centered CO unit at 1 920～2020 cm-1 (1, 2, 3, and 4) and (2) stretching vibration of CO unit of diamine ligands at 1 690 cm-1 (3) and 1 694 cm-1 (4), respectively. The increased electronegativity of diamine ligands and Mn-centered ligand strengthens the CO bond, which increases the stretching vibration frequency of the CO units from 1 to 4. The calculation results, including molecular structures, the UV and IR spectra show good agreement with those obtained by experiment and provide the reasonable theoretical guide for the synthesis and regulation of diamine-coordinated manganese tricarbonyl catalysts.

In order to quickly and accurately detect the phosphorus content information of rice leaves in the cold northern region, analyze the growth of rice, and provide the basis for precision fertilization and scientific management of rice fields, this study takes rice in the cold northern region as the research object, based on the plot experiment, and uses the marine optical HR 2000+ spectrometer to obtain the hyperspectral reflectance data of rice leaves, the content of phosphorus in rice leaves was determined by vanadium molybdenum yellow colorimetry.SG smoothing and multiple scattering corrections (MSC) were used to preprocess rice leaf hyperspectral data. The spectral data were selected by using SPA and UVE. Eleven features were screened by the SPA algorithm, including 6 in the visible band (411, 420, 428, 442, 467 and 689 nm) and 5 in near-infrared band (797, 850, 866, 965 and 976 nm). A total of 47 features were obtained by the UVE algorithm, which was all located in the visible band range and distributed between 405 and 603 nm. Taking the characteristic reflectance selected by the two methods as input, three inversion models of phosphorus content in rice leaves were constructed and analyzed, including extreme learning machine (ELM), BP neural network and wolf pack algorithm (WPA-BP).The results show that the verification set R2 of the three models constructed with the characteristic reflectivity filtered by the UVE algorithm as the input is between 0.705 2~0.724 5 and RMSE is between 0.017 4~0.020 4; Under the condition of the same inversion model, the prediction effect of the model constructed by using the characteristic reflectivity filtered by SPA algorithm as the input is good. The verification set R2 of the three models is between 0.726 4~0.829 3, and RMSE is between 0.018 0~0.021 1; In addition, when using the features screened by the two algorithms for modeling, comparing the prediction results of the three models, it is found that the accuracy of the BP neural network model optimized by the wolf swarm algorithm is significantly higher than that of the ELM and BP neural network, and the determination coefficient R2 of the verification set is 0.803 4 and RMSE is 0.018 0. Because of this, the combination of SPA and WPA-BP has certain advantages in a hyperspectral inversion of phosphorus content in rice leaves in the cold northern region, which can be used as a reference for the detection of phosphorus content in rice leaves and accurate quantitative fertilization.

Anthracnose of Camellia oleifera is a highly destructive disease commonly occurring in the Camellia oleifera industry, seriously restricting the development of the Camellia oleifera industry. In theearly stage of Camellia oleifera anthrax. It only need to repair the sick part of the tree in time. As the disease worsens, the affected branches must be eradicated, and seriously sick strains should be cut down in time. Aiming at, the current problem that the detection of Camellia oleifera anthrax is complex and the judgment accuracy is low, this paper proposes a method to determine the detection of the degree of Camellia oleifera anthracnose using laser-induced breakdown spectroscopy (LIBS) and Fourier transform near infrared spectroscopy (NIR), to achieve rapid, efficient and high-precision determination of the degree of anthracnose of Camellia oleifera. Fe, Ca, Mn, CaⅡ, and other elements in the LIBS spectrum of healthy Camellia oleifera leaves and diseased Camellia oleifera leaves were significantly different， and the characteristic peak intensities of the elements increased with the degree of disease. The main reason is that these elements are all necessary elements for the growth of Camellia oleifera. The absorbance of the Fourier transform near-infrared spectra of healthy Camellia oleifera leaves and Camellia oleifera leaves with different degrees of anthracnose also differs, mainly due to the ability of Fourier transform NIR to extract the physical properties of the sample. Using normalization, multivariate scatter correction (MSC), standard normal variate (SNV) preprocessing method combined with competitive adaptive reweighted sampling (CARS), Successive projection algorithm (SPA) variable screening method to establish fusion spectral classification model of anthracnose grades of Camellia oleifera by partial least squares discrimination analysis (PLS-DA) and support vector machine (SVM). Among them, the Root mean square error of prediction (RMSEP) and the prediction determination coefficient R2p of LIBS (Normalization-CARS)-NIR (Normalization-CARS)-PLS-DA of prediction set are 0.173 and 0.987 respectively and the misjudgment rate is 0. In the SVM model, the accuracy of the modeling set of LIBS-NIR-CARS-SVM is 100%, and the accuracy of the prediction set is 97.59%. The experimental results show that: the PLS-DA model based on the fusion of LIBS and Fourier transform NIR spectra for detecting anthracnose grades of Camellia oleifera leaves higher qualitative analysis accuracy and more stability than the SVM model. The results showed that: the LIBS spectrum combined with Fourier transform NIR spectrum could be used to separate healthy Camellia oleifera leaves from various grades of anthracnose of Camellia oleifera leaves efficiently, quickly and accurately.

Amber is a kind of organic gemstone with high value, and blood amber is a new kind of amber appearing in recent years, which is generally believed to be artificially irradiated. Four pieces of natural amber and one piece of copal resin were selected to be irradiated by an electron beam combined with a 60Co radiation source to change their color. It indicates that the irradiated samples showed the appearance characteristics of blood amber, and the color is bright red. Magnification observation was used by a 3D microscope system. UV-Vis, FTIR, and electron paramagnetic resonance were used to systematically study the samples before and after color modification. It can draw the following conclusions. After irradiation, the dendritic lightning striations were formed in amber and copal resins, which is the lightning streak discharge channel produced by insulator material under electron beam bombardment. Because there are pits, cracks and other positions in amber and copal resins, it often becomes the trigger point of breakdown and discharge. Therefore, the discharge trace will be left when the solid medium is broken down. There is no difference in the dendritic lightning striations between amber and copal resins after irradiation, and they are distributed in a single branch.IR and UV-Vis spectra have no significance in identifying whether amber and copal resin have been irradiated. It is no difference in the characteristic absorption peaks of the samples before and after color modification, indicating that the molecular skeleton was not seriously damaged. Electron paramagnetic resonance (EPR) spectra reveal that the irradiated amber andcopal resins will lead to the splitting of covalent bonds of some hydrocarbon groups in their structure and the formation of atomic groups with unpaired electrons, which is the free radical. The longer the irradiation time, the larger the formant area, indicating the higher the spin concentration of free radicals and the more obvious the color deepened. The formant area of all samples before irradiation is almost zero, indicating the spin concentration of free radicals in natural amber and copal resin is close to zero. It can be inferred thatthe generation of free radicals is the fundamental cause of color change. Electron paramagnetic resonance testing can determine the relationship between the spin concentration of free radicals produced by irradiation and color by scraping the small amount of power. Combined with the generation of the dendritic lightning striations, it can be used as the distinguishing characteristic of blood amber produced by irradiation.

“Promoting the national water conservation action” and “encouraging the reutilization of reclaimed water” is officially described in “The 14th Five-Year Plan for National Economic and Social Development of China (2021——2025)” and “the Long-Range Objectives Through the Year 2035”. The reuse of reclaimed water in the water-deficient areas is important for the sustainable development of water resources. Reclaimed water-irrigated agriculture can achieve comprehensive economic, ecology and society benefits, and conforms to the national policy requirements of Carbon Peak and Neutrality (namely “Double Carbon”). The fundamental investigation on reclaimed water-irrigated agriculture is generally beginning, and the related achievements can seldom be found. Two kinds of reclaimed water samples from wastewater treatment plants in Xi’an City, the capital of Shaanxi Province in northwestern China and Shenyang City, the capital of Liaoning Province in northeastern China, were recorded as “water sample A” and “water sample B”, respectively, were used in the research, and methods of ultraviolet-visible spectroscopy (UV-Vis), elemental analysis, three-dimensional excitation-emission matrix fluorescence spectroscopy (3D-EEMs) and Fourier transform infrared spectroscopy (FTIR) was applied to analyze DOM in water samples. The characteristic constants were further discussed to study the detailed information of DOM. The results show taht the absorption regions of 250~270 nm appear in UV-Vis spectra, indicating double-bond structure and aromatic components, and the absorption peaks at 210~250 nm are caused by unsaturated double-bond conjugate structure in DOM. DOM in “water sample A” shows a higher humification degree and a greater relative molecular mass, while it suggests a low polymerization degree of C-skeleton in the benzene ring and a more obvious endogenous feature in DOM from “water sample B”. The nature of DOM can be significantly affected by micro-organisms activities. The DOM fluorescence peaks appear at three locations, which are Ex/Em=280/310 nm (Tyrosine-like components), Ex/Em=230/320 nm (Tryptophan-like components) and Ex/Em=250/460 nm (Humus-like components), respectively. DOM in two water samples are autochthonous and of low-aromaticity, and DOM in “water sample B” is relative fresh and new generated. The FTIR spectra of the two DOM samples are similar, and the representative peaks appear at 3 460～3 420 cm-1 (stretching vibration of —OH and —NH2), 2 925 cm-1 (asymmetric stretching vibration of —CH2), 1 639 cm-1(stretching vibration of CO and bending vibration of N—H) and 1 412 cm-1 (shear deformation vibration of —CH2). Functional groups of DOM in two water samples very little, and the main elements are the protein-like components, carbohydrates and aromatic organic compounds. The achievements are significant to pre-evaluate the environmental behavior and ecological effect of DOM in reclaimed water used for agricultural irrigation.

Honeysuckle is an essential medicine for clearing away heat and detoxifying. However, the sources of honeysuckle on the market are complicated, and the most famous honeysuckle produced in Pingyi, Shandong, is often counterfeited. Most existing identification methods are time-consuming, costly, and complex to operate. Therefore, a fast and efficient way to trace the honeysuckle’s origin is urgently needed. The current one-dimensional convolutional neural network (1D-CNN) identification model based on honeysuckle near-infrared spectroscopy (NIRS) data has the problems of too many parameters and too low model efficiency, high computational complexity and is prone to overfitting. This paper improves the traditional 1D-CNN structure. We use the more efficient VD (Very Deep) structure to replace the hidden layer structure in the conventional 1D-CNN and make adaptive improvements for NIRS data so that the model can be directly applied to one-dimensional NIRS data. The improvement method is divided into three steps: firstly, the design of the feature layer is converted into two constraints optimization design: the first constraint is to set the C value of each convolution layer (the ratio of the size of the convolution kernel and the receptive field) to 1/6, which can improve the efficiency of the network model; the second constraint is to take the size of the top-level sensory domain as the size of the data vector, which can achieve feature extraction of deeper data and reduce overfitting. Secondly, this design minimizes the output feature vector of the feature layer to a smaller size through the downsampling operation. Finally, use two convolutional layers of size 1×5 and a pooling layer with dropout to downsample the data size to a vector of only one vector instead of a fully connected layer for classification, thereby reducing the number of parameters. In the experiment, 500 honeysuckles samples were collected from the main producing areas of Henan, Shandong, Hebei, and Chongqing. The spectral range used in the test is 908～1 676 nm. The sample set was preprocessed by the KS algorithm, and the training set, validation set and test set were divided by the shuffle algorithm. At last, a honeysuckle origin identification model based on improved 1D-VD-CNN and near-infrared spectroscopy was constructed. The results show that the 1D-VD-CNN training set and test set’s accuracy reach 100%, and the loss value converges around 0.001. Compared with the traditional 1D-CNN model, the training set and test set accuracy of the 1D-VD-CNN model are improved by about 0.5% and 1.4%, respectively, and the number of parameters and FLOPs are reduced by nearly 1 M and 20 M, respectively. At the same time, compared with the original spectral data analysis method and the PLS-DA method, it shows that the 1D-VD-CNN model has higher efficiency and better recognition performance for honeysuckle near-infrared spectral classification.

Carbendazim (Methyl-1H-2-benzimidazole carbamate),a kind of broad-spectrum fungicide, is extensively used in the prevention and treatment of ring rot and brown spot in apple planting. However, excessive use and residue of Carbendazim induce toxic effects on humans.In order to rapidly and accurately detect Carbendazim in apples, the Surface-Enhance Raman Spectroscopy combined immunoassay (SERSIA) was used in this study. This method was based on SERS’s highly sensitive molecular “fingerprint” characteristics, combined with immunospecific selectivity. The core-molecule-shell “sandwich” structure of Au@M@Ag nano SERS material and SERS immune probe binding antigen were prepared. Combined with Fe3O4 magnetic nanomaterial coated antibody,the specific detection of Carbendazim was achieved with the separable function.Transmission electron microscopy (TEM), UV-Vis spectroscopy and Raman spectroscopy were used to characterize the material properties and optimize the experimental parameters.The results showed a good linear relationship between 0.5～300 nmol·L-1 carbendazim concentration and the intensity of Characteristic peak at 2 227 cm-1 of the labeled molecule 4-mercaptobenzonitrile. The average recovery of Carbendazim in apples with different spiked concentrations was 95.6%～98.3%, and the RSD value was 0.15%～0.99%. Itis a sensitive, selective and stable method for detecting Carbendazim in apples without complex pretreatment.

Phytoremediation is a green and attractive technique for remediating heavy metal pollution. Understanding the distribution of heavy metals in different plant parts can provide insight into the molecular mechanisms of heavy metal phytoremediation. Laser-induced breakdown spectroscopy (LIBS) has outstanding technical advantages for the rapid in situ analysis of elements, especially for the direct analysis of solid samples without complex sample pre-treatment. Elemental scanning imaging is currently an important research and application direction of LIBS technology. A nanosecond pulsed laser-based elemental imaging LIBS device was developed with a spot resolution of 50 μm, a sample movement step of 100 μm and an imaging analysis speed of 6.25 mm2·min-1. The spectra of the target elements were normalized by baseline deduction, peak area fitting and normalization to produce a distribution thermogram with a pseudo-color representation of the elemental distribution in different regions of the sample. The in situ elemental imaging of pea plants was carried out using a hydroponic model, and the in situ elemental imaging of Ni, Cu, Cr and Pb were carried out using the imaging LIBS device. The results show that the device can effectively analyze the major matrix elements, such as C, Mg, Fe, Ca, Na and K, present in the plant and that after heavy metal stress, there is a significant accumulation of heavy metals in pea plants and different distribution trends of different heavy metals in the plant. Different from the distribution of nickel ions, the plant absorbed many copper ions and enriched it in the primordial root structures.The chromium accumulatedin the middle of the pea roots andin the germ and embryonic axis. However, large amounts of heavy metal lead was enrichedin the germ and embryonic axis, with the least amount in the root tip. This study shows that the elemental imaging LIBS technique enables the simultaneous in vivo analysis of multiple heavy metals in plants, which has implications for aiding the study of the mechanisms of phytoremediationin environmental waters. The imaging LIBS is a potential deviceand new method for related research in plant physiology and ecotoxicology.

Methanol gasoline because of its high octane number, low cost advantage to become the new fossil fuel alternatives, the methanol content of accurate detection is an important link in determine its quality, the quantitative analysis of methanol gasoline components is of great practical significance for alleviating the shortage of traditional petroleum resources but increasing demand in China. The conventional methods of methanol detection in methanol gasoline, such as alcohol analyzer determination, quick test box determination, etc., are complicated in operation and low in accuracy and quality , the conventional methods of methanol detection in methanol gasoline, such as alcohol analyzer determination, quick test box determination, etc., are complicated in operation and low in accuracy and quality. Near infrared analysis method is widely used in qualitative or quantitative analysis of components in many industries due to its detection speed and accuracy, Methanol gasoline near infrared spectrum are studied non-destructive detecting method, made up of 0.5%~30% methanol gasoline, nearinfrared spectrum acquisition system is designed and detect 60 components of methanol gasoline spectral data, Moving average smoothing method, S-G convolution smoothing and multiple scattering correction(MSC) were used to establish a prediction model after comparative analysis of spectral data, BP Artificial Neural Network(ANN) and Principal Component regression (PCR) were used to predict the determination coefficient and root mean square error of the mode, comparing the results and prediction effects of the two algorithms. The results show that the root mean square error of each model is less than 1%, and the fitting degree of SG smooth-principal component regression prediction model is the best, and the determination coefficient is 0.998 98, the model based on SG convolution smoothing algorithm and neural network algorithm has the smallest deviation between the predicted value and the true value, and its root mean square error (RMSEP) is 0.322 84%. This study shows that the performance of SG smooth-neural network prediction model in the application of near infrared spectroscopy detection and analysis technology to detect methanol content in methanol gasoline is good, and meets the application requirements, this study provides a theoretical basis for the practical detection and application of methanol gasoline components, and provides technical support for the effective development and utilization of methanol gasoline.

Firing temperature is a key technical parameter for the study of the development of ancient ceramic technology. Different firing temperatures might be used for ceramics from the varied chronological periods and functions. Thermal expansion analysis, can only be used to study high-fired ceramics and porcelains. It is challenging to determine the firing temperature of low-fired potteries from Neolithic and Bronze Age China. Current research developed an innovative method for studying ceramic firing temperature based on FTIR analysis. The structure of clay minerals in ceramics would be altered when exposed to high temperature, and a series of changes would happen to their FTIR bands at 3 625， 1 030， 640， 560 and 525 cm-1. These altered bands are potential indicators of ceramic firing temperature. Seven experimental ceramic samples fired between 450 ℃ and 1 050 ℃ were analyzed with FTIR, and the result shows a strong correlation between firing temperature and spectral features. In this light, the original firing temperature of ancient ceramics could be inferred based on their IR spectrum within an interval of 100℃. This method analysed two Shang period pottery samples and two casting mould samples. It is revealed that one pottery sample was fired between 550 ℃ and 650 ℃ while the other one was higher than 850 ℃. This result indicates highly varied firing techniques for different types of potteries. Both casting mould samples were fired below 550 ℃, demonstrating that the ceramics with different functions were consciously fired at different temperatures. Since FTIR is a fast analytical technique and requires only a small sample, it can be used to determine firing temperatures for large assemblages of archaeological ceramics without significant damage to the integrity of artefacts. It has great potential in investigating the development of ceramic firing techniques and revealing their technological and cultural significance.

Gansu Province is one of China’s most widely distributed areas of grottoes. Many of the grottoes are civilized in the world. Yunya Temple is located in Zhuanglang County, which was first developed in the Northern Wei Dynasty, developed in the Northern Zhou Dynasty and flourished in the Ming Dynasty. It is an extremely important cave in the Longdong area. The Ming Dynasty sculptures are unique in the grottoes because of the high value of cultural relics, beautiful shapes, and preservations and integrities. They are very important to the study of Buddhist art. In this paper, taking Cave 6 and Cave 7 of Yunya Temple Grottoes as the research objects, the integrated use of visible light photography, infrared photography, depth of field microscope ( OM ), scanning electron microscopy and energy spectrum ( SEM-EDS ), X-ray diffraction ( XRD ), portable X-ray fluorescence ( XRF ), laser Raman ( Raman ) and other technologies to produce materials and process analysis. The results show that the painted sculpture production process here is consistent with the traditional painted sculpture production process in Gansu, and the structures are the skeleton layer, clay plastic layer and color painting layer. Among them, the skeleton layers are divided into two kinds: wood skeleton and stone combined wood skeleton. Stone is a cliff body, mainly composed of quartz, albite, potassium feldspar, micro plagioclase, and goethite, a typical sandstone. The mud layer is divided into coarse and fine mud layers, and wheat straw and hemp are added as additive materials during production. The main components are quartz, albite and plagioclase. The painting layer is composed of a pigment layer and a gold layer. The red pigment is cinnabar and plumbum. The brown pigment is iron red, the green pigment is copper chloride, the white pigment is chalk, and the black pigment is carbon black. Drpiment is used as the orange pigment for some painted sculpture in Cave 7. Leaching powder paste is used in many places, such as collars, sleeves, patterns, etc. Compared with gold paste, mud gold, and appropriation, the leaching powder paste is more three-dimensional, which enhances the solemnity and aesthetic effect color plastic, and has a wide range of applications. Clothing patterns and base patterns were obtained using infrared photography ofoil smoke pollution, dust and other diseases. This work comprehensively analyzes the production process and materials of the colored sculptures in the Ming Dynasty of Yunya Temple, which is helpful for the systematic study of small and medium-sized grottoes in China. Also, it provides a scientific data base for the follow-up protection of grottoes and temples.

With the launch of high-resolution series satellites and the development of UAV hyperspectral, the available hyperspectral data are further expanded. Hyperspectral unmixing is a crucial task to improve hyperspectral images’ fine utilization value. With the rapid development of computer and artificial intelligence technology, deep learning theory has been introduced into the image processing field. The autoencoder network has been taken into hyperspectral unmixing because of its great feature extraction ability. This study improves the autoencoder structure and proposes a deep stack autoencoder network (DSAE) for hyperspectral image unmixing. The network consists of endmember extraction (EDSAE) and abundance estimation(ADSAE). Firstly, the EDSAE network is constructed by adding batch normalization, sparse constraint, “sum-to-one” constraint and deleting the bias term. Then unsupervised training is carried out for endmember extraction. Secondly, the obtained endmember spectral data are enhanced based on the HAPKE and LINEAR models. Finally, the supervised training network ADSAE is constructed based on the original stack autoencoder network, and the activation function of the last layer is set as the Softmax function. The simulated dataset is used as a training set, and the hyperspectral images are used as a test set. Based on the DSAE method proposed in this study, end member extraction and abundance estimation are carried out on three hyperspectral images, including Samson, Jasper Ridge and Urban. The results are compared with those obtained by the traditional methods N-FINDR, VCA, MVC-NMF and other deep learning methods SNSA and EndNet. The experimental results show that theD SAE method has obvious advantages over the other five methods in endmember extraction for the three real hyperspectral data set. It also shows the best abundance estimation results based on the synthetic datasets generated by the HAPKE mixing model. The DSAE method has good stability and robustness, which provides a new idea for the quantitative analysis and utilization of hyperspectral images.

Dissolved organic matter (DOM) is an active organic component in soil, and its source and composition can indicate the degree of soil humification and its relationship with the external environment. In this study, in order to scientifically monitor and evaluate wetland soil environmental quality, we collected the surface (0～20 cm) soil under different vegetation types of the Nianzishan Yalu River National Wetland Park in Heilongjiang Province, and applied three-dimensional fluorescence spectroscopy-parallel factor analysis method to measure the DOM fluorescence spectrum, and further analyzed the effects of soil physicochemical indexes on DOM composition. The results showed that the soil DOM humification index was between 2.562 and 9.052 under five vegetation types (including nine formations). The soil humification degrees of the deciduous broad-leaved forest and the deciduous broad-leaved shrub were higher than that of the meadow, followed by the marsh, and the soil humification degrees of the aquatic vegetation were the lowest. The fluorescence index was between 1.407 and 1.586. The soil DOM source had both exogenous and autogenic characteristics. The deciduous broad-leaved forest and the Phragmites australis (Cav.) Trin. ex Steud. marsh had obvious exogenous characteristics, but the aquatic vegetation and the Echinochloa crus-galli (L.) P. Beauv. Marsh had obvious autogenic characteristics. The biological index ranged from 0.482 to 0.662, and the contribution rate of recent autogenic characteristicwas low. Three types of five kinds of organic components were identified fromthe soil DOM. Humus-like substance (ultraviolet fulvic-like acid componentand visible fulvic-like acid component) had the largest relative proportion, and the soil samples with obvious exogenous characteristics had high content. Followed by protein-like substance (tyrosine-like componentand tryptophan-likecomponent), its content was higher in the soil sample with strong autogenic characteristics. Humic-like acid substance (humic acid component) was the lowest, and mostly existed in the xerophyte-mesophyte environment. Soil moisture content, pH value, and total organic carbon content had significant or significant effects on DOM composition. The three physicochemical indexes were positively correlated with the protein-like substance content and negatively correlated with the humus-like substance content. The correlation of moisture content, total organic carbon content with humic-like acid substancecontent were negative, respectively. Overall, in this wetland park, the soil samples of deciduous broad-leaved forest and shrubs are weakly acidic, with low moisture content and total organic carbon content, high degree of humification, obvious exogenous characteristic, high humus-like substance and humic-like acid substance content. However, under aquatic vegetation, the soil is nearly neutral, with high water content and total organic carbon content, a low degree of humification, obvious autogenic characteristic, and high protein-like substance content. The results of this study can provide basic data for the monitoring and evaluating soil environmental quality in permanent river wetlands in the cold northern region represented by this wetland park.

Plant nitrogen content (PNC) is an essential indicator of crop growth and nitrogen nutrition status. Therefore, accurate and efficient access to PNC information is vital for dynamically monitoring potato growth and proper N fertilizer application. In this study, the UAV hyperspectral images were obtained at the budding stage, tuber formation stage, tuber growth stage, starch accumulation stage, and maturity stage of the potato. After preprocessing, the original canopy spectrum and first-order differential spectrum of five growth stages were extracted; Secondly, the correlation analysis was carried out between the extracted canopy spectrum and potato PNC, and the sensitive wavelength of PNC was screened out; Then, the texture and color of two image features of the hyperspectral image at the wavelength of the original spectral features of the canopy were extracted using the gray co-generation matrix and the 1st to 3rd-order color moments, respectively, and the extracted features were correlated with the potato PNC to filter out the top five image features with higher correlation; Finally, based on spectral features, image features, and map fusion features, potato PNC estimation models were established by using elastic network regression (ENR), Bayesian linear regression (BLR), and limit learning machine (ELM). The results showed that: (1) there are differences in the characteristic wavelengths of canopy spectra in the five growth stages of potatoes. Still, most of them were located in the visible region. (2) The correlation between the texture and color characteristics of the original spectral characteristic wavelength image of the canopy and PNC was high. The correlation from the budding stage to the starch store stage was significantly higher than that in the mature stage. (3) The estimation models of potato PNC based on a single spectral feature and a single image feature have a good effect from the budding stage to the starch accumulation stage but a poor effect at the maturity stage. (4) From the budding stage to the starch accumulation stage, the estimation effect of potato PNC based on the map fusion feature was significantly better than the single spectral feature and the single image feature. (5) In each growth period of potato, the PNC estimation models constructed by ENR based on the same variable were better, BLR was the second, and ELM was poor. Among them, the accuracy and stability of the PNC estimation models constructed by ENR with fusion characteristics as model variables were the best. The modeling R2 of five growth periods were 0.91, 0.75, 0.82, 0.77 and 0.69 respectively; RMSE were 0.24%, 0.31%, 0.26%, 0.22% and 0.29% respectively, and NRMSE were 6.59%, 9.79%, 9.58%, 7.87% and 11.03% respectively. This study can provide a fast and efficient technical tool for monitoring the nitrogen nutrition of potatoes.

Plant nitrogen content (PNC) is essential for evaluating crop growth and nutritional status. Obtaining crop PNC information quickly and accurately can provide an important basis for formulating and implementing farmland management strategies. Existing studies have shown saturation in estimating crop PNC using only the spectral information of images. Therefore, this research attempted to use vegetation indices (VIs) combined with two-dimensional discrete wavelet decomposition technology (DWT) to extract high-frequency information(HFI) at multiple scales. It was constructing a spectral, spatial feature (VIs+HFI) and exploring the ability of VIs, HFI, and VIs+HFI to estimate PNC. First, the UAV was a remote sensing platform to obtain digital images of the five critical nitrogennutrient growth periods of potato budding, tuber formation, tuber growth, starch accumulation, and maturity. It measured PNC data for each growth period. Secondly, based on the pre-processed UAV images, the spectral information of the canopy of each growth period was extracted to construct VIs, and the DWT was used to extract the HFI of each growth period 1~5 scales. Then, the VIs and HFI extracted from each growth period were correlated with the ground-truthed PNC data. The top 7 VIs and the top 10 HFI with larger absolute correlation coefficient values were screened, respectively. To reduce the effect of covariance on the experimental results, the screened HFI were subjected to principal component analysis (PCA) for dimensionality reduction according to the KMO test results. Finally, two methods, ridge regression and extreme learning machine (ELM), were used to construct and evaluate the PNC estimation model of each growth period of potato with VIs, HFI principal components, and VIs+HFI principal components as model variables.The results showed that: (1) HFI at different scales contributed to the estimation of PNC in each growth period of potato. (2) The accuracy and stability of the potato PNC estimation model for each growth period constructed with VIs+HFI as model variables werehigher than that of a single VIs and HFI. (3) In each growth period of the potato, the PNC estimation model constructed by the ridge regression method was better than the ELM method. Among them, the PNC estimation model constructed with VIs+HFI as the model variable had the best effect. The modeling R2 of the five growth periods were 0.833, 0.764, 0.791, 0.664, 0.435, and the RMSE were 0.332%, 0.297%, 0.275%, 0.286%, 0.396%; NRMSE were 9.113%, 9.425%, 10.336%, 9.547%, 15.166%, respectively. This research can provide new technical support for real-time and efficient potato nitrogen nutrition status detection.

It is still difficult to identify black pear spots in the early stage of infection because the changes in the appearance of the infected area are very small and difficult to be observed by the naked eye. This study combined hyperspectral imaging technology and Stacking integrated learning algorithm to realize gley identification and detection of pear black spot. Firstly, a hyperspectral imaging system was used to collect the hyperspectral images of healthy pear samples and different disease grades. The region of interest (ROI) was selected based on the images, and the average spectrum was extracted. Then, First derivative (FD), Second derivative (SD), Standard Normal Variable Transformation (SNVT), SNV-FD and SNV-SD pretreatments were performed on the extracted original spectral data. Then, the Competitive Adaptive Weight Sampling (CARS) method was used to extract the spectral information of the characteristic wavelength. Finally, the Least Square support vector machines (LS-SVM), K-nearest neighbor method (KNN), Random Forest (RF) and Linear discriminant Analysis (LDA)classification models are established respectively based on the screened feature information. Among them, the combination of SNV-FD-LSSVM, SNV-KNN and SNV-FD-RF was better, with test set accuracy of 94%, 88% and 88% respectively. In the models established by LS-SVM, KNN, RF and LDA algorithms, the number of test set accuracy not less than 85.00% are 5, 3, 2 and 0 respectively. Therefore, three classifiers, LS-SVM, KNN and RF, are selected for subsequent ensemble learning. In order to improve the model accuracy, the optimized LS-SVM, KNN and RF models were used as the base classifier to construct the Stacking learning framework, and the modeling results of a single classifier were compared and analyzed. The results showed that the overall recognition accuracy of the integrated learning model is 98.68%, which is 4.64% higher than that of the single classifier model, and the recognition rate of gley samples is 11% higher. The results confirmed the feasibility of hyperspectral imaging combined with an integrated learning method to identify pear samples with a black spot in the gley stage. The integrated model significantly improved the accuracy of the single model. Moreover, it provides a new method for early detection and disease classification of black pear spots, and lays a foundation for further study on applying integrated learning algorithms in qualitative spectral analysis.

Rapid and non-destructive detection of crop disease types are essential to improve crop quality and yield. Traditional disease classification methods are time-consuming and difficult to detect in real-time. Therefore, the classification of soybean diseases was carried out by the hyperspectral technique. In this paper, healthy soybean was used as the control, frogeye leaf spot and bacterial blight diseases were the research objects, and hyperspectral data of three types of leaves were obtained. Changes inthe reflectance of diseased and healthy leaves were analyzed based on hyperspectral curves. Two single methods, principal component analysis (PCA) and spectral index (SI), were used to extract effective disease information. A total of 30 SI were used. A combination method of PCA and SI (PCA-SI) was proposed on this basis. Extracting the effective principal component (PC) and the effective SI, which were divided into two groups (9SIs and 18SIs) according to the score, and then grouped corresponding to each effective PC respectively to form the variable set of effective information of the disease spectrum. Three methods were used to extract effective disease information respectively. Based on the extracted spectral variables, the least square support vector machine (LSSVM) and support vector machine (SVM) was used to establish the disease classification model. With the original hyperspectral as the benchmark and the accuracy of disease classification as the index, the disease classification performance of the model, the effective information extraction methods of different diseases and the effectiveness of the classifier were evaluated. The results showed that the hyperspectral reflectance of diseased leaves was higher than that of healthy leaves in the visible band of 450~700 nm, while the characteristics of diseased leaves were opposite in the near-infrared band of 760~1 000 nm. A single PCA method was used to extract 34 effective PCS for disease classification. Based on the PCA-SI combination method, 5 effective PCs (PC1—PC5) and 18 effective SIs were extracted and grouped to obtain 10 groups of variables, and 13 groups of variables were used as modeling sets. The spectral variables extracted by the three methods have better disease classification ability than the original hyperspectral, and the proposed PCA-SI combination method has the optimal disease-effective information extraction ability. PC1-18SIs and PC4-18SIs were the best modeling sets, and the LSSVM classifier performed the best classification. PC1-18SIs-LSSVM and PC4-18SIs-LSSVM models were the optimal disease classification models. The total disease classification accuracy of the training and prediction sets was 100% and 98.85%, respectively. Compared with the original hyperspectral classification model, the overall classification ability of these two models was improved by 6.47% and 21.74%, respectively, and the model classification ability was good. It can provide a reference for real-time and non-destructive classification and identification of diseases.

Hyperspectral monitor on CarbonSat focuses on the detection of vegetation carbon sink and forest stock in the terrestrial ecosystem by detecting the spectrum of 670~780 nm, mapping the temporal and spatial distribution vegetation fluorescence to meet the needs of global carbon sink quantitative monitoring and forest vegetation productivity assessment. How to effectively calibrate the spectral parameters of HSM, establishing the relationship between the detector and the measured spectral information is the basis of quantitative inversion is the basis of quantitative radiance inversion. This paper gave the spectral data error model of HSM by grating equation, combined with the spot distribution function of the optical system, the Instrument Line Shape (ILS) of HSM is obtained. The simulation results show that the ILS changes slowly, and the ILS is approximately the same in a small spectral range. The wavelength error is mainly caused by grating manufacturing. It can be eliminated by the spectral line calibration method. In order to realize the spectral calibration of HSM on the ground, a calibration system that includes the tunable laser, wavelength meter, and rotating engineering diffuser is established in the vacuum tank. A monochromatic light with a linewidth less than 0.001 nm is provided, and the automatic data processing program is used to test the relationship between the response curve of the detector and the monochromatic light. The spectral sampling rate of HSM is about 2.5 pixels, and the effective data points of the single wavelength spectrum are limited. For getting accurate data of ILS, the spectral sampling rate is increased by two orders of magnitude by wavelength scanning at 0.015 nm wavelength interval, and the spectral resolution is obtained by Gaussian fitting. The results show that the spectral resolution of the HSM is 0.24~0.26 nm. The wavelength calibration data of all pixels is obtained by selecting the characteristic wavelength and cubic polynomial fitting. The characteristic wavelength is selected to verify the fitting residual. The results show that the calibration accuracy is better than 0.005 nm. In order to further verify the spectral calibration results, the ground push broom imaging experiment of HSM was carried out. The spectral data of pine forest and gravel pavement were obtained at the Huailai test station of the Aerospace Academy of the Chinese Academy of Sciences. The comparison results between the atmospheric absorption line measured by HSM and the atmospheric absorption line of HITRAN show the deviation of the central wavelength of the oxygen absorption line is less than 0.003 nm. The calibration result satisfied the system requirement HSM.

Milk powder is the main nutritional product for infants when they are not breast-fed. The frequent occurrence of melamine, vanillin and other additives has aroused great concern for society. How to more quickly detect and efficiently identify the adverse substances in milk powder is a research hotspot in the field of dairy product safety at home and abroad, which has very important practical significance to ensure the growth and health of infants. Due to a large number of absorption peaks in IR and Raman spectra of starch and vanillin, the overlap region of spectral signals between substrate and additive was wide, and the signal-to-noise ratio was low, which led to serious speak overlap. The more sensitive the instrument is, the more difficult it is to detect and screen. In this study, a non-destructive, rapid and highly sensitive method for determining vanillin as a food additive was established based on X-ray diffraction (XRD). The matrix materials, such as starch and protein of food dairy products, did not have complex diffraction patterns under XRD. Therefore,the detection method of vanillin based on XRD eliminated the background interference of various substrates in food and dairy products.The doping of vanillin could be quickly identified and qualitatively identified by combining the fingerprint characteristics of XRD patterns. The results showed a sharp diffraction peak at 13.03° with good peak shape and high response value in the diffraction pattern after doping vanillin, which would be a characteristic fingerprint peak for rapid detection of vanillin. Compared with gas chromatography and electrophoresis, this method could be more convenient, rapid, accurate in qualitative confirmation to obtain more reliable results and strong applicability while removing matrix influence by direct injection and fine scanning. In this study, the minimum doping amount of vanillin was 202 μg·g-1. The method can be used to rapidly detect vanillin in commercial flavoring foods, and the sensitivity can meet the requirements of recommended limit and excessive application.

Sulfur quantum dots (SQDs), as a new kind of quantum dots without metal elements, not only have the advantages of environmental protection and non-toxicity but also the advantages of simple preparation, low cost, good quality solubility, and stable photoluminescence (PL) characteristics. Sulfur quantum dots have aroused great interest from researchers in the quantum dot field and have good application prospects in nanoelectronics, optics, catalytic chemistry, biomedicine, and sensors. Currently, the research on sulfur quantum dots mainly focuses on synthesising sulfur quantum dots and improving photoluminescence properties. Like carbon dots, these quantum dots can also display different light colors under the irradiation of ultraviolet lamps, but the green fluorescence properties need to be further improved. At present, the research on sulfur quantum dots mainly focuses on the synthesis of sulfur quantum dots and the improvement of photoluminescence properties. Like carbon dots, these quantum dots can also display different light colors under the irradiation of ultraviolet lamps, but the green fluorescence properties need to be further improved. At present, sulfur quantum dots are mainly prepared by the ultrasonic-assisted liquid-phase reaction. In this paper, long-chain thiol molecules of 1-dodecanethiol are used as a sulfur source, and blue elemental sulfur quantum dots are successfully synthesized in a short time (2 h) by one-step heating at a high temperature (240 ℃). The synthesized quantum dots are characterized by fluorescence spectrum (PL), absorption spectrum (Abs), Raman spectrum, infrared absorption spectrum, elemental analysis, and morphology analysis. It can be seen from the experiment that the absorption of sulfur quantum dots began gradually from 550 nm, which was mainly due to many surface defects of quantum dots. There is an obvious absorption edge at 450 nm, corresponding to band absorption; The absorption at 372 nm is attributed to the absorption caused by S2-8 in quantum dots. Under the excitation of 330 nm, the synthesized quantum dots show obvious blue light, with the main emission peak at 450 nm and the main half-peak width of about 50 nm. Then, different quantum dots were synthesized by changing the reaction temperature and time, respectively. It was found that with the increase in reaction temperature and reaction time, the synthesized sulfur quantum dots (SQDs) showed a change from blue to yellow-green when excited at 330 nm, and the main peak wavelengths of the fluorescence spectrum (PL) were at 400, 450 and 525 nm, respectively. The luminescence quantum yield (PLQY) of the synthesized sulfur quantum dots (SQDs) could be 1.48%. In addition, we used synthesized sulfur quantum dots (SQDs) to prepare electroluminescent devices for the first time, with the structure of ITO/PEDOT:PSS/PVK/S-QDs/B4PyMPM/LiF/Al. Then we tested the electroluminescent characteristics of the devices and successfully obtained the blue light emission of the sulfur quantum dots (SQDs) at 472 nm. By changing the thickness of the electron transport layer B4, the luminance of S-QLED devices can be changed, which has a specific guiding role in realizing the electroluminescence of sulfur quantum dots (SQDs).

Stellar spectrum classification is one of the important tasks of stellar spectrum analysis. Chinese large-scale survey project LAMOST can obtain massive stellar spectral data. In order to efficiently classify massive stellar spectral data, especially stellar spectral subtype data, we need to study fast and effective stellar spectral automatic classification algorithms. This paper proposes a hybrid deep learning algorithm based on Transformer feature extraction, Bert+svm (abbreviated as Besvm), to classify the spectral subtypes of type A stars automatically. The algorithm takes 26 line indices of the spectrum of A-type stars as input features and uses the Bert model to perform a deeper learning of the 26 line indices. By learning the internal correlation of the 26 line indices, it extracts the spectrum more conducive to the A-type stars classification characteristics. The extracted new features are input into the classifier algorithm Support Vector Machine (SVM for short), and then the three subtypes A1, A2, and A3 of the A-type star spectrum are automatically classified. Previously, the SVM algorithm has been applied in the stellar spectrum classification task, and some derivative SVM algorithms also have a higher classification accuracy rate in the stellar spectrum classification task. Compared with the SVM algorithm previously applied to the stellar spectral classification task, our hybrid deep learning algorithm is less affected by the signal-to-noise ratio of the data, and the low-signal-to-noise ratio data can also have a higher classification accuracy. The amount of data used is relatively small. This paper verifies the effectiveness and superiority of the algorithm through five sets of experiments: Experiment 1 is used to compare and select excellent kernel functions, and through the matching experiment of spectral data, the radial basis kernel function RBF is finally selected; Experiment 2 compares the performance indicators of the Besvm algorithm with the other four traditional excellent algorithms verify the superiority of the Besvm algorithm; Experiment 3 is used to test the stability of the Besvm algorithm; Experiment 4 analyzes the influence of the amount of data on the Besvm algorithm; Experiment 5 analyzes the influence of different signal-to-noise ratios data on the classification accuracy of Besvm algorithm. The analysis of comprehensive experimental results shows that the hybrid deep learning algorithm Besvm proposed in this paper can still maintain a high classification accuracy rate on a small-scale data set with a low signal-to-noise ratio. The overall classification error rate of Besvm is below 0.01, which is much lower than the error rate of the classic traditional machine learning algorithm LDA algorithm, Bp neural network algorithm, SVM algorithm and Xgboost algorithm. The classification accuracy is too limited by the number of hidden neurons.

In the face of the increasingly abundant airborne and spaceborne hyperspectral sensors and the accompanying increase in hyperspectral data, the problems of excessive data volume and band redundancy have always been the major difficulties in hyperspectral image processing and interpretation. At the same time, the use of hyperspectral remote sensing technology to reveal camouflaged targets has always been the key point of modern remote sensing application technology research. In detecting massive spectral data of ground objects and redundant spectral information, the design of appropriate data dimensionality reduction technology plays a vital role. The band selection method among the main methods of dimensionality reduction processing can not only reduce the spectral information of the image data without distortion but also accurately distinguish the camouflaged target and its background based on it. Today, hyperspectral technology is an important technical means for military applications, and it is also a research hotspot for many scholars at home and abroad.It is a commonly used research method to use various indicators to calculate the different performances between the bands and to select the most representative bands according to their parameters for feature identification or classification to test the pros and cons of the method. However, few experimental studies still exist on specific band selection methods for special features, such as vegetation camouflage targets. In the study, green steel plates, green camouflage nets, and green fake turf were selected and placed in a background environment containing healthy green vegetation, wet bare ground, and dry, bare ground. Band selection and classification experiments were carried out by simulating camouflage targets and background objects in the real environment. First, analyze the spectral curve, and select a significant feature band. Secondly, the band screening is performed based on the sub-space divided according to the phase relationship between the band. The visual model is then established according to the image brightness of the property target. Finally obtains a band selection collection with relative in dependence and optimal recognition. Next, using support vector machine classifier and Mahalanobis distance classifier, the proposed algorithm is compared with two common algorithms in band selection results and full band combination for classification experiments. The experiment shows that the band selection results of the proposed method are better than those of the common algorithms in band selection results and full band, and the classification accuracy and speed are improved. Among them, compared to using full band classification, the overall classification accuracy of the two types of classifiers has been improved by 4.559 2% and 2.364 8%, respectively. The Kappa coefficient has been improved by 0.059 4 and 0.031 2, and the classification time has been reduced by 6.83 seconds. Experiments show that this method can effectively classify vegetation camouflage targets and background objects and has great practical application value.

Camellia oleifera, which has high nutritional value and is known as oriental “olive oil”, is an important economic forest in southern China, and China has the widest distribution of Camellia oleifera in the world. Extracting the distribution and planting area of Camellia oleifera is significant for forestry departments to carry out macro-management and production guidance of Camellia oleifera. Changning City, Hunan Province, located in a subtropical zone with complex object features and many mountains and hills, is the study area. Many farmland and forests are distributed in this subtropic area, and some vegetation varies greatly in different seasons, which brings great challenges to remote sensing extraction of Camellia oleifera. This paper uses GF-2 high-resolution satellite images in spring, summer and autumn. Combining vegetation index, texture features, PCA principal components, and four different time series combinations in spring and summer, spring and autumn, summer and autumn, and random forest algorithm, 17 classification scenes (S1—S17) were constructed. Three classification algorithms, random forest, support vector machine and maximum likelihood, were used to carry out remote sensing extraction experiments of Camellia oleiferato select the optimal feature combination, classification season, time series combination and optimal classification method. The results show that the classification accuracy based only on spectral information is low, and the addition of texture features can greatly improve the accuracy, while PCA has a weak effect on improving the accuracy; By comparing the classification results of single-period remote sense data in different seasons, it is found that the season with the highest extraction accuracy of Camellia oleifera is summer. With the summer image of the optimal feature combination (S8), the producer accuracy of Camellia oleifera is 94.06%, and the user accuracy of Camellia oleifera is 92.57%. In the classification scenes S10—S17, it is found that the accuracy of using time series information is improved compared with that of single-period images, and the classification accuracy of time series combination from high to low is: spring, summer and autumn, spring and summer, spring and autumn, summer and autumn. Random Forest, Support Vector Machine and Maximum Likelihood are used to extract Camellia oleifera by integrating spectral, texture and time series information, and the classification accuracy of random forest algorithm is the best in general. The therandom forest method (S17) using multi-temporal remote sensing vegetation index, texture and PCA in spring, summer and autumn is the scheme with the highest classification accuracy. The overall accuracy and Kappa coefficient are 96.85% and 0.961 0 respectively, and the producer accuracy of Camellia oleifera is 98.31%, and the user accuracy of Camellia oleifera is 94.33%. The random forest method (S10) using remote sensing vegetation index and texture in spring and summeris the best scheme with calculation efficiency and classification accuracy. The overall accuracy and Kappa coefficient are 95.62% and 0.945 8, respectively. The producer and user accuracy of Camellia oleifera are 96.93% and 95.09%, respectively. The best remote sensing extraction scheme of Camellia oleifera proposed in this paper can provide a reference for remote sensing monitoring of Camellia oleifera and other economic forest extraction in subtropical areas.

Unlike traditional passive optical sensors, hyperspectral LiDAR emits the active, full-waveform and gaussian laser pulse. After interacting with the vegetation leaf surface, the backscattered intensities for different waveforms return to the receiver and are then recorded. Previous research on spectral reflection characteristic have only focused on the circumstance at the incidence angle of 0°, the reflection characteristics at other multiple incidence angles and their impacts and errors on leaf chlorophyll content estimation have seldom been studied. This study used the hyperspectral LiDAR with 32 bands developed by our lab to obtain the leaf reflection spectrum over different incidence angles, and the intricate reflection characteristic was then analyzed at the bands with the high signal-to-noise ratio. After that, spectral indices were chosen to study the impact of leaf bi-directional reflection characteristics measured by hyperspectral LiDAR on leaf chlorophyll content estimation. The results show that, (1) the returned intensity of vegetation leaf measured by hyperspectral LiDAR gradually decreases as the incidence angle increases, but the bi-directional reflectance factor (BRF) does not show the same trait with intensity. There are two different features with the incidence angle in the visible and near-infrared bands. The maximum BRF value for visible bands occur in the incidence angles of 0° to 10°, but 60° for the maximum BRF values for the near-infrared bands. The minimum BRF values for all bands both occur in the incidence angle of 45°, and the difference between the maximum and minimum values is about 0.1. The changing trend of the BRF at the incidence angle of 10° to 60° for the visible and near-infrared bands decreases first and then increase; (2) BRF has a great impact on chlorophyll retrieval accuracy based on the relationship analysis on spectral indices and chlorophyll content at different incidence angles. However, there is not a synchronous decrease or synchronous increase trend for R2 and RMSE. Specifically, R2 decreases first, then increases, and finally decreases to the minimum at the incidence angle of 50°, in which the increase occurs at 60°. RMSE presents the opposite changing trend. For the different spectral indices, R2 owns a 4 times fluctuation with 0.14~0.63. RMSE owns about 1.5 times fluctuation with 0.5~0.8 mg·g-1. The great changes of R2 and RMSE reveal the essential impact of bi-directional reflection characteristics on leaf chlorophyll content retrieval.

Rapid and accurate monitoring of paddy rice planting areas distribution plays an important role in formulating regional agricultural production policies and protecting regional food security. With the successful launch of FY series satellites, domestic satellite data have been increasingly used in crop information monitoring, but there are few studies on the extraction of paddy rice planting distribution information based on FY data. In order to quickly and accurately obtain paddy rice planting distribution information and explore the application potential of FY remote sensing data in monitoring paddy rice planting distribution, the study was conducted to extract paddy rice planting distribution based on FY-3 MERSI data in Panjin county, Liaoning Province. Five images of FY MERSI data during the growth period of paddy rice in 2019 were used to calculate the Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), Ratio Vegetation Index (RVI) and NDWI-NDVI. The temporal sequence analysis of these vegetation indices was carried out on the interest areas of six land cover types in Panjin county, including paddy rice, building land, water body, natural vegetation, natural wetland and dry land. The optimal recognition mode and threshold were determined using NDVI, NDWI, RVI and NDWI-NDVI time series curves, and the remote sensing extraction model of paddy rice planting distribution was established. First, the paddy rice planting distribution was roughly extracted according to NDWI-NDVI>-0.14 at the transplanting stage and NDWI-NDVI<-0.4 at the heading stage. Then, other land cover types were masked based on the difference of NDVI, NDWI and RVI curve characteristics between paddy rice and other land cover types, and the spatial distribution of paddy rice planting in the study area in 2019 was obtained. Based on field survey data, the accuracy of paddy rice planting distribution in the study area was verified, and the overall accuracy was 75%. Accuracy verification was also conducted based on remote sensing visual interpretation data, the overall accuracy, Kappa coefficient, paddy rice mapping accuracy and user accuracy were 80.80%, 0.61, 80.00% and 86.96%, respectively. The paddy rice planting area in the study area in 2019 was 116 618.75 hm2, consistent with the data published in the 2019 Panjin Statistical Yearbook. The study shows that extracting paddy rice planting distribution based on FY-3 MERSI remote sensing image can satisfy the requirements of remote sensing monitoring of regional crop planting distribution. FY-3 MERSI has great application potential in extracting crop planting distribution information. The study enriches the remote sensing data sources for crop planting distribution monitoring and provides a theoretical basis for promoting the practical application of FY data.

Exploring the relationship between spectral characteristics and nutrient content of the grassland communities is of great significance for promoting the application of rapid non-destructive testing technology in grassland fertilization management, which can be used to diagnose the nutritional status of the grassland communities by hyperspectral technology. A typical grassland community in the Loess Hilly-gully region on the Loess Plateau, Bothriochloa ischaemum community, was investigated with treatments of four nitrogen (N) addition (0, 25, 50, and 100 kg·N·ha-1·yr-1) and four phosphorus (P) addition treatments (0, 20, 40, and 80 P2O5·kg·ha-1·yr-1). Based on hyperspectral and community N and P nutrient content measurements, combined with the first derivative treatment in the red-edge region, 18 characteristicspectral parameters consisting of vegetation indexes, characteristic bands and red-edge parameters, the characteristicspectral parameters sensitive to the N and P content and N∶P ratio of B. ischaemum community were screened by multiple linear stepwise regression (SWR) methods, and an inverse model was established to estimate the aboveground total N content and total P content and N∶P ratio in the community. Results showed that the N and P content of the B. ischaemum community increased with N application, and the N∶P ratio decreased with P application; the spectral reflectance under N and P addition is inversely proportional to fertilizer application in the visible band and positively proportional to fertilizer application in the near-infrared band, and the “double-peak phenomenon” of the first derivative in the red-edge region was significantly affected by N and P addition. Among them, TBSI, R910 and AMP contributed significantly to the model for N estimation (R2=0.87, F=18.8***), while DVI, mSR705, R430, R660 and AMP contributed significantly to the model for P estimation (R2=0.91, F=20.51***), and Slope725 contributed the most to the model for the estimation of N∶P ratio (R2=0.54, F=5.14***). This study used hyperspectral technology to achieve a rapid estimation of the N and P content of the B. ischaemum community, and based on the significant correlation between N and P content and N∶P ratio and spectral feature parameters, the parameter combination with the highest accuracy was selected, which laid a foundation for the method and parameter selection of monitoring grassland nutrient content after N and P addition at large scale.

Inner Mongolia can mine up to 100 tons of yellow to colorless feldspar yearly, with good transparency and concentrated distribution. It can be treated to meet the market demand. Guyang feldspar can be used as a gem resource with great development prospects. In this paper, the basic gemological characteristics, chemical composition and vibration spectral characteristics of feldspar in Guyang County, Inner Mongolia are systematically studied by using a series of testing technologies such as laser Raman spectroscopy, X-ray fluorescence spectroscopy, Fourier transforms infrared spectroscopy, electron probe and conventional gemological testing methods. The results show that the crystal form of feldspar raw stone samples in this area is mostly gravel, the refractive index is 1.555~1.570, the birefringence is 0.008~0.010, and the density is 2.65~2.68 g·cm-3. The UV fluorescence characteristics of the samples show that they are inert under long waves (365 nm) and short waves (254 nm). X-ray fluorescence spectrometer analysis shows that all samples contain a certain amount of Al, Si and Ca, and a small amount of Ti, Fe, Mn, Mg and Sr. According to the chemical molecular formula calculation and the proportion of end groups of feldspar according to the test results of electron microprobe, this kind of sample belongs to medium feldspar. The infrared absorption peak of feldspar is mainly between 1 200~400 cm-1. From albite to anorthite, it increases with the grade of feldspar. In the infrared absorption spectrum, the absorption peaks of 590 and 650 cm-1 are shifted to the low wavenumber range, to 575 cm-1± and 624 cm-1± respectively. The absorption peaks of Guyang middle feldspar studied in this paper are located at 578 and 632 cm-1 respectively, which is in line with the characteristics of the infrared spectrum changes of feldspar sequence and belongs to the infrared absorption spectrum characteristics of typical middle feldspar. The Raman peaks of this kind of feldspar are composed of seven main Raman peaks: 102, 186, 290, 489, 516, 572 and 800 cm-1. The 102, 186 and 290 cm-1 peaks below 450 cm-1 are caused by the vibration between metal cations and oxygen (［M—O］). The splitting degree of the Raman peaks at 290 cm-1 and 490 cm-1 can indicate the order of Al/Si in silicate minerals. The Raman peaks at 489, 516 and 572 cm-1 belong to the bending vibration spectrum of O—Si(AL)—O and the antisymmetric stretching of Si OBR Si (AL). The comparative analysis with feldspar from other producing areas can be used as one of the identification basis. Based on the above analysis, this feldspar’s composition and main structure are analyzed and discussed.

Changes in environmental conditions could alter the composition structure and chemical properties of dissolved organic matter (DOM), then affect its biogeochemical cycling processes. In this study, synchronous fluorescence spectroscopy, three-dimensional fluorescence spectrum, and Fourier transform infrared spectroscopy (FTIR) combined with two-dimensional correlation spectroscopy were applied to evaluate the effect of pH on DOM characteristics and its interaction with Cu2+. (1) Our results suggested that the fluorescence intensity of different DOM components increased remarkably along with the pH value increasing from 5 to 10. Humic-like components showed the most significant changes, and fulvic-like components responded fastest to pH changes. It was caused by pH changes induced exposure of some functional groups, such as carbonyl, phenolic, and carboxyl. (2) Two-dimensional correlation analysis of fluorescence spectra revealed that changes in pH could significantly affect the Cu2+-binding capacity of different DOM components but could not affect the binding sequence to Cu2+ with DOM. Three fluorescent components were identified by parallel factor analysis of the three-dimensional fluorescence spectrum. The nonlinear fitting of quenching curve for fluorescent components quantitatively verified our results. (3) FTIR results showed that, under hinger pH conditions, DOM has more binding sites and stronger binding affinities with Cu2+. The structural change of DOM upon Cu2+ addition under pH 5 and 10 conditions followed the order ofpolysaccharide C—O>phenols>aldehyde and ketone CO>aromatic C—H and polysaccharide C—O>amide II C—N>phenols>aliphatic C—H>aldehyde and ketone CO>carboxyl C—OH>aromatic C—H>carboxyl CO, respectively. Furthermore, two-dimensional hyperspectral correlation analysis of fluorescence spectra and FTIR indicated that humic-like fractions of DOM participated in the Cu2+binding after the phenolic and aryl groups.

Lake Bosten is the largest inland throughput freshwater lake in the northwest arid zone of China. In recent years, the lake ecosystem and the drinking water safety of the surrounding residents have been seriously affected by the increase in human activities and wastewater discharge in the basin, and the impact of riverine input on the lake water quality needs to be focused on. In this study, the three-dimensional fluorescence spectra of coloured dissolved organic matter (CDOM) measured in summer and autumn were analysed in parallel.Three fractions of CDOM from Lake Bosten were identified: terrestrial humic fraction C1, tyrosine-like fraction C2 and tryptophan-like fraction C3. The effect of river input on the DOM of Lake Bosten was also analysed based on Pearson correlation. The results show that river input’s influence on Lake Bosten’s DOM differs between seasons and is directly related to the seasonal changes in river water quality. In summer, the water entering the lake from the Kaidu River mainly comes from winter snowmelt and carries a large amount of terrestrial humus into the lake, while in autumn, it is mainly glacial meltwater. The content of terrestrial humus is reduced as the flow into the lake increases, showing that the overall DOM concentration near the estuary is higher in summer and lower in autumn. The three components C1, C2 and C3, were found to be more abundant in the western part of the lake where the Kaidu River and the Yellow Water Ditch enter the lake, and the three components were similarly affected by seasonality. The correlation between DOM and conductivity was also carried out at sampling points near the mouth of the Kaidu River and other areas. It was found that the influence of external river input on the DOM of Lake Bosten was mainly concentrated near the mouth of the river, and the relationship between DOC and CDOM in the area near the mouth of the river was significant so that the CDOM could be inferred by remote sensing before estimating the DOM content. The study of the influence of river inputs on the composition of dissolved organic matter in Lake Bosten in different seasons is of great importance to protect the lake’s ecological environment and improve the lake’s water quality.

The extensive existence of microplastics in the environment has potential exposure risks. So it is very necessary to develop a reliable and accurate monitoring method. This paper has studied two quantitative methods of microplastics based on Raman spectroscopy. The first is the quantitative analysis of the mass concentration of microplastics in suspension. Raman spectroscopy was used to measure the characteristic peak intensity of microplastic suspension calibrated by silica,and the relationship between the characteristic peak intensity, and the concentration of microplastics was studied. The results showed a good correlation between the concentration of microplastics and the characteristic peak intensity (R2=0.96), but this method needed stirring and ultrasonic oscillation to prepare the suspension in advance. Before the measurement, the characteristic peak intensity of microplastic suspension should be calibrated by silica. Moreover, the result was not good when the concentration was high. All these problems reduced the practicability of this method. The second is quantitative analysis of microplastics’ quality on the filter membrane’s surface. The aerosol generator generates microplastic aerosol samples, which were collected on the surface of the filter membrane. Raman mapping mode was used to measure the characteristic peaks of microplastic samples on the filter to determine the existence of microplastics to determine the correlation between peak recognition frequency and the mass of microplastic on filter. The results showed a good correlation between peak recognition frequency and the mass of microplastic measured by this method (R2=0.95). However, the traditional sampling method made microplastic samples unevenly distributed in a large area and made a small proportion of the measurement area. All these problems led to a large relative measurement error (2%~6%). In this study, aerosol micro-concentration technology was used to sample under the condition of keeping the surface deposition density of samples unchanged to reduce the deposition area of samples, reduce the collection amount of microplastic samples and increase the proportion of measurement area. The results showed that the correlation between peak recognition frequency and mass of microplastic could be effectively improved by aerosol micro-concentration technology (R2=0.97), and the relative measurement error could be reduced to 2%~4%. This method does not need complex operations such as preparation and calibration of the microplastic suspension, and it can directly collect microplastics in the air on the filter for quantitative analysis of microplastic quality, reducing the time required for measurement. This method is expected to be applied in the real-time measurement of microplastics in the environment.

Sulfur dioxide (SO2) and nitrogen oxides (NOx) as important primary emissions in the atmosphere. Anthropogenic activities of SO2 and NOx excessive emissions will cause great harm to the ecological environment and human health, in 2018 the Ministry of Environmental Protection “Announcement on the implementation of special emission limits for air pollutants in cities in the Beijing-Tianjin-Hebei air pollution transmission corridor” on the “2+26” cities need to implement special emission limits for SO2, NOx and other air pollutants, so it is important to understand the distribution and emission of SO2 and NOx in these cities for air pollution prevention and control. As one of the heavy industrial cities with the most serious air pollution among the “2+26” cities, Tangshan City has implemented many air pollution prevention and control measures in recent years, but the air quality is still not optimistic. Therefore, in order to obtain the time and space distribution of the main pollutants in the urban area of Tangshan City, to quantitatively analyze the emissions of different regional sources, and to identify the sources of the main pollutants, a mobile pollution gas monitoring system based on mobile DOAS techniques was used from February 26 to March 1, 2021, to conduct aerial observation experiments for the urban area of Tangshan City and some industrial parks (steel, thermoelectric and coking enterprises) to obtain the Spatial distribution of NOx and SO2 along its course and the emission fluxes in the moving area. The experimental results show several areas with high NO2 values in the first ring of Tangshan City, all of which are located at interchanges and junctions where vehicles are concentrated. The NO2 and SO2VCD obtained in the walkway of the industrial park are both higher, 1.75~1.99 times and 2.21~3.44 times higher than those of the first ring, respectively, and there are high NO2 and SO2 emissions from some enterprises in the industrial park. Combining the ratio of vertical column concentration SO2/NO2 and the ratio of near-ground concentration CO/NO2 and using the Pearson correlation coefficient to determine the correlation between SO2 and NO2 column concentration and NO2 near-ground concentration and column concentration, further analyzing the main pollution sources in different areas, the results show that the lowest SO2/NO2 obtained by the first ring walkway is 0.42, CO/NO2. The highest correlation r between NO2 surface and column concentrations reached 0.56. The highest SO2/NO2 and lowest CO/NO2 was 0.81 and 7.13 in the March 1 aerial walk in Fengnan Industrial Park, with a good correlation r between SO2 and NO2VCD of 0.787. The air pollutants in the first ring area of Tangshan City are mainly vehicular traffic exhaust emissions. The sources of air pollutants in the Fengnan Industrial Park are dominated by a large amount of NO2 and SO2 released from elevated point sources (chimneys) during industrial production.