Non-destructive spectral technologies are important methods of information extraction of cultural relics, which can obtain the relevant historical and artistic information on cultural relics in situ, identify the conservation and damage status of cultural relics, and traces the previous restoration. These technologies provide a scientific basis for the assessment of cultural relic conservation conditions, the discussion of disease mechanisms, the research of load information, and the exploration of production materials and craftsmanship. The application of X-ray fluorescence spectroscopy (XRF), laser-induced breakdown spectroscopy (LIBS), X-ray photoelectron spectroscopy (XPS), Raman spectroscopy (RS), infrared spectroscopy (IR), diffuse reflectance spectroscopy (DRS), multispectral (MSI) and hyperspectral imaging (HSI) techniques for non-destructive analysis of cultural relics were presented. Because of the different types, sizes and preservation status of cultural relics, portable and fixed instruments have their characteristics in the non-destructive analysis. Portable and micro-XRF can perform qualitative and quantitative analysis of cultural relics, and Macro-XRF can analyze multi-layer structures and obtain elemental distribution and hidden information of imaging patterns. LIBS can detect low atomic number elements such as lithium and carbon that cannot be detected by XRF and conduct depth and profile analysis of cultural relics. XPS can get the chemical state and elements’ content on the sample’s surface. RS can identify the phase composition of cultural relics, confirm the composition and deterioration of cultural relics, and evaluate the protective effect. Resonance Raman spectroscopy is sensitive to aromatic compounds with RS activity and can analyze organic dyes on textiles and paper. Surface-enhanced Raman spectroscopy can identify spectral peaks that conventional RS cannot identify. In infrared spectroscopy, the application of near-infrared spectroscopy has been expanded from organic to inorganic cultural relics. IR reflection spectroscopy can compensate for the deficiency of IR absorption spectroscopy, and it has been used in the research of color painting craftsmanship, surface deterioration layer of cultural relics, and analyzing the multi-layer structure of paint layer. DRS has unique advantages in pigments and dye analysis. MSI and HSI have the characteristics of the integration of spectrum and image, which can perform qualitative, quantitative and localization analysis on the study area. They have been used in the restoration, extracting hidden information and identifying cultural relics. Each spectrum has its characteristic and limit in the test function of cultural relics. In order to obtain comprehensive information heritage as nondestructively as possible, elemental and phase structure analysis, compositional analysis and imaging technology are often used together, combined with metrology and algorithm analysis, to improve the detection results and expand the application scope of non-destructive technology. Finally, the development prospect of non-destructive spectroscopy technology has prospected.

There are two kinds of oil spills on the sea surface: non-emulsification and emulsification. The scientific detection and evaluation of oil spills on the sea are helpful to the recovery and treatment of oil spill pollution and formulate of an emergency plan. For non-emulsified oil spill, mainly exists in the form of oil film, and its thickness becomes an important evaluation index of oil spill; for emulsion spills they mainly exist in the form of water in oil or oil in water, and the oil-water ratio can be used as the basis for evaluation. Laser-induced fluorescence (LIF) technology is considered one of the most effective methods for oil spill detection. Oil film thickness inversion based on LIF detection technology has seen the relevant research algorithm, but there is no corresponding quantitative method of oil spill on the sea surface, and the oil spill on the sea surface will bring greater harm to the marine environment. Therefore, it is an urgent task to analyze and study the information on emulsion oil spills on the sea surface. Based on the detection mechanism of LIF system, an equivalent estimation model for water in oil emulsion spillage is proposed: Firstly, the oil spill of continuous phase in water in oil is regarded as an oil film with the same optical properties, the water droplets of all dispersed phases are regarded as a whole, which is equivalent to a thin water layer, in order to keep the equivalent model consistent with the external environment of the actual emulsion, an oil surface is covered on the thin water layer, so that the estimation of oil spill volume of water in oil emulsion is converted into the calculation of equivalent oil film thickness; Secondly, according to the radiation transmission process of light, the equation of fluorescence information that the system can receive is established, and the calculation formula of oil film thickness is sorted out. That is, based on knowing the oil type, the corresponding thickness value can be obtained by substituting the fluorescence intensity value measured by the system, and then the oil spill amount can be estimated. The paper also analyzes the error caused by the equivalent model through an example to verify the applicability and effectiveness of the estimation method under the equivalent model. That is, when the oil content and thickness of water in oil emulsion are in a certain range, there is a small error between the actual oil spill thickness and the equivalent oil film thickness. The equivalent treatment method studied in this paper can provide a new method for estimating emulsified oil spills on the sea surface, which has important guiding significance and certain innovative value for solving this problem.

Computed tomography imaging spectroscopy (CTIS) has the ability of traditional imaging spectrometers to acquire two-dimensional images and one-dimensional spectra of the target space. It also has the characteristics of high-throughput measurement and non-scanning imaging, which has a wider range of applications in the field of spectral imaging. According to the Center Slice Theorem, the performance of the CTIS is mainly restricted by the performance of the Focal Plane Array(FPA) and the two-dimensional dispersive element. The previous research mainly focused on improving the design of the two-dimensional dispersive element to increase the diffraction order and projection angle to enhance the sampling and spectral accuracy. This paper focused on the FPA two-dimension dispersion projection measurement. It proposed a method of combining parallel block-compressed sensing and CTIS to establish block-compressed sensing computed tomography imaging spectroscopy (BCSCTIS) model, which uses the low-resolution FPA to achieve the measurement of the high-resolution dispersive projection and further achieves the performance higher than traditional direct computed-tomography measurement. In order to verify the correctness and feasibility of the BCSCTIS model, this paper carried out a BCSCTIS simulation experiment and carried out the corresponding optical system experiment in turn. The system matrix from a three-dimensional spectral cube to a two-dimensional dispersion projection was simulated in the simulation experiment. The hyperspectral data set was used to quantitatively compare the reconstruction results of the direct measurement model of dispersion projection with the that parallel block compressed sensing measurement model, and the results showed parallel block compressed sensing computed tomography imaging spectroscopy can obtain higher spectral reconstruction quality. It can achieve a significant improvement in spectral projection acquisition resolution and spectral reconstruction quality higher than the performance of FPA itself. Furthermore, the CTIS optical experimental data were processed by parallel block compressed sensing, and the effectiveness and feasibility of BCSCTIS were further verified. In the optical experiment, the system matrix was accurately calibrated point-by-point using a supercontinuum and a reflective digital micro-mirror device (DMD). Also a parallel calibration method to improve the calibration efficiency was proposed. In the experiment, the calibration time is reduced to one-fourth of the single-point calibration. The conclusion is consistent with the simulation experiment, which further confirms the correctness and feasibility of the proposed BCSCTIS.

Terahertz spectral imaging not only includes the intensity information in two-dimensional image space but can also obtain spectral information in the terahertz band, constituting a three-dimensional data matrix. Due to the limitation and influence of the internal hardware of the Terahertz imaging system, the signals in the higher frequency band of the terahertz frequency domain have weak energy and low signal-to-noise ratio, resulting in the problems of low resolution and low contrast of the terahertz images. Therefore, this paper improves the spatial resolution and edge detail visibility of terahertz spectral imaging by using a three-dimensional data matrix and a suitable algorithm. In this paper, a three-dimensional portable Terahertz time-domain spectroscopy imaging system is built to realize the two-dimensional scanning of standard high-resolution plates. The signals collected by the system were compared in the time domain and frequency domain, respectively. The spatial resolution and depth of field of the imaging system were calibrated by combining the Rayleigh criterion and resolution scale, and the spatial resolution algorithm for improving THZ spectral imaging was studied. Then, aiming at the characteristics of low SNR, low contrast and complex noise causes in the high-frequency region of the Terahertz frequency domain, combined with the image denoising theory of deep residual learning, a terahertz image depth denoising network is proposed, which introduces the real “terahertz residual noise” in the imaging system in the training set. Finally, the reconstructed images are compared with the original images and the traditional terahertz denoising algorithm results. The denoising effects of different algorithms on the high-frequency images in the terahertz frequency domain are evaluated from subjective and objective aspects. Experimental results show that the limit spatial resolution of the proposed algorithm is about 157 μm, the saddle-peak ratio of the Rayleigh criterion at the limit spatial resolution of the denoised image is 0.623, and the overall image contrast is 46.635. The spatial resolution is about double that of traditional imaging methods, and the contrast is about 26% higher. The results of this study provide a new standard for high spatial resolution and high visibility THZ spectral imaging and provide a new solution to the problem of image noise in the higher frequency region of the THZ frequency domain.

The vibrational and rotational energy levels of many biomolecules are in the terahertz band, so terahertz time-domain spectroscopy can be used to detect biomolecules. In addition, the photon energy of terahertz wave is low, only in the order of MeV, it will not damage biological samples in the detection process, so terahertz time-domain spectroscopy has an extensive application prospect in the research fields of biochemical detection in the future. Studies have shown that most biomolecules need to be in a liquid environment to give full play to their biological activity. However, the hydrogen bond in an aqueous solution will produce strong absorption in the terahertz band. What’s more, water molecules are polar molecules, and terahertz wave has strong resonance absorption to polar molecules, which makes it very difficult to detect active biomolecules in a liquid environment by terahertz technology. Therefore, many research teams combine terahertz spectroscopy with microfluidic technology to reduce the impact of various factors on biomolecular detection. Microfluidic technology reduces the distance between liquid sample and terahertz wave by reducing the depth of liquid pool in microfluidic chip, reducing the absorption of terahertz wave by aqueous solution. In this study, a double-layer microfluidic chip was prepared by cycloolefin copolymer (COC: Zeonor 1420R), and its transmittance to terahertz wave was as high as 95%. The length and width of the liquid pool in the microfluidic chip are 4 cm, and the depth is 50 μm. Furthermore, there are a large number of free-moving cations and anions in the electrolyte. Therefore, we used an external electric field device to apply voltage to the microfluidic chip injected with liquid samples to explore the influence of free-moving cations and anions in the electrolyte on terahertz transmission characteristics. The external electric field device includes a power supply, a ZVS circuit encapsulated in a plexiglass box and a DC high voltage package with an output voltage of 10 000 V.On this basis, we studied the terahertz wave transmission characteristics of five potassium salt solutions with the same concentration and these five potassium salt solutions in a constant electric field for different times, which provides a basis for further enhancing the application of terahertz time-domain spectroscopy in biochemistry. Moreover, the electrolyte contains many positive and negative ions, which will move under the action of the external electric field. And it provides technical support for terahertz time-domain spectroscopy to study the dynamic characteristics of electrolytes.

Wavelength modulation spectroscopy (WMS) based on harmonic detection, a modality of the tunable diode laser absorption spectroscopy (TDLAS), has been widely employed for gas properties measurement and combustion diagnosis, given its advantages of contactless, rapid, high sensitivity and accuracy. In recent years, to expand the application scope of the WMS method and reduce the calibration error of spectral parameters, the research on the calibration-free strategy of the WMS method has gradually become a hotspot. Traditional calibration-free WMS method generally needs to simulate the absorption spectrum according to the spectral database combined with the laser modulation parameters, which puts forward high requirements for the prior spectral parameters and hardware parameters. A rapid and accurate calibration-free wavelength modulation spectroscopy (WMS) method for gas parameter measurement based on 2nd and 4th order harmonics is proposed. Compared with the traditional calibration-free WMS method, the proposed method has the following characteristics and advantages. First, the proposed method, analytically deduced from a much more accurate Voigt function model, enables speedy measurement down to milliseconds and general suitability for various degrees of line-shape broadening. Second, the proposed method only needs the algebraic calculation of the central peak height parameters of the 2nd and 4th harmonics to obtain the key spectral parameters, such as absorption line broadening and integrated absorption area, to realize the measurement of gas parameters such as concentration and temperature.Third,the proposed method does not need square iterative fitting calculation and high-order harmonic calculation, which reduces the requirements of hardware system. Fourth, instead of acquiring the entirely scanned absorption line-shape, the proposed methodonly requires extracting the peak values of the harmonics. This characteristic significantly benefits gas diagnosis at elevated pressure and/or temperature. Fifth, the proposed method only needs to use the spectral parameters such as absorption line intensity and low state energy in the spectral database. However, it does not need to use the prior parameters such as the self-broadening coefficient, temperature index and collision broadening coefficient of other components, which reduces the dependence on the spectral database. In order to verify the feasibility of the proposed method, a WMS measurement system was built in the laboratory environment. The absorption line of the CH4 molecule near 6 046.95 cm-1 was selected, and the 2nd and 4th harmonic peaks were used to measure the mole fraction of CH4 at room temperature. Experimental results show that under the absorption optical path length of 20 cm, the relative error of CH4mole fraction measurement is 1.19%, and the detection limit of the system is 4.28×10-6.

Solar far-ultraviolet radiation is one of the main sources of energy input into near space, and the response of the near space environment to solar eruptions is an important scientific issue to be further studied. Studying the radiation characteristics of the solar far-ultraviolet in the middle and upper atmosphere is an important basis for studying atmospheric composition and density changes, photochemical reactions and dynamic processes in near space. In this paper, using the far-ultraviolet data calculated by the FISM2 flare model and the earth’s middle and upper atmosphere data provided by the MSIS-E-00 model, the solar far-ultraviolet radiation from 120 to 190 nm is divided into 7 bands, and numerical simulations were performed using an atmospheric radiative transfer method based on the Lambert-Beer law. A total of 150 sets of flare data in 11 years from January 2010 to December 2020 were selected, and time-lag cross-correlation (TLCC) was used to evaluate the flare peak time difference between solar far-ultraviolet radiation and soft X-rays, using least squares (LS) method to calculate their flare peak flow relationship. Atmospheric radiative transfer equations were used to calculate the spectral properties, flux changes, and heating rate changes of the solar far-ultraviolet in near space (20～100 km) during flares. Finally, the deposition of solar far-ultraviolet radiation in the earth’s atmosphere is calculated. The results show that in the process of solar flare eruption, the flux of far-ultraviolet radiation changes significantly, and the flux peak is about 240 s earlier than that of soft X-rays. The wavelength increases with the increase of the wavelength; in the near space range of 20～100 km, the solar far-ultraviolet spectrum is almost completely absorbed, but due to the special absorption window structure of the atmospheric composition, part of the spectrum in the 185~190 nm band can reach an altitude of 20 km. In the near space region, the ratios of the far-ultraviolet fluxes at the time of the solar flare eruption and before the eruption were all around 2.0 in the seven bands, and the ratios of the peak heating rates were 1.22, 1.88, 1.35, 1.42, 1.23, 1.08 and 1.11. This paper verifies the feasibility of using far-ultraviolet radiation to sense solar flares in near space, provides a theoretical basis for optical detection experiments in near space and provides a reference for related research fields such as atmospheric inversion.

Numerous diseases in ancient earthen cultural relics are relative to water. Free water and absorbed water in earthen cultural relics’ pore structure bring different influences. During the evaporation of water, free water easily evaporates from soil in favor of drying soil. Relatively, the volatilization of absorbed water is slow, leading to salt damage. Therefore, it is the foundation for relic protection to identify free water and absorbed water in earthen relics and study volatilization behavior. The free water and absorbed water have different polarities in chemistry. FE fluorescence substance with double emission peaks is sensitive to the change of chemical polarity and hydrogen bond, used to indicate polarity difference. This paper introduced FE as florescence probe into the stimulated soil samples with different moisture time. By detecting fluorescence characteristic peaks online during water evaporation, the water form and the volatilization behavior of free water and bound water in soil samples were revealed. In the same way, the fluorescence spectra of the stimulated soil samples before and after consolidation were determined to research the effect of consolidation treatment on the volatilization behavior of different forms of water. FE probe has separable double emission peaks, and the excitation wavelength is in the visible region, avoiding soil’s highly absorptive ultraviolet region. The method proposed in this paper can sensitively detect the water form in earthen cultural relics and evaluate the properties in breathability and water resistance for consolidated earthen cultural relics through the volatilization behavior of different forms of water.

With the rapid development of infrared temperature measurement technology, infrared thermometer plays an important role in both military and civil fields, so it calls for higher measurement accuracy. It should be observed that a couple of years have witnessed an increasing number of people who are concerned that radiation source consisting of area blackbody is the vital device to calibrate the non-contact thermometer. Spectral emissivity is an essential parameter to describe the performance of radiation sources, while there is less research involved in the effect of blackbody surface topography on emissivity. It is manifested that the emissivity of the area blackbody is related to the combination structure of the evagination cone and coating. In order to design a remarkable performance area blackbody with high emissivity, the paper takes the area blackbody with convex cone structure as the basic model, adding components of spacing resembling grating and coating, and establishes an area blackbody model with different structural parameters, cell spacing and coating thickness. Graphite and silicon nitride is set as the base material and the coating material, respectively. Reflectivity is obtained by measuring the variation of reflected radiation with related soft. Then the emissivity is calculated through the relationship between reflectivity and emissivity, drawing the curve of spectral emissivity in the range of 3~14 μm. According to the electric field distribution of the surface and curve of spectral emissivity, we analyze the influence of parameters such as the ratio of width to length, coating thickness and spacing on emissivity. It is shown that the height of the structure unit is proportional to spectral emissivity, and emissivity is optimized due to narrower width, which increases as the ratio of width to length decreases. The dropping trend of spectral emissivity is changed by the coating structure in which the emissivity increased at the wavelength above 11 μm, and the emissivity rise with the increasing coating thickness. What’s more, the spacing structure is proportional to emissivity. The height and width of the unit structure of the original area blackbody are set at 10 and 1 μm, respectively. A coating with a thickness of 2 μm and a spacing structure with a width of 2 μm are sequentially added to the original model for simulation calculation. The optimized blackbody radiation source has advantages in the long wavelength band, and the spectral emissivity with a minimum value of 0.966 is stable in the range of 3~14 μm. The electric field distribution shows the influence of these structural factors on the surface energy of the area blackbody, and the optimized model parameters can be used to manufacture a realistic area blackbody. Simulation results illustrate that a smaller aspect ratio, thicker coating and larger spacing improve spectral emissivity, which provide a theoretical reference for the manufacture of radiation sources consisting of area blackbody with high radiation performance.

Remote plasma can effectively avoid the etching effect caused by electron-ion collision, improve the free radical reaction, and achieve a better modification effect. All of which have important applications in the field of membrane materials. In order to study the electron states and their variation laws in remote plasma, emission spectroscopy was used to diagnose the remote Ar plasma. Further, the effects of RF power, the pressure inside the reaction chamber, and distance from the discharge center on the emission spectral intensity, electron density, and electron temperature of the remote Ar plasma were also investigated. The results show that the characteristic peaks are more concentrated in the 690～890 nm region, which is dominated by ArⅠ atomic spectral lines, and that the variation pattern of the intensity of the spectral lines is the same as that of the electron density. The electron temperatures at different discharge parameters were calculated using the Boltzmann slope method and three selected ArⅠ spectral lines. The electron temperature changes with the change in the RF power, the pressure inside the reaction chamber, and the distance from the discharge center. When the RF power is increased from 30 to 150 W, the electronic temperature drops from 3 105.39 to 2 552.91 K. When the pressure rises from 15 to 25 Pa, the electron temperature drops from 3 066.53 to 2 593.32 K, then rises to 2 661.71 K when the pressure increases to 35 Pa. The electron temperature increases from 0 to 10 cm from the discharge center due to the increase in plasma potential, while after 10 cm, the electron temperature decreases and tends to 0 K at 80 cm from the discharge center. By analyzing the Stark spread of the Ar Ⅰ 696.894 spectral line, the electron density of the remote Ar plasma was calculated, and it was discovered that the electron density is up to 1016 cm-3 in order of magnitude. When the RF power was increased from 30 to 150 W, the electron density increased from 2.15×1016 to 2.88×1016 cm-3, from 2.36×1016 to 2.90×1016 cm-3 when the pressure was increased from 15 to 25 Pa, and decreased to 1.89×1016 cm-3 when the pressure was increased to 35 Pa. The electron density decreases rapidly as the axial distance increases, reaching 0 cm-3 at 80 cm from the discharge center. The discharge parameters and axial distance can be adjusted to achieve a low concentration of electrons and ions in the atmosphere, thereby effectively avoiding the etching effect caused by electron-ion collision and achieving a better modification effect.

Concentration and temperature measurement are of great significance for engine design, CFD model establishment and numerical simulation verification. Planar laser-induced fluorescence (PLIF) technology has the advantages of simple experimentation, object selectivity and high measurement sensitivity. Among them, the OH-PLIF measurement of concentration and temperature is more widely applied due to their theoretical maturity and technical convenience. However, the application of OH-PLIF in the kerosene combustion field is limited because of the great interference of residual kerosene. This paper focus on the kerosene interference problem in the measurement of OH fluorescence distribution. The absorption and fluorescence emission spectra of OH and kerosene were comparative analysis. The absorption spectrum of kerosene vapor is obtained from the intensity ratio of the strontium halogen lamp before and after its pass through the kerosene vapor. Compared with the isolated absorption line of OH in the 260~320 nm band, the kerosene absorption is a wide absorption band from 240 to 320 nm. The kerosene absorption band completely covers the OH excitation line. When the OH is excited in this band in kerosene combustion, it is inevitable to stimulate the kerosene to produce fluorescence. On the other hand, kerosene fluorescence and the OH/kerosene mixed fluorescence were measured by adjusting the excitation wavelength. The kerosene fluorescence emission centered at 290 and 340 nm, while OH fluorescence is mainly concentrated at wavelength 300~320 nm. Combined with the absorption spectrum, kerosene fluorescence covers OH fluorescence indicating that the OH measurement in the 280 nm band can not be carried out by a single frequency domain filter when there are residua. In this paper, the kerosene interference was eliminated by measuring kerosene simultaneously and deducted from mixed fluorescence then. The fluorescence was detected selectively by adding an extra camera based on the ordinary OH PLIF system. The two ICCD cameras combining (315±15) and (360±6) nm bandpass filters were used for OH/kerosene mixed fluorescence and kerosene fluorescence measurement respectively. Then the OH fluorescence of a kerosene Bunsen burner and an engine model kerosene combustion was obtained by subtracting the kerosene from mixed fluorescence pixel by pixel. These results confirmed the feasibility of the kerosene interference elimination method and laid the foundation for the measurement of temperature distribution in kerosene combustion.

P-dibromobenzene (C6H4Br2) has a wide range of applications in the chemical industry, but it is also one of the organic pollutants that threaten the ozone layer. The study of the dissociation characteristic of the molecule under an external electric field has important reference value for protecting the ozone layer. Under the action of different external electric fields (-0.025~0.025 a. u.), the p-dibromobenzene molecule is optimized based on B3LYP/6-311+G(d,p) based on density functional theory (DFT), the ground state geometric structure is computed. They are then using the Time-dependent density functional theory (TD-DFT) and B3LYP/6-311+G(d,p) basis set to calculate the ultraviolet absorption spectrum of the p-dibromobenzene molecule to guess the dissociation property of the molecule. Finally, scanning the potential energy of the two C-Br bonds of the molecule gave direct evidence of the dissociation property of the p-dibromobenzene molecule. Studies have shown that under the action of an external electric field, the ground state geometric structure, spectral characteristics, potential energy curve and potential energy surface of p-dibromobenzene molecules have undergone major changes. With the increase of the external electric field, the 3C-12Br bond length and total energy of the molecular system of p-dibromobenzene gradually decrease, and the 6C-11Br bond length and dipole moment gradually increase; the increase of the 6C-11Br bond length indicates that the 6C-11Br bond energy of p-dibromobenzene molecule is reduced, and the 6C-11Br bond is the easiest to break. The energy gap first increases and then decreases with the increase of the external electric field. The decrease in the energy gap indicates that the molecule is more prone to chemical reactions. The peak intensity at the stretching vibration of the C-Br bond gradually decreases. The peak intensity at the stretching vibration of the C-Br bond gradually decreased, and the absorption peak of the ultraviolet absorption spectrum increased slightly at first and then decreased abruptly. The energy-enhancing properties indicate that the vibrations intensify, and the chemical bonds become active. The 3C-12Br bond of the p-dibromobenzene molecule was scanned under an external electric field, and the potential energy curve of the 3C-12Br bond of the molecule was obtained. When the intensity of the external electric field is -0.02 a.u., the highest energy of the right barrier of the molecule is the same as the lowest energy. Flat, the molecular 3C-12Br bond is broken; continue to scan the potential energy of 6C-11Br under this electric field strength, the molecular 6C-11Br bond is broken, so the p-dibromobenzene molecule will gradually decompose the separation phenomenon. Under the action of an external electric field, scan the two C-Br bonds of p-dibromobenzene simultaneously to obtain the molecule’s potential energy surface. When the external electric field intensity is 0.02 a.u., the diagonal potential energy of the potential energy surface decreases, and another dissociation channel appears. Therefore, synergistic dissociation of p-dibromobenzene molecules may occur. These results provide data guarantee for the experimental study of the degradation mechanism of p-dibromobenzene molecules in an external electric field and have important reference significance for the study of the dissociation characteristics of the molecular system.

X-ray fluorescence (XRF) spectrometers have been developed rapidly and applied widely. They are quick and accurate without complicated pretreatment or consumables. In some industries, XRF has partially replaced traditional AAS, ICP and ICP-MS. The limit of detection (LOD) is one important evaluation indicator for its application performance. There are various LOD calculating methods, generally equal to 3 times standard deviation (SD) of blank samples. Analyte cannot be detected when it is lower than the LOD, could be qualitatively analyzed when higher than LOD and lower than the limit of quantification (LOQ), and could be accurately analyzed when higher than LOQ. XRF LOD calculation method is different from other common methods because measured values of traditional analysis are continuous distribution in accord with Gaussian distribution, however, XRF measured values conform to Poisson distribution, which is a discrete distribution and could approach Gaussian distribution only with high enough counts. In practical analysis, accumulating countsis not worth taking too long. This paper introduces seven methods for calculating LOD, including X-ray Poisson distribution method, K times SD method, linear calibration method, RSD method, SD method, environmental monitoring analysis method, marine monitoring specification and analysis method. Taking the detection data of XRF heavy metal instruments as an example, Pb, As and Cd elements were tested in six kinds of rice powder reference samples, and each method’s calculation processes and consideration factors were compared in detail. On account of the difficulty off inding the completely blank sample, the approximate blank sample was used this paper instead. The poisson distribution methods are quick and accurate, requiring only two measurements at the fastest. The linear calibration method, which considers comprehensive factors, is generally considered the most accurate method of LOD calculations and can be used as a reference value compared with others. RSD method and SD straight line extrapolation method need more test times, which can be used without blank samples or spectrum intensity. The RSD method can be used as a necessary condition for determining LOD. When RSD>43%, the analyte cannot be qualitatively detected.

Large amounts of particulate and volatile matter derived from incense burning have become one of the important sources of indoor air pollution. At present, the research about the effect of burning incense on the local air environment mainly focused on determining particulate matters, volatile gases, and organic pollutants after incense burning. The research about the emission efficiencies and emission factors of heavy metals and the effect of particulate heavy metals on indoor air quality after incense burning has not been reported. To better understand the emission characteristics of heavy metals in Tibetan incense, nine kinds (S1-S9) of line-type Tibetan incense powder samples and their total suspended particulates (TSP) after burning were digested by microwave digestion method with programmed temperature procedure, and four primary kinds of heavy metal elements (i.e. As, Cd, Cu, Pb) in these powder samples. Their TSP were determined by the inductively coupled plasma optical emission spectrometry (ICP-OES) with the wavelength 188.980 nm for As, 214.439 nm for Cd, 327.395 nm for Cu and 220.353 nm for Pb, respectively. The experiment results are listed as follows: (1) The contents of four heavy metals are different in different Tibetan incense samples, and the average content of heavy metal from high to low is in the order of Cu, Pb, As and Cd. The source of Cd and Pb in different Tibetan incense samples is similar, remarkably different from As and Cu. (2) After burning, the amounts of smoke and dust from different Tibetan incense samples are different. The largest and minimum amounts are 94.75 and 38.52 mg·kg-1, respectively. (3) The emission amounts of As, Cd and Pb in the total suspended particles from different Tibetan incense samples depend on their contents in the incense sample. However, the emission amount of Cu is not only dependent on the incense samples but also affected by other factors. (4) The emission efficiency of heavy metals in Tibetan incense samples from high to low is in the order of Cd, As, Pb, and Cu. The emission factor from high to low is in the order of As, Pb, Cd, and Cu. (5) The emission of As and Cd from Tibetan incense samples after burning may adversely affect local atmosphere quality. Among these samples, the highest contents of As and Cd in the TSP of S3, S5 and S9 from the same manufacturer. The manufacturer is suggested to control the source and content of As and Cd stricter to decrease the heavy metal pollution in indoor air and the risk to human health after burning Tibetan incenses.

Compared to cell and sliced tissue samples, blood samples could be collected easier, and its biomedical constitution would show some relavant variations before clinical pathological symptoms. Raman spectroscopy provides molecular-related information about biomedical contents for clinical investigations in a rapid, nonlabeled, nondestructive and noninvasive way, presenting a significant application prospect for blood sample-based diagnosis. In this study, we present a reliable method for detecting breast cancer using blood serum combined with multivariate analysis methods. The blood serum samples were divided into healthy, early, and advanced cancer groups based on clinical pathological diagnosis. Using a quatz capillary tubes as sample holder, the spectral information was acquired to illustrating the biomedical constitution nature of the serum sample. The spectral classification models, which were built on the method of principal component analysis (PCA), linear discriminant analysis (LDA), supporting vector machines (SVM) and partial least squares discriminant analysis (PLS-DA), were utilized for unveiling the spectral variances among different investigated groups. And the leave-one-out cross-validation (LOOCV) method was adopted for evaluating the model classification performance. After that, we not only observed the resonance Raman spectral phenomena of carotenoid contents in serum but also identified the spectral variations of protein and lipid contents during breast cancer progression. By using the multivariate analysis methods, the representative spectral identities were recognized. Since then, the spectral classification accuracy of PCA-LDA model was found to be 99%. For three types kernel based PCA-SVM model, it was found that the linear kernel model reached 92% accuracy with parameter c=0.003, the classification accuracy of the RBF kernel model was 94% with parameter c=0.125 and γ=256, and the polynomial model presented 92% accuracy with parameter c=0.003 and d=11. Meanwhile, the spectral classification accuracy of PLS-DA was 80%. The obtained results could pave a theoretical and experimental foundation for serum Raman spectroscopy-based breast cancer early screening and diagnosis.

Yellow peaches are rich in nutrients and very popular with consumers. However, they are susceptible to bruising during processing and transportation, causing great economic losses to farmers and sellers. When using hyperspectral technology to detect fruit bruises, researchers only use spectral information or image features to build a bruise detection model but rarely use spectral fusion technology for fruit bruise detection. Moreover, spectral reflectance is easily affected by external stray light, some real information will be lost, and the image contains little information, It is difficult to identify images only relying on a few specific bands accurately. Therefore, to achieve an accurate classification of the degree of bruising of yellow peaches, different treatment methods can be developed according to the different degrees of bruising, and economic damage can be reduced. This study proposes a model that uses the spectral information of hyperspectral combined with image features to detect yellow peaches with different degrees of bruising. In this experiment, 180 yellow peaches were used as experimental samples. Firstly, the images of lightly, moderately and severely bruised yellow peaches were collected using hyperspectral images, A 100×100 pixel area was selected as the region of interest at the bruise location of each yellow peach, and the spectral information of the bruised area was extracted using ENVI4.5 software. Then, the principal component analysis (PCA) algorithm was used to reduce the dimensionality of the collected hyperspectral images, and the PC1 image was finally selected as the principal component image of the bruised yellow peaches among the first 5 PC images. The images corresponding to 6 characteristic wavelengths were selected as the feature images according to the weight coefficient curve of PC1 image, and the average grey value was used as the image feature of the bruised yellow peaches. Finally, PLS-DA models for yellow peach bruising were established using spectral information, image features and spectral information combined with image features respectively, and the performance of each model was evaluated using the correct classification rate. The results showed that the PLS-DA model based on spectral information combined with image features had the best discriminative effect, with classification accuracy of 85%, 90% and 100% for light bruises, moderate bruises and severe bruises respectively, and the acorrectoverall rate reached 91.7%. In order to further improve the accuracy and operational efficiency of the PLS-DA model, this paper uses the competitive adaptive reweighted sampling (CARS) algorithm to filter the spectral data in the fused data for the feature bands. The model built using feature bands combined with image features had the best classification prediction, with 95%, 90% and 95% correct predictions for lightly bruised, moderately bruised and heavily bruised yellow peaches, respectively, and acorrectoverall rate of 93.3%. In conclusion, this study shows that it is feasible to establish a PLS-DA model based on hyperspectral spectral data combined with image features to detect the degree of bruising of yellow peaches, which provides a theoretical basis for the post-harvest treatment of yellow peaches.

As an important metal material, steel is widely used in manufacturing and construction due to its good plasticity, toughness, and low price. China’s annual steel production and export volume ranks among the top in the world, and the scrap produced in the steel production process is an important resource. Accurate classification of steel scrap is a key link in electric furnace steelmaking, and it is also of great significance to the sustainable development of an environmental and energy. In order to improve the efficiency of scrap steel recycling, a method for intelligently identifying scrap steel grades using laser-induced breakdown spectroscopy (LIBS) combined with XGBSFS is proposed. Firstly, the LIBS data in the range of 170~400 nm were collected by Lapa-80 solid-state pulsed laser for 3 types of 18 different scrap steel samples. The gross error in the spectral data is eliminated by k-value verification, and the remaining data after the elimination is averaged. 28 groups for each sample, a total of 504 groups of average spectral data were obtained. Then, the spectral data is subjected to baseline correction, normalization and other preprocessing to reduce the influence of matrix fluctuations. Finally, the processed spectral data is divided into the training set and test set, and 16 characteristic spectral lines of Si, Cu, C and other elements in the spectral data are extracted as classification features for the model’s input. After the variables are optimized by the XGBSFS feature selection algorithm based on XGBoost, kNN and SVM are used to establish an intelligent identification model of scrap steel. The accuracy rates of the XGBSFS-SVM and XGBSFS-kNN algorithm models established in this study are 100% and 98.8% respectively, on the test set, and the input dimensions are also reduced from 16 dimensions to 2 dimensions, and the modeling times of the two models are respectively From 3.1 to 2.79 s, 3.26 s to 1.64 s. Compared with the algorithm model using SVM and kNN alone, the optimized model proposed in this paper has higher prediction accuracy, modeling efficiency, and better generalization ability. After comparing the comprehensive effects of modeling time and accuracy, the XGBSFS-SVM model was selected for the intelligent and rapid identification of different scraps. The experimental results show that the LIBS and XGBSFS feature optimization methods proposed in this study can effectively optimize the modeling of feature variables and provide a new technology for the rapid and intelligent identification of scrap steel types in industrial production and the recovery of steel.

Engine oil is the core component of the engine. It is easy to mix with water in the engine oil, which can easily accelerate the deterioration and deterioration of the engine oil, and then harms the safe operation of the engine. Detecting water in the engine oil is an important indicator to ensure the quality of the engine oil. Moisture is easy to accelerate the deterioration and degradation of engine oil, and it is harmful to the safe operation of the engine, and its detection is an important index to ensure the quality of engine oil. Therefore, near-infrared (NIR) spectroscopy combined with partial least squares (PLS) regression method was used to detect engine oil with different water content. Firstly, the mechanism of 931, 1 195~1 212 and 1 391~1 430 nm wavelengths with strong absorption peaks were analyzed according to the NIR characteristics of water-containing engine oil. Orthogonal signal correction (OSC) and several other spectral pretreatment methods were used to construct the PLS regression model, and the characteristic wavelength was selected according to the regression coefficient. The results showed that the PLS model pretreated by OSC had the better predictive ability, while the pretreated by MSC and SNV reduced the correction ability of the model. The 166 feature wavelengths were selected, accounting for 32.42% of the spectrum. The fourteen oil samples in the prediction set were predicted using the established near-infrared full spectrum PLS model and the characteristic wavelength selected PLS model. Both models can achieve good prediction, and the standard deviation of prediction is 0.000 7 and 0.000 6, respectively. The PLS model selected by characteristic wavelength had the most robust prediction and the best performance index (R2P was 0.993 0, R2CV was 0.988 7, RMSECV and RMSEP were 3.140 1×10-4 and 2.419 0×10-4, RPD was 11.988 4). Compared with the full-spectrum model, the PLS model with characteristic wavelength selection can eliminate much useless information in the full spectrum, predict the water content of engine oil most robustly and have the best performance index so that the performance of the model has been significantly improved. The prediction set of oil samples was verified according to the established full-spectrum PLS model after OSC pretreatment and the PLS model for characteristic wavelength selection. The prediction effect of the PLS model after characteristic wavelength selection was good, and the predicted value of each oil sample was closer to the measured value. It indicates that the PLS model established after characteristic wavelength selection does not reduce the accuracy and prediction ability of the model, but eliminates the information of unrelated variables, making the model more generalized. Therefore, the near-infrared spectroscopy technology has good accuracy and reliability in detecting moisture in engine oil, which provides a feasible solution for engine condition monitoring.

The positive polarity Marx nanosecond pulse power supply was designed independently. The electron excitation temperature and electron density of argon discharge in a vacuum were measured and calculated by emission spectroscopy under different discharge frequencies and voltage amplitudes. Through the double line method to choose the appropriate Ar atomic spectrum, electron excitation temperature between 1 550～3 400 K in the regular polarity pulse power voltage source, and under a specific voltage, electron excitation temperature with the increase of power frequency and rise, in power frequency, electron excitation temperature increase with the increase of the supply voltage as well. The electron density in vacuum volumetric dielectric barrier discharge is measured and calculated according to the Stark broadening principle. The order of magnitude of electron density can reach 1013 m-3. When the power supply voltage is constant, the electron density shows a rising trend with the increase in voltage frequency. When the power supply frequency is constant, the electron density also gradually increases with the increase of power supply voltage.

In order to improve the accuracy of the land cover classification model based on near-infrared spectroscopy, the soil near-infrared spectrum data released by Eurostat was adopted as the research object in this study, and the land cover classification model based on short-time Fourier transform (STFT) preprocessing method and different convolution scale fusion are studied to achieve the rapid identification of cultivated land, forest land and grassland. In order to meet the requirements of two-dimensional convolution, 4200 wavelength points in the 400~2 500 nm band of the one-dimensional spectrum were transformed into two-dimensional images by SFTF, so the spectrum information about spectral data was extractedfor the later modeling. The samples were randomly divided into the training set, validation set and test set according to the ratio of 6∶2∶2. Two kinds of convolution neural network(CNN) models, including the single-size kernel CNN and themulti-size kernel fusion CNN, were established to classify the land cover. The CNN models adopted the ReLU activation function, batch normalization (BN), and dropout methods to prevent the model’s gradient disappearance. The Early Stopping method trains the network to prevent the model from overfitting. Firstly, this paper discussed the influence of different STFT window lengths (64, 100 and 128) and convolution kernel sizes (3×3, 5×5 and 7×7) on the classification effect of the different models. The experimental results show that the overall classification accuracy of all model shit a high point when the STFT window length was 100 and the window overlap length was 50%. The model’s classification accuracy decreases with the convolution kernel size. That is, the accuracy of the model with a smaller convolution kernel size is relatively higher. The overall classification accuracy of the CNN model with a convolution kernel size of 3×3 reaches 78.76%, which is higher than that of the CNN model with a convolution kernel size of 5×5 and 7×7. The CNN models with different convolution kernel sizes have good classification results for a certain land cover type.That is, the model with 3×3 convolution size has the best classification effect for cultivated land, 5×5 convolution size has the best classification effect for woodland, 7×7 convolution size has the best classification effect for grassland. Secondly, a fusion-CNN model based on the hybrid of multi-size convolution kernels is proposed. The model integrates the advantages of different size convolution kernels. The classification accuracy of the fusion-CNN for three land cover types has been improved to varying degrees, with an overall classification accuracy of 84.39%. The Fusion-CNN model overcomes the shortcomings of the single size convolution kernel CNN model, such as a long selection period for appropriate convolution kernel size and cumbersome parameter adjustment steps, and can simplify and speed up the modeling process. Fusion-CNN convolution fusion network can more effectively extract the internal characteristic information on soil near-infrared spectroscopy to obtain high and stable land cover classification accuracy.

The residues of dimethoate pesticides in the environment and agricultural products pose a great threat to human health and the ecological environment. Therefore, it is urgent to establish an efficient, simple and inexpensive method for detecting dimethoate. This study used a simple method to prepare N-doped carbon quantum dots (N-CQDs) via hydrothermal treatment of xylose and ammonium bicarbonate. The excitation spectrum of N-CQDs exhibited dual fluorescence centers at 238 and 330 nm, while both of their emission centers were located at around 402 nm. The fluorescence center obtained at the excitation wavelength of 238 was quickly quenched by dimethoate within 1 min, while that obtained at 330 nm showed little change. On this basis, a dual fluorescence center-based ratiometric fluorescent probe for dimethoate detection was constructed. Under the optimal reaction conditions, the linear ranges for dimethoate detection were 2~100 and 100~180 ng·mL-1, with a limit of detection of 0.67 ng·mL-1. Common ions and pesticides showed little effect for dimethoate detection, suggesting a good selectivity of the constructed probe without needing specific enzyme labeling. Moreover, the as-constructed probe was applied to detect the dimethoate residual in the actual pitaya samples, and the results were compared with the standard GC-MS method. The recoveries of the established probe were between 92.55% and 102.24%, with RSD lower than 3.62%. The recovery of the GC-MS method was between 84.22% and 100.64%, with RSD lower than 10.95%. Results reveal that the established probe shows higher accuracy and precision than the standard GC-MS method, the result is satisfactory.

The purpose of this paper is to promote the use of additives in purple clay products for food contact in line with national standards. Through instrumental analysis, this paper has established a non-destructive testing method for toxic barium salt added purple clay products, so that the industry can effectively use this technology for purple clay production quality management and material safety assessment. The method is to directly test the samples by energy dispersive X-ray fluorescence spectrometry, establish a set of standard samples with components to complete the function relationship between photon number and content, and carry out a database of comparison and the precision analysis by a large number of purple clay raw ore, mud raw material and purple clay products in various periods. The process is to analyze the consistency of repeated tests under the selected instruments and operating conditions with RSD in 1.75%～4.83%. The relative standard deviations (RSD) of homogeneity of standard samples are 3.10%～5.59%. The relative standard deviations of 10 purple clay samples are 1.23%～12.30%. The test of the specimens with thickness ranging from 1.9 to 14.1 mm shows the change rate between the test value and the real value is -6.94% to 78.44%. The average content of barium in 109 samples of purple clay ore is 0.029 5%. Tests are carried out on three kinds of clay materials without barium carbonate and two kinds of purple clay materials with barium carbonate of about 3‰ and the results show mud materials with barium carbonate, and the test value of barium is 0.380%～0.398%. In the mud without barium carbonate, the test value of barium is 0.016 2% to 0.046 1%. The barium content of 11 purple clay ware samples from 1979 and later is 0.238 8% to 0.387 7%. The actual analysis data of 1 085 purple clay samples show that barium content in purple clay products increased gradually after 1979. The conclusion is that this method has the characteristics of non-destructive, high efficiency, high speed, high accuracy, good consistency of repeated tests and easy operation, and it is more suitable for the determination of barium carbonate additive in food contact violet purple sand compared with the chemical test method. At the same time, through a large number of experimental data, it is found that the beginning of the artificial addition of barium carbonate was about 1979. Since then, the content of barium element in purple clay products has shown a rising trend, indicating that in the absence of effective measurement methods and technical supervision, the behavior of adding barium carbonate could not be substantially controlled.

Pork is the main meat consumption product in China. Its freshness is closely related to the health of residents. At present, the most common detection methods for meat quality include sensory testing, physical and chemical testing, and microbiological testing, but sensory detection is less reliable and comparable. Physical and chemical testing and microbiological testing have many problems, such as time-consuming, complicated operation and destroying samples, thus establishing a fast and nondestructive detection method has great significance. Raman spectroscopy is fast and nondestructive as a detection technology. Moreover, portable Raman spectroscopy provides a new way for food spot detection and is expected to achieve rapid real-time mass detection in the processing industry. At present, there is no study on the rapid detection of physical and chemical indexes of pork freshness by Raman spectroscopy. Therefore, a portable Raman spectrometer was used in this study to detect the freshness of cold storage lean pork. Collecting the Raman spectroscopy of samples with time and monitoring the corresponding freshness index, including total volatile base nitrogen (TVB-N), pH, L*, a*, and b*. Raman spectra were pre-processed by standard normal variable transformation（SNV）, curve smoothing（SG）, normalize（NL）, multiple scattering correction（MSC）, baseline（BL）, and Detrending（DFA）. Partial least squares regression (PLSR) was used to establish a quantitative prediction model of pork freshness indicators based on full displacements of Raman spectroscopy. The results indicated that the PLSR model based on the Raman spectrum had a good performance predicting pork freshness. The optimal model for TVB-N and pH was SNV-PLSR, and the correlation coefficient was 0.948 and 0.886, respectively. The optimal models for color L*, color a*, and color b* were SNV-PLSR, DFA-PLSR, and MSC-PLSR, respectively. The correlation coefficients were 0.827, 0.858 and 0.900, respectively. The regression coefficient method (RC) was used to screen the optimal spectral bands of each index model, and the PLSR model of the optimal spectral bands of each index was established. The results showed that the TVB-N and pH models could be simplified, and only 20% of the spectral bands can achieve a good prediction effect. TheRP of the TVB-N model and pH model were 0.933 and 0.880, respectively. Raman spectroscopy providing us with a spot detection method shows great potential in rapidly detecting pork freshness, especially in predicting TVB-N content.

The balance of human energy intake and energy consumption is one of the standards for maintaining health. Unbalanced intake may cause consequences such as cell damage and obesity. The estimation of energy intake is of great significance to human health management. The current method of assessing energy intake is mainly through dietary review, but it is time-consuming because of increasing the burden of the person to be evaluated. Therefore, developing a simple and fast way to estimate energy intake is urgent. After energy intake, metabolites generated by digestion and metabolism are excreted as waste. Wastes such as urine, etc., contain many chemical species, which can systematically reflect the dietary status and disease processes. This research aims to establish a classification model based on SERS techniques, which is highly sensitive, non-destructive, and identifiable molecular fingerpring. Peak statistics, and unsupervised and supervised clustering algorithms are utilized to analyze SERS data collected from volunteer groups of energy intake with 1 500, 2 030, 2 700 kcal·day-1. Since there is a certain amount of overlapping of Raman peaks in many organic molecules, it is difficult to analyze and assign SERS peaks. This study adopts a comparative analysis of an unsupervised PCA and a supervised OPLS-DA algorithm for classification and prediction. It was found that the scattering distribution of different categories in PCA has a large extent, so the model shows poor categorization. After correcting the background by first-order derivative difference, the scatter map presents the classified trend. The OPLS-DA algorithm can decompose the X matrix information into the Y-related and unrelated two components by presetting the Y’s label to achieve good classification after orthogonal signal correction processing. The results show that the OPLS-DA algorithm can be well-classified for three or each two different energy intake levels. Both the specificity and accuracy of the ROC analysis have reached 100%. The permutation test of 200 times also illustrates the model with good accuracy and predictability. The results indicate that the levels of human energy intake can be directly estimated by analyzing the SERS signal of the urine. This method can rapidly analyse urine in 2 minutes with simple manipulation and accurate discriminant results, which shows great potential in medical health applications.

In the past two decades, terahertz spectroscopy has been proven to be a powerful tool potentially applied in biomedical research, industrial quality inspection, national defence, and public security. However, the dependence of traditional Terahertz time-domain spectroscopy (THz-TDS) on femtosecond lasers seriously limits the application of THz technology in practical industrial fields. Recently, terahertz frequency-domain spectroscopy (THz-FDS) based on the photo-mixing mechanism of two commercially available mid-infrared lasers has been widely studied as an alternative due to its better compactivity, lower cost, and outperformed frequency resolution. However, the data processing algorithm for THz-FDS is not comparable to the traditional THz-TDS. Previous spectroscopic studies using THz-FDS generally focus on the absorption spectra of target material extracted from the signal amplitude. Nevertheless, the phase information of the THz wave provided by the coherent detection mechanism has not been well explored. Hence, various valuable parameters such as refractive index, dielectric constant, and polarizability closely related to the phase information cannot be accurately obtained. In this paper, the relationship between the phase information of the THz wave and the periodic oscillation of the FDS original photocurrent data is well explained, accompanied by presenting actual measurement data. A theoretical model is then constructed to calculate the refractive index spectra from the original photocurrent data oscillation period. The primary reason the previous DC-basepoint algorithm could not output the correct refractive index over the low-frequency band is then pointed out. We propose an improved dual-basepoints index algorithm capable of extracting accurate refractive index spectra over the whole spectrum. In order to verify the reliability of our proposed method, polytetra fluoroethylene, a polymer widely used in the THz band, was selected as the characterization object. Multiple samples with different thicknesses were prepared. The refractive index of all those samples was 1.456±0.006. The measurement uncertainty of the refractive index between different samples is only 0.5%, and the average value is consistent with the previously reported value, which thoroughly verifies the reproducibility of this method. In addition, one sample was selected to investigate the influence of the experimental parameter setting on the refractive index measurement results. The original photocurrent data were collected on the same sample with THZ-FDS using various integration times and frequency step-size combinations. The results show that the experimental parameter setting does not significantly affect the refractive index measurement results, which verifies the algorithm’s robustness. The refractive index measurement method proposed in this paper explored the characterization parameters of the terahertz frequency domain spectral system. Therefore, this method is of great significance to the practical application and development of terahertz spectral technology.

In order to study the three textiles P1, J1 and J2 on the bronze blister and the bronze mirror from the early Western Zhou Dynasty unearthed from the shangmiao cemeteries of Xitou site, Xunyi County, Shaanxi Province, the structures of textiles were analyzed by Infrared Spectrometer, Three-dimensional Video Microscope System and Scanning Electron Microscope. It is found that there are obvious peaks near 1 434 cm-1 due to the bending vibration of CH2 in cellulose and lignin; 1 080 and 1 033 cm-1, due to the stretching vibration of C-O ether bond in glucose ring in cellulose; J1 and J2 show a peak near the wave number of 898 cm-1, which is the characteristic absorption vibration band of D-glucoside bond. A peak appears in the region of 1 645~1 610 cm-1, which is the overlapping absorption peak of conjugated carbonyl and the CC stretching vibration of lignin. These characteristic peaks are consistent with the infrared absorption spectra of hemp fibers, so it can be inferred that the three textile traces are hemp textiles. The observation shows that the surface of these fibers is rough, their radial shape is cylindrical and flat, and the cross-section is triangular and oval, similar to hemp. It is suggested that they are hemp products. The textile density on the bronze blister is 23×25 pieces·cm-2, the fabric of bronze mirror is loose, which are 13×6 pieces·cm-2 and 13 pieces·cm-2×14 pieces·cm-2. It is found that there are significant differences in the warp and weft diameter data of these three textiles, which should be the result of different use of warp and weft. According to the above analysis results, due to the differences in tomb specification, the fabric on the bronze blister is finer than that on the mirror, and the weaving density of the two kinds of fabrics on the mirror is quite different. According to the location of their attachment, it can be inferred that J2 is for wrapping, J1 is for padding during burial, and the mirror is placed face down during burial. During the Shang and Zhou dynasties, hemp crops were widely used by the people who belonged to the cemetery, and the textile level was mature; At the same time, it is precise because of the attachment of these textiles to bronze that the antibacterial effect of copper ions and the custom of wrapping bronzes with textiles enable them to be preserved so far.

To achieve the best living effect, the ancients used their accumulated experience to decorate and transform the interior of houses. This paper analyzes the white surface of F33 and F50 at the Hamin-Mangha site in Tongliao, eastern Inner Mongolia. It is found that the reason for the hard and white ground of the site is the use of white ash as the ground decoration layer; The change of infrared spectrum peak in absorption intensity is used to judge the information of atomic disorder in calcium carbonate, to judge its manufacturing process. The results show that the ratio of calcite ν2/ν4 in the white bedding material of the site is about 5, which is much higher than that of natural limestone, which reflects the difference in the disorder of calcite crystal structure between natural limestone and artificially prepared lime mortar. It proves that the white ash layer of the site is artificially fired, indicating that our ancient ancestors began to use artificially fired lime as decorative building materials a long time ago.

The use of ochre is one of the signs of modern people’s behavior, in addition to stone artifacts that can reflect modern people’s cognitive ability and other deep-seated spirituals. In order to better understand the scientific information of ochre in archaeological sites from China regional, we used Raman, polarized light microscopy, and EDX to analyze the ochre in two archaeological sites of the Late Paleolithic period (Zhongshanshan site and Yahuai Cave site). The AMS dating results of the Zhongshan site show that the age was about 14 000 years ago, and the characteristics of the stone products of the two sites belong to the late Paleolithic period. Raman analysis showed that the peak values of all samples were consistent with ochre, the peak value was mainly 221 s, 290 s, 608 s and 662 m. Polarized light microscopic analysis showed that all samples were red ochre (oxide red). The sample is dark red under single polarizing light, with rounded edges. It is completely extinct under orthogonal slice light and has a large refractive index. The results showed that all ochre samples were mainly iron red, and no red pigments, such as cinnabar, were found. EDX analysis showed that all mineral samples contained Fe2O3. The content of Fe2O3 and SiO2 in chemical element composition is high, while the content of Al2O3, CaO, TiO2, TiO2 is low. Through the analysis of the SiO2 content of samples, it is shown that the SiO2 content is positively correlated with their hardness. Generally speaking, the main components of ochre found at the three sites were iron red pigments, and no other red pigments were found. This study is the first Raman and polarized microscopic analysis of the late Paleolithic ochre in China. The use of ochre by ancient humans in southern China can be further clearly dated to about 14 000 years ago (the Zhongshan site). The analysis of ochre materials from the two sites provides a basis for the scientific analysis data of late Paleolithic ochre in China and confirms the reliability of naked eye judgment of ochre materials. They were providing important potential value for the careful study and utilization of modern human behavior and ochre, providing important research materials and scientific data for the international study of ochre in China.

The gem Cr-grossular (also called tsavorite) from Canada often has a special green color zoning. The black or dark green core is often surrounded by an emerald green edge, which looks like a “frogspawn”. To study the gemological and spectral characteristics of the Canadian tsavorite with “frogspawn” color zoning, the samples from this area were systematically studied by standard gemological methods, as well as LA-ICP-MS, UV-Vis-NIR spectrum and Raman spectrum, which reveal the causes of the color zoning and the changes of composition and spectroscopy in different color zones. The chemical composition analysis shows that the Canadian tsavorite is mainly composed of grossular (Core: Gro>55.64 mol%; rim: Gro>83.90 mol%), but the content of Cr2O3 varies significantly in different color zones. The deep-colored core in the center shows high uvarovite content (Uva average 21.49 mol%). The black core also has high Ti content (TiO2>1.9 Wt%). In addition, the samples also contain a small amount of Fe and a trace amount of V, Mg and Mn. UV-Vis-NIR spectra show that Cr3+ is the main ion causing green color and Fe mainly induces yellow color in samples. The absorption bands at about 435nm in the blue violet region and 603 nm in the red region are mainly due to Cr3+. The double peak at about 700 nm of Cr3+ can be used as the characteristic peak to distinguish the presence of V3+. Fe3+ causes an absorption peak at about 370 nm and contributes to the absorption band at 435 nm in the blue-violet region. Fe2+ causes the broad and weak absorption band at 1 200 nm. By analyzing the ratio of Fe2O3 to Cr2O3+V2O3, color tones of grossular can be effectively distinguished. When the ratio is less than or near 1.61, it often shows pure emerald green; When the ratio is near 2.71, it often shows yellowish-green; When the ratio is around 4.38, it often shows greenish-yellow. Raman spectrum analysis shows that the sample’s yellowish-green to emerald green regions are typical grossular. The Raman peaks between 800 and 1 100 cm-1 are mainly caused by the stretching vibration of ［SiO4］ tetrahedron; The Raman peaks between 400 and 700 cm-1 are mainly caused by the bending vibration of ［SiO4］ tetrahedron; The Raman peaks below 400 cm-1 are mainly caused by lattice vibration. Based on the comparison between the Raman spectra of color regions of the sample and natural uvarovite, the increase of Cr3+ content in the mineral structureleads to regular Raman peak shift, as the testing point moves from the edge of the sample to the black core.

Zhaxun is usually divided into four grades. Zhaxun is of better quality with black color, heavy quality and fewer feces, according to the experience of traditional Tibetan medicine. Different grades have little difference in appearance, and it is more difficult to distinguish after boiling into pastes. Fourier transform infrared spectroscopy(FTIR) has the advantages of being fast, and nondestructive and has been widely used in medicinal material identification. They were exploring the feasibility of FTIR to identify Zhaxun of different grades with FTIR. 56 batches of medicinal materials and substitutes were collected and divided into four grades according to the proportion of feces and morphology. Each batch was processed into dry pastes according to the Tibetan medicine standard of six provinces. Different chemopmetric methods established models according to different samples in the range of 4 000～400 cm-1. Preconditioning methods contain Saitzky-Golay (S-G) smoothing, ordinate normalization, and second derivative transformation for preprocessing, and the main absorption regions of Zhaxun are 3 500～3 200, 3 000～2 800, 1 800～1 350, 1 350～900, 900～400 cm-1. There are great differences in the indices of the absorption peaks of substitutes. Only Zhaxun of grade Ⅰ and grade Ⅱ have absorption peaks near 1 779 cm-1. The intensity of Zhaxun of grade Ⅰ and grade Ⅱ near 1 768 cm-1 is significantly stronger than grade Ⅲ and substitutes. Only substitutes near 1 660 cm-1 have no absorption peak, and substitutes at 1 257 cm-1 have an absorption peak. These can be used as the identification of Zhaxun’s different grades. Changes absorption peaks in area ③ and area ⑦ are related to the traditional classification that the stronger peak intensity is related to quality. Principal Component Analysis (PCA) is difficult to distinguish Zhaxun of different grades. However, Partial Least Squares Discriminant Analysis (PLS-DA) can better distinguish medicinal materials of four grades and results of external verification show that the model can well distinguish Zhaxun of different grades according to the statistical chart of results of SIMCA. The hierarchical cluster method(HCA) can distinguish some batches of Zhaxun of grade Ⅲ and substitutes through SPSS21.0. FTIR combined with chemometrics provides a rapid method for the quality evaluation of Zhaxun that can quickly identify the grades of Zhaxun and distinguish the substitutes.

In order to solve the premature rupture of degradable mulch film due to degradation, the stability and degradation of PPC plastic film were studied under different mulching methods. The six mulching methods in the field were as follow: single-layer black PPC plastic film (SB), single-layer white PPC plastic film (SW), double black PPC plastic film (DB), double-white PPC plastic film (DW), double-layer mulch which upper layer was black and the lower layer was white (DBW), the double layer mulch which upper layer was white and the lower layer was a black (DWB). The infrared spectrum characteristics, mechanical properties and molecular weight of film periods were determined at a different time to judge the degradation of the plastic film under different treatments. The results showed that plastic film was exposed to conditions for a long time, and the toughness of plastic film has become worse. The main chain in the molecular structure was broken, furthermore, the relative intensity of the absorption peak of the C-H bond and C-O bond decreases with the increase of time. Due to the degradation of mulch film under the action of water molecules and microorganisms, the relative intensity of theCO stretching vibration peak and C-O-C antisymmetric absorption peak increased in different degrees. Due to the adhesion of molecular chain activity under sunlight, the stability of double-layer plastic film was higher than that of single-layer plastic film. Therefore, the decreasing trend of peak intensity for black and white double-layer plastic film DWB and DBW was more stable than that for single mulching film. The degradation degree was more slowly. The molecular weight and mechanical properties of each treatment showed a downward trend. The molecular weight. Tensile strength and elongation at the break of the double-layer plastic film were significantly higher than that of single-layer plastic film treatment. The stability of DWB and DBW treatments was stronger than that for the single mulching film within 30~150 days. The elongation at break and tensile strength of DWB and DBW treatments were 117.66%～120.39%, 151.69%～175.20% and 18.28～13.95, 15.35～9. 81 MPa at 150 days. The average molecular weights and number average molecular weights of DWB and DBW treatments were 72 663, 66 555 and 62 416, 66 555 g·mol-1. Comprehensive analysis showed that black and white double-layer plastic film could slowly cause the rupture of the plastic film caused by the degradation of PPC plastic film, Which could improve the time for the performance of water conservation and soil moisture conservation and better take advantage of the double-layer mulch film. After the optimization of the coverage methods of PPC plastic film, PPC plastic film could maintain its stability during the crop growth period, and it also could be degraded like a single layer mulch film after the crop growth period, which provided a theoretical basis for the further use of PPC film in agricultural production.

Continuous cropping obstacle is the greatest difficulty in Rehmannia glutinosa planting. It must be replanted after 8～10 years. However, the mechanism of the continuous cropping obstacles is not clear yet. An analytical method based on Fourier transform infrared spectroscopy (FTIR-ATR) and Two-Dimensional Infrared Correlation Spectroscopy (2D-IR) is proposed. Collected from the main crop (YA-One), second crop (YA-Two), 10-year-old interval (YA-Ten) and blank (CK) of the spectra of four kinds of Rehmannia glutinos soil samples. Norris noise filtering and smoothing, second derivative and two-dimensional correlation spectrum analysis were performed to extract characteristic spectra, analyzed the fingerprint of compounds in the soil. The results showed that the peaks of different continuous cropping soils were similar, and the main characteristic absorption peaks of spectra were around 3 621, 3 439, 2 932, 2 860, 2 513, 1 798, 1 636, 1 433, 1 029, 887, 783, 695, 546 and 469 cm-1. Compared with CK, the absorption intensity of YA-Two is the strongest, and these absorption peaks mostly represent the functional groups such as hydroxy -OH, carbonyl CO, C-O, and substituted absorption of the benzene ring in phenolic acids, indicating the accumulation of phenolic acids in Rehmannia glutinos in continuous cropping. The reverse stretching vibration and symmetric stretching vibration of saturated methylene CH2 in the glycosides were observed At t 2 925 and 2 857 cm-1. The absorption intensity of Ya-Two was the smallest, and Ya-Ten was the largest. The results indicated that the glucoside components decreased in continuous cropping soil. The position, number and strength of automatic peak and cross peak were different in the two-dimensional infrared correlation spectra of the soil of Rehmannia glutinos in bands 920～1 750, 1 490～1 710 and 1 500～1 548 cm-1, which clearly showed the difference of functional groups. The results showed that the rapid detection of organic compounds in soil could be achieved by using Fourier transform infrared spectroscopy (FTIR) and two-dimensional infrared correlation spectroscopy (TD-IR), which provided a theoretical basis for studying Rehmannia glutinos continuous cropping obstacles.

As a cactus plant, the dragonfruit plant is leafless and mainly relies on succulent stems for physiological functions such as photosynthesis and transpiration. There are obvious differences in tissue structure and morphology between the succulent stems of dragon fruit and the leaves of common green leaves, and there is also obvious differences in plant canopy structure, which will directly affect the spectral characteristics of plant canopy. Furthermore, it affects the monitoring of photosynthetic pigments based on spectral technology. In order to explore the method to improve the estimation accuracy of chlorophyll content in the stem and branch of dragon fruit, taking the planting base of Yanshan dragon fruit in Longping Town, Luodian County, Guizhou Province, as the experimental area, the tissues of the stem branch and branch of dragon fruit were collected and determined by ethanol extraction. The chlorophyll content of the tissue was determined by ethanol extraction. Then the spectral data were processed and analyzed by traditional mathematical transform, continuous wavelet transform, discrete wavelet transform and discrete wavelet-differential transform respectively. The correlation analysis algorithm was used to extract and screen the sensitive feature bands. Finally, the partial least square algorithm is selected to construct the estimation model of chlorophyll content in the stem and branch of dragon fruit. The conclusions were as follows: (1) under the discrete wavelet-differential transform algorithm. The peaks and valleys of high-frequency and low-frequency information appear alternately, and the segments of available information have strong stability. With the increase of scale, the amplitude of the curve increases, and the frequency decreases. (2) the methods of differential transform, continuous wavelet transform, discrete wavelet transform and discrete wavelet-differential transform in mathematical transform can improve the sensitivity of spectrum to chlorophyll content in stem and branch of dragon fruit, among which the method of discrete wavelet-differential transform was the best, and the determination coefficient of the spectrum and chlorophyll content in stem and branch of dragon fruit could reach 0.565 (located at H1 decomposition scale 737.5 nm). (3) discrete wavelet-differential transform can effectively improve the ability of the spectrum to estimate chlorophyll content in stems and branches of dragon fruit, and the estimation model based on the H2 scale of discrete wavelet-differential transform was the optimal model. This study analysed the effects of four kinds of spectral processing algorithms on improving the sensitivity and estimation ability of spectrum to chlorophyll content in stem and branch of dragon fruit. The results show that the discrete wavelet-differential transform algorithm proposed in this paper can effectively improve the ability of the spectrum to estimate chlorophyll content in the stem and branch of dragon fruit, which provides basic technical support for the non-destructive diagnosis of chlorophyll content in stem and branch of dragon fruit.

Catalpa bungei has the advantages of straight texture, excellent material, and versatility characteristics and it is a precious wood species unique to China. Bending property, an important mechanical property of wood, research on its rapid determination method can provide a scientific basis for genetic improvement, processing, and utilization of Catalpa wood. The “Luoqiu 1”, “Luoqiu 4” and “Tianqiu 2” of the new C. bungei clones were used as the experiment materials. The modulus of rupture (MOR) and modulus of elasticity (MOE) was determined according to the national standard bending property test method. Near-infrared spectroscopy (NIRs) combined with the partial least squares (PLS) method was used to predict the bending properties of three newly bred C. bungei clones. The best modeling method based on different wood sections, pretreatment methods, and the number of sampling points were explored. The results indicated that the maximum Rp and RDP of the MOR model based on the average spectra of two sections were 0.843 and 1.88, and the maximum Rp and RDP of the MOE prediction model were 0.846 and 1.88. In descending order of accuracy of MOR models based on average sections, pretreatments were: MSC+S-G, 2ndDer+S-G, and 1stDer+S-G. In descending order of accuracy of MOE models based on average sections, pretreatments were: MSC+S-G, 1stDer+S-G, and 2ndDer+S-G. In conclusion, NIRs can be used to predict the MOR and MOE of valuable C. bungei wood. Models established with different sections, pretreatments, and the number of sampling points have certain differences in modeling results. This paper obtained the best modeling methods for the MOR and MOE of C. bungei wood. NIR models of MOR and MOE based on average spectra of radial and tangential sections were the best. MSC+S-G was the most suitable pretreatment method for the bending properties of C. bungei wood. The five-point sampling method has the highest model accuracy. The number of sampling points can be reduced to quickly estimate the bending property of a large number of samples. It is possible to collect only one spectra point in the middle part, the loading position, to reduce the workload of collecting spectra and improve the efficiency of rapid evaluation of the bending property of C. bungei wood.

NIR spectrometer has been widely used in many fields such as aerospace, biomedicine, environmental testing, food safety etc. High performance, miniaturization, and low-cost are the main bottlenecks of developing micro-detection equipment based on near-infrared (NIR) continuous spectrum analysis, which is the leading research direction of the spectrometer. This paper proposes a miniature NIR spectrometer system structure based on micro-optical electro-mechanical system (MOEMS) integrated scanning grating micromirror and improved asymmetric Czerny-Turner(C-T) optical structure, analyzes the working principle of the spectrometer system and integrated scanning grating micromirror, and determines the maximum scanning angle of the integrated scanning grating micromirror based on grating-related parameters and spectrometer performance index requirements. The aberration of the improved asymmetric C-T initial optical structure is analyzed, the simulation and optimization design of the spectrometer optical system is completed based on the ZEMAX optical design platform, and the system’s key parameters are determined. The influence of the plano-convex cylindrical lens on the performance parameters, such as resolution and detection sensitivity of the improved asymmetric C-T optical structure system, was simulated and analyzed. Based on the simulation and optimization results, the mechanical structure design, processing and mounting of the miniature NIR spectrometer were completed, and the experimental platform was built to complete the testing of the relevant performance parameters of the spectrometer. The results show that the miniature NIR spectrometer based on MOEMS integrated scanning grating micro-mirror and improved asymmetric C-T optical structure was designed. This paper uses the MOEMS integrated scanning grating micro-mirror with the resonance frequency of 677.1 Hz developed by Chongqing University to realize simultaneous scanning and spectroscopy and can complete a wavelength range of 800~1 800 nm in 0.8 ms. The spectral accuracy is not significantly different from that of foreign brand spectrometers, the overall spectral resolution (FWHM) is ≤11 nm, and the wavelength stability is ≤±1 nm. The optical structure design based on the plano-convex lens can increase the detection output light intensity value by more than 15%, effectively improving spectral measurement sensitivity. At the same time, the spot size is smaller after the second focus of the plano-convex lens so that the single tube detector with a small sensing area and large frequency can be used to achieve spectral detection, which can reduce the system costs, suppress external optical noise, and meet the scanning grating type spectrometers with high scanning frequency. It can reduce system cost, suppress external optical noise, and meet the demand for spectral resolution of scanning grating spectrometer with high scanning frequency. Therefore, the NIR spectrometer based on MOEMS integrated scanning grating micromirror and improved asymmetric C-T structure proposed in this paper meet the development needs of high-performance, miniaturization and low-cost spectrometer.

Due to the uneven market of rattan pepper oil, based on near-infrared spectroscopy technology, rattan pepper oil is the research object, and the research on the adulteration detection of rattan pepper oil is carried out. First, pure rattan pepper oil was used as the base oil, and the adulterated soybean oil, corn oil, and sunflower oil were prepared in proportion to obtain oil samples. The near-infrared spectroscopy was used to collect the spectral data of the oil samples to obtain the adulterated near-infrared spectral data of rattan pepper oil. The spectral data are normalized and preprocessed by Standard Normal Variation (SNV) and MultivariativeScatter Correction (MSC). And then, the feature data is processed by Competitive Adaptive Reweighting Sampling (CARS) and SuccessiveProjection Algorithm (SPA). Extraction, combining different preprocessing algorithms and feature data extraction algorithms, and establishing a prediction model of vine pepper oil adulteration through Support Vector Machine regression (SVR). The results show that the coefficient of determination (R2) of the calibration set and prediction set of the MSC-CARS-SVR model is the highest, the calibration set R2 reaches 0.756 1, and the prediction set R2 reaches 0.705 2; the root mean square error (RMSE) is the smallest, and the calibration set RMSE reaches 0.743, The prediction set RMSE reaches 0.794. In order to improve the accuracy of the model, the Whale Optimization Algorithm (WOA) and the Improved Whale Optimization Algorithm (BAS-WOA) are used to optimize the SVR model. The left and right beards are moved forward, and the objective function after the advance is calculated. If the objective function is better than the current optimal whale value, the position of the whale is replaced by the position of the beetle after the move forward, thereby realizing the improvement of the beetle operator on the whale algorithm. When WOA optimizes the SVR model, compared with the MSC-CARS-WOA-SVR model with the highest accuracy, the R2 of the calibration set can reach 0.859 1, and the R2 of the prediction set can reach 0.821 6; the RMSE of the calibration set is reduced to 0.374, and the RMSE of the prediction set is reduced to 0.495. Compared with the traditional SVR model, the accuracy and performance of the SVR model are significantly improved. When BAS-WOA optimizes the SVR model, the MSC-CARS-BAS-WOA-SVR model has the highest accuracy. The calibration set R2 is as high as 0.955 1, and the prediction set R2 is as high as 0.943 9; the calibration set RMSE is reduced to 0.054, and the prediction set RMSE is reduced to 0.081. Compared with the WOA optimization algorithm, the model accuracy and performance of the BAS-WOA optimization have been further improved. The model prediction set R2 is increased from 0.821 6 to 0.943 9, and the prediction set RMSE is reduced from 0.495 to 0.081. Whale Optimization Algorithm easily falls into local extremum and convergence rate problems when optimizing the SVR model. The improved Whale Optimization Algorithm uses the left and right baleen search of the beetle algorithm to improve the lack of the Whale Optimization Algorithm, thereby improving the global optimization ability of the algorithm. The research shows that near-infrared spectroscopy technology combined with an intelligent optimization algorithms can effectively identify the adulteration of vine pepper oil, which provides a reference for the research on the adulteration of vine pepper oil.

Pleurotus ostreatus is popular with consumers because of its delicious taste and rich nutrition. Pleurotus ostreatus is widely cultivated in China, and its producing areas are scattered. The differences in climate conditions, cultivation matrix and cultivation mode of each producing area make the Pleurotus ostreatus produced in different producing areas different in taste and nutritional value. In order to standardize the market management of Pleurotus ostreatus products and create regional characteristics of Pleurotus ostreatus brands, with the help of the characteristics of non-pollution, high efficiency and low cost of mid-infrared spectroscopy, this paper broke through the limitations of chemical analysis and biological identification methods at present, and put forward a method of identifying Pleurotus ostreatus from different producing areas by mid-infrared spectroscopy combined with machine learning. The infrared spectrum data of fruiting bodies of Pleurotus ostreatus from 10 different producing areas were collected, and 60 samples were collected from each area. The analysis of the spectral data showed that the correlation of the infrared spectra showed significant differences in the band 530～1 660 cm-1. At the same time, based on the K-S method, the samples were divided according to the ratio of the training set to test set of 7∶3, 420 training sets and 180 test sets were obtained. Multiplicative scatters correction (MSC), standard normal variable transformation (SNV), Smoothing(SG), first derivative (FD), second derivative (SD) and other preprocessing methods were used to optimize the spectrum and remove the noise. In addition, it combined with a support vector machine (SVM) for preliminary modeling comparison. It was concluded that the difference in spectral data after MSC pretreatment was the largest, and the recognition performance of the prediction set was the best at 84.44%. The MSC spectral data is normalized in 0-1, and principal component analysis (PCA) was used to reduce the dimension. The first seven principal components, which satisfy the cumulative contribution rate of principal components in the training set ≥85% and the variance percentage of principal components ≥1%, were selected as input variables for modeling identification comparison with support vector machine (SVM), random forest (RF) and extreme learning machine (ELM). The experimental results showed that the SVM model had the best recognition effect in identifying Pleurotus ostreatus models from different producing areas, and the recognition rate of the training set and test set was 100%. The recognition rate of the RF model training set was 100%, and the recognition rate of the test set was slightly lower, 98.89%. Compared with other models, the recognition rate of the ELM model was poor, the recognition rate of the training set was 99.28%, and that of the test set was 98.33%. The recognition rates of the three models were all higher than 98%, indicating that the identification of Pleurotus ostreatus from different producing areas can be realized, quickly and at low cost using infrared spectroscopy combined with machine learning. This provided a method basis for the producing areas identification of Pleurotus ostreatus products and a reference for the identification of other kinds of edible fungi products’ producing areas.

Due to simplicity, rapidity and non-destructiveness, Raman spectroscopy is very suitable for mineral classification and identification. A Raman spectral model-fitting method does not need to build a reference spectral database and complex spectral matching, which is advantageous in mineral classification. However, there is a lack of comprehensive comparison of the existing model-fitting methods based on machine learning and deep learning since they use relatively single-learning models. To this end, this paper comprehensively evaluates the model-fitting classification methods of mineral Raman spectral using the RRUFF mineral Raman spectrum dataset. It compares the classification performance of four traditional machine learning methods of KNN, XGBoost, SVM, and RF, and three deep learning models of CNN, DNN, and RNN, as well as four data preprocessing methods and sample size on the classification effect. To improve the classification performance, we also propose a data preprocessing method of Raman spectral intensity curvature, which calculates the curvature of the baseline-corrected Raman spectral sequence intensity as a construction feature so that the model can extract the position of the spectra peaks more effectively. The experimental results showed that data preprocessing greatly improved the classification performance of machine learning models but had little effect on deep learning models. Additionally, the size of the sample is a key factor of the model performance. When the size is large, the deep learning models outperform the traditional machine learning models, whereas when the size is small, it is difficult for the deep learning models to exert their advantages, while the traditional machine learning models combined with data preprocessing work better.

To enhance the visibility of infrared and visible image fusion and overcome the problems of detail loss, insignificant target, and low contrast in infrared and visible image fusion results, a novel infrared and visible image fusion method based on two-scale decomposition and saliency extraction is proposed. Firstly, based on the theory of human visual perception, the source image is decomposed at different levels to avoid mixing high-frequency and low-frequency components to reduce the halo effect. In this paper, we use a two-scale decomposition method to decompose the source infrared and visible images and obtain the basic layer and detail layer, respectively, representing the image well and having good real-time performance. Then, a weighted average fusion rule based on a visual saliency map (VSM) is proposed to fuse basic layers, and the VSM method can extract the salient structures and targets in the source images. The VSM-based weighted average fusion rule is used to fuse the base layer, effectively avoiding the contrast loss caused by the direct use of the weighted average strategy and making the fused image perform better. The Kirsch operator is used to extract the source images separately to obtain the salient maps for the fusion of the detail layer. Then the VGG-19 network is applied to get the weight maps by extracting features from the salient maps and fusing them with the detail layer to obtain the fused detail layer. The Kirsch operator can quickly extract the image edges in eight directions, and the significant map will contain more edge information and less noise. The VGG-19 network can extract deeper feature information from the image, and the obtained weight map will have more helpful information. Finally, the fused basic and detail layer images are superimposed to get the final fusion result. Four sets of typical infrared and visible images are selected for testing and compared with six other current mainstream methods in the experimental part. The experimental results show that the method in this paper has the advantages of high contrast, prominent target, rich detail information and better retention of image edge features in terms of subjective quality. The objective metrics such as information entropy, mutual information, standard deviation, multiscale structural similarity measure and difference correlation sum also show relatively good results.

In view of the fragmented planting landscape and complex planting structure of Chinese farmland, achieving high accuracy identification of target crops is of great importance for subsequent crop yield estimation, grain policy adjustment and national food security guarantee. Based on Google Earth Engine (GEE) remote sensing data processing cloud platform, this study proposes a planting structure extraction method applicable to different fertility stages of winter wheat. The method adopts multi-temporal Sentinel-2 images covering key fertility stages of winter wheat in 2021 as the data source, and integrating multi-dimensional feature variables, including spectral band features, index features, texture features and topographic features. In this study, the GBDT (gradient boosting decision tree) classifier was employed to extract the planting area and spatial distribution information of winter wheat at different fertility stages at the field scale. The best fertility period for winter wheat identification was discussed. In addition, the crop recognition performance of different common classification models at the field scale was compared and analyzed. The experiments were conducted in Chengu Town, Henan Province, China, and the experimental results showed that the accuracy of planting area recognition was relatively high in the standing and jointing stage (3.11-4.10) of winter wheat, with an overall classification accuracy of 94.61% and a Kappa coefficient of 92.68%. The highest recognition accuracy was achieved in the heading and flowering stage (4.11-5.10), with an overall classification accuracy of 97.01% and a Kappa coefficient was 95.52%; however, the classification accuracy was low in grain-filling and milky stage (5.11-6.10), with an overall classification accuracy of 86.23% and a Kappa coefficient of 81.33%. The results showed that the GBDT classifier could effectively extract land cover information under field-scale conditions and achieve better feature classification recognition during winter wheat’s heading and flowering stage. In addition, this study compared GBDT with traditional classifiers such as Random Forest (RF), CART (classification and regression tree) and Naive Bayesian (NB). The results show that the GBDT classifier has the best performance in feature recognition, with an overall classification accuracy of 1.20 and 5.99 percentage points higher than the RF and CART classifiers, respectively, and a Kappa coefficient of 1.61 and 8.04 percentage points higher than the RF and CART classifiers, respectively. Moreover, the NB classifier has the worst recognition precision, with an overall classification accuracy and a Kappa coefficient of 84.43% and 78.69%, respectively. The results of this study can provide effective technical support for fine-grained crop extraction at the field scale.

Chlorophyll-a, the main pigment in phytoplankton, is an indicator of the degree of eutrophication in the water, so accurately obtaining and predicting the concentration of chlorophyll-a can provide a theoretical basis for protecting the marine environment. In the experiment, remote sensing images obtained by the moderate resolution imaging spectrometer were used as data sources, and images of chlorophyll-a concentration in the same water area were taken as the true relative value. The convolution neural network was used to establish the relationship model between remote sensing reflectance and chlorophyll-a concentration, and then the inversion of marine chlorophyll-a concentration was realized. The experimental procedure starts with pre-processing the global ocean reflectance data (band combinations of 412, 469, 488, 547, 667 nm) and chlorophyll-a concentration data in 2020 by multiplicative amplification and logarithmic transformation. Then, a convolutional neural network inversion model for the concentration of marine chlorophyll-a was constructed based on the data set of the water area at the boundary of the Pacific Ocean and the Indian Ocean in January 2020, which was divided into the training set and validation set. The coefficient of determination (R2), Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) were used as evaluation indicators to optimise the inverse model. Finally, the data of chlorophyll-a concentration from January to December 2020 were used as the test set to verify the inversion accuracy of the model. The results show that the proposed inversion model has an accuracy of R2=0.930, RMSE=0.130, MAE=0.102 and demonstrate that the inversion results of chlorophyll-a concentrations given by the model are in high agreement with ground truth values. It can be applied in inversion studies of global marine chlorophyll-a concentration based on remote sensing imagery.

Hyperspectral imaging technology is an advanced image acquisition technology which can not only obtain the spatial information of ground objects but also the spectral information of ground objects to obtain the three-dimensional image data of “atlas integration”. Its spectral resolution is very high, and its curve is nearly continuous. It can effectively detect ground objects that cannot be detected in multispectral imaging technology and has been widely used in target detection, ground object classification and image compression. Among them, ground object classification can distinguish different types of ground objects in images, and the classification results are the basic data of thematic mapping, which has achieved good results in military, agricultural, geological and other fields.Ground object classification refers to assigning category labels to pixels in images, that is, assigning the same labels to similar ground objects and different labels to different kinds of ground objects. According to whether the target spectral information has been obtained before classification, ground object classification can be divided into supervised classification, semi-supervised classification and unsupervised classification. However, the spectrum of ground objects is greatly affected by imaging conditions, especially for land-based imaging. The spectrum of ground objects will change to a certain extent under different imaging conditions and is no longer strictly unique so that it cannot be accurately classified according to the spectral data of ground objects under unknown imaging conditions. However, the scattering ratio coefficient of the same ground object is unique, has nothing to do with imaging conditions or detection direction, is not affected by the characteristics of bidirectional reflection, and is only related to the type of ground object. It is a physical quantity reflecting the essential attributes of ground objects, so it can be used as the basis for classifying them. In this paper, the scattering of various ground objects are measured under ground-based imaging conditions, the measurement process of scattering weights is described in detail and the fitting ability of the kernel-driven model is verified. Through comparison, it is found that the scattering coefficients of different ground objects have great differences, and then the method of using scattering coefficients to classify ground objects is put forward.In this paper, two sets of data are used to verify the classification method, and three similarity measures, including projection, distance and amount of information, are used to measure the classification results quantitatively.The experimental results show that the scattering coefficients of the same ground object are almost the same, which has nothing to do with the imaging conditions but only with the type of ground object. The scattering coefficients of different ground objects are different, and using scattering coefficients can effectively achieve the classification of ground objects, and good results have been achieved.

White LEDs emit white light at room temperature after the rare-earth-doped fluorescent powder is excited by blue light chips or ultraviolet chips. The realization of the photoluminescence phenomenon develops a new type of all-solid-state lighting source that is lauded as the fourth-generation lighting source due to its energy-saving, environmental protection, and green lighting advantages. For modern-facility agriculture, blue light between 480 and 500 nm regulates plant rhythm, which is beneficial to plant growth. Blue light plays an important role in the light form and green plant photosynthesis. Green plants capture sunlight for photosynthesis by chlorophyll, carotenoids, lutein, and phytochrome, and LED lights suitable for plant growth can improve the efficiency of photosynthesis. However, it is difficult for traditional light sources to adjust the wavelength of light due to their light quality. In this case, ultraviolet light below 380 nm in the solar spectrum needs to be converted into blue light to improve crops’ light efficiency. Therefore, blue phosphor, with high light efficiency and high thermal stability, has become an important material in full-spectrum lighting and photobiological agriculture. Blue fluorescent materials play an important role in manufacturing white light-emitting diodes (W-LEDs) excited by near-ultraviolet (NUV) chips. This paper uses a high-temperature solid-phase method to prepare YVO4∶Tm3+ blue phosphor. X-ray diffractometer, scanning electron microscope, fluorescence spectrometer and other detection methods are used to characterize and analyze the samples’ phase structure, apparent morphology and luminescence properties. The results show that YVO4∶Tm3+ blue phosphor can be prepared by high-temperature solid-phase method calcination at 1 100 ℃ for 2 h. The powder is about 2 μm spherical, the excitation peak is in the 319 nm ultraviolet region, and the emission peak is in the 479 nm blue region. The color coordinates of the sample are located at (0.104 4, 0.122 4), a blue phosphor that is expected to be applied to white LEDs.

Protein glycation reaction based on traditional heating methods has disadvantages, such as long reaction time consumption and easy to produce harmful advanced glycation end products. In this study, the ovomucoid (OVM)-reducing sugar (1∶0.03, W/W) system served as the research object, effects of superheated steam, a new high-temperature treatment technology, on the OVM glycation and protein structure were studied. The results showed that the glycation of OVM was induced by a short time (1~3 min) of superheated steam treatment (110 and 120 ℃), and the sequence of glycation reaction activity of different sugars was ribose (pentose) > glucose (hexose) > lactose (disaccharide). After glycation, the free amino content of protein decreased from 19.97 mg·mL-1 to the minimum of 2.93, 5.04 and 6.69 mg·mL-1, respectively. The ultraviolet spectra of OVM glycation products under different processing conditions showed no significant changes in wave displacement. However, the maximum absorption had some changes, with most of them were reduced, indicating OVM globulin molecular structure was altered, and the chromophore groups (phenylalanine and tyrosine) of OVM were masked by the reducing sugars. The effect was most significant under ribose glycation. After glycation, the fluorescence intensities were significantly decreased, in which the ribose glycation decreased the most, followed by glucose and lactose, indicating that the fluorescence quenching occurred, which resulted from the folding of exposed fluorophore in protein. In addition, reducing sugar as a quenching agent penetrated the protein framework and reacted with fluorophore, inhibiting the fluorescence emission. Fourier transform infrared spectroscopy showed that, after superheated steam treatment, the protein peak at 3 300 cm-1 became narrow, indicating that the content of N-H decreased or a reaction occurred, thus reducing the stretching vibration frequency of N-H. Some fingerprint peaks in the vicinity of 2 000～2 500 and 500 cm-1appeared, indicating that the protein degradation or glycation middle product formation occurred. MALDI TOF MS analysis showed that there were about 6 glycation reaction sites in each the OVM-ribose system and OVM-glucose system, and about 2 glycation reaction sites in the OVM-lactose system. Moreover, after glycation, OVM dimer, trimer and multimer were formed. Ribose and glucose produced the most significant polymers with OVM, while lactose produced fewer polymers with OVM. This study can provide technical and theoretical guidance for the glycation reaction featured by trace amounts of sugar participation, short time consumption, and application of superheated steam in food processing.

Tea, with much nutrition, is one of the most popular drinks in the world. Good and bad tea are mixed at the market, so it is difficult to make a classification among them. Therefore, using a fast and accurate method to identify tea varieties is meaningful. Most chemical compound’s fundamental frequency absorption bands are within the wavelength range of 2 500~25 000 nm (Mid-infrared region). Large amounts of feature discriminant information in the mid-infrared spectra of tea can be applied to classify tea varieties. This paper proposed a fuzzy covariance the learning vector quantization (FCLVQ) based on the Gustafson-Kessel (GK) clustering. It introduces learning rate of learning vector quantization (LVQ) to control the update rate of cluster centers. Combined with mid-infrared spectroscopy, FCLVQ realizes fast and accurate identification of tea varieties by iteratively calculating the fuzzy membership values and fuzzy clustering centers of samples. Three different kinds of tea(i. e. Emeishan tea, high-quality bamboo-leaf-green tea and low-quality bamboo-leaf-green tea) were selected as 96 samples in total at the market. Each variety corresponds to one group, which consists of 32 samples. The Fourier mid-infrared spectra were collected using an FTIR-7600 spectrometer, and the average spectral data were computed as the final experimental spectra. Firstly, the original spectral data contained noise data, so they were pretreated with multiplicative scattering correction(MSC) to reduce noise. Secondly, principal component analysis(PCA) was employed to reduce the dimensionality of data from 1 868 to 14, and the cumulative contribution of the 14 principal components was 99.74%.Thirdly, the dimensionality of the processed data was reduced to 2, and the discriminant information was extracted by linear discriminant analysis(LDA). Finally, fuzzy C-means clustering(FCM) was run to get initial cluster centers for FCLVQ. The experimental results showed that when the weight index m=2, the accuracy rate of FCLVQ was 95.25%. On the condition of m=2, for the same spectra, the classification accuracy rates of FCM, GK and fuzzy Kohonen clustering(FKCN) were 90.91%, 92.41% and 90.91% respectively. The experimental results showed that compared with the other three algorithms, FCLVQ had a better classification accuracy when m=2 and the number of principal components were 14. Thus, it can be used to classify different tea varieties.

Ambient air was severely polluted due to the elevated level of PM2.5 via burning fireworks. The custom of burning fireworks is prevalent on important occasions around the world. However, there is only a little research concerning heavy metals in PM2.5 from fireworks. In this work, the grade of air quality, potential ecological and health risks of heavy metals in PM2.5 from fireworks in Beijing were evaluated by Inductively Coupled Plasma-Mass Spectrometry (ICP-MS) combined with Geo-accumulation index, potential ecological risk index, and health risk assessment model. A sample was first digested with a purified nitric acid solution at 120 ℃ for 2 hours. The solution volume was precisely adjusted with pure water. Finally, the sample was measured with an ICP-MS. The tested elements included As, Ba, Cd, Co, Cr, Cu, Fe, Mn, Ni, Pb, V, and Zn. The Rh was added to the sample as the standard internal element to control the instrument signal drift and attenuation. The results show that the 24-hour average concentration of PM2.5 from fireworks during the fireworks discharge period is (93±117) μg·m-3, which is higher than China’s ambient air quality standards. Cu, Cr, Ba, Pb, Zn, and As levels in PM2.5 are 17.8, 16.6, 8.1, 5.8, 1.8, and 1.3 times higher than the background concentrations, respectively. They are influenced significantly by fireworks. The order of geo-accumulation index of heavy metals in PM2.5 from fireworks is Cu>Pb>Cd>Zn>As>Ba>Cr, Mn, Ni, Co, V, Fe. The pollution of heavy metals in PM2.5 from fireworks is mainly Cu, Pb, and Cd. The As, Zn, and Ba may are caused moderate pollution. The pollution degrees of other heavy metals are lower than moderate pollution in this work. The pollution caused by fireworks is severed from 20:00 on New Year’s Eve to 8:00 on the Spring Festival. The pollution degrees are all higher than the moderate level in this period. They are strongly to extremely contaminated from 22:00 on New Year’s Eve to 5:00 on the Spring Festival. The order of ecological risk index of heavy metals in PM2.5 from fireworks is Cd>Cu>Pb>As>Zn, Cr, Co, Ni, Mn, V. The potential ecological risks mainly stem from Cd, Cu, and Pb. The levels of potential ecological risks from 22:00 on New Year’s Eve to 5:00 on the Spring Festival are higher than moderate. The levels of potential ecological risks are up to very high from 00:00 to 4:00 on the Spring Festival. The health risk assessment results show that the values of HI and ILCR are less than 1. It indicates that the non-carcinogenic and carcinogenic risk of heavy metals in PM2.5 from fireworks to humans can be ignored.

Chemical Oxygen Demand (COD) is a commonly used water quality indicator in water pollution monitoring. Traditional collection methods are time-consuming and labor-consuming, but the inversion of COD concentration by remote sensing method can quickly obtain the spatial distribution of COD concentration in the whole water area, which is of great significance for water pollution control and water environment protection.Using multi-spectral remote sensing data inversion of COD concentration is low precision. Because at present, a lot of the inversion models based on the Pearson correlation coefficient index selection experience method, modeling band for multi-spectral remote sensing data, its wide spectral bands, and band combination of quantity is limited, hard to find effective variables as modeling.In order to solve this problem, this study in Zhengzhou city, lake as an example, based on the Planet multi-spectral high-resolution remote sensing image and the remote sensing image preprocessing and hyperspectral data for analysis of water samples, using convolution neural network method to inversion of days lake COD concentration. At the same time, choose the single variable regression model, a multivariate regression model accuracy comparison test. The main conclusions are as follows:(1) Compared with the inversion method using Pearson correlation coefficient as the measurement standard to select different band combinations, convolutional neural network inversion has higher spatial inversion accuracy, with the determination coefficient of 0.89 and RMSE of 2.22 mg·L-1. This is because a convolutional neural network not only makes full use of the spectral characteristics of its remote sensing images. Moreover, the spatial information of the domain around the target pixel can be extracted. The abstract features of the deep layer of the image, as well as the"internal law" between the water quality parameter concentration and remote sensing data, can be learned, which can avoid the instability caused by the traditional modeling method to a certain extent. (2) Select the optimal convolutional neural network model to make the thematic map of the spatial distribution of COD concentration in Tiande Lake water quality. Tiande Lake has typical spectral characteristics of inland water, and its spatial distribution of COD concentration is generally characterized by high in the west, low in the east, low in the southeast inlet and high in the northeast outlet.The average value of concentration in the Tiande Lake region retrieved by the convolutional neural network is 23.96 mg·L-1, the standard deviation is 7.11 mg·L-1, and the coefficient of variation is 0.29, which is closer to the statistical value of actual sampling points.The results of COD retrieval based on a convolutional neural network model and multi-spectral image show that the convolutional neural network has good application potential in remote sensing COD retrieval of water quality parameters.

In order to explore the influence of fault tectonism on the characteristics of the chemical structure of coal and the characteristics of the microcrystalline structure. The primary coal and fault tectonic coal of the Hongqingliang mine and Duanwang mine were analyzed by Fourier Transform Infrared Spectroscopy FTIR and X-ray diffraction XRD. The following results showed that because of the influence of fault tectonism, the contents of benzene ring disubstituted and benzene ring tetrasubstitution in Hongqingliang and Duanwang fault tectonic of coal were significantly higher than those in primary coal, the contents of benzene ring trisubstitution and benzene ring pentasubstitution were lower than those in primary coal, and the contents of ether bond C-O-C stretching vibration, methylene -CH2 antisymmetric bending vibration, both methyl, methylene and aromatic ring substituted by hydroxyl -CH3 and -CH2 in fault tectonic of coal were higher than those in primary coal. In comparison, the contents of the polar bond of ether vibration, carbon-carbon double bond, methine and phenol of aromatic hydrocarbon in fault tectonic of coal were lower than those in primary coal. The aromaticity fa of fault tectonic of coal in Hongqingliang and Duanwang mine was 1.01 and 1.03 times that of primary coal, the aromatic cyclocondensation DOC was 1.01 and 3.7 times that of primary coal, the ratio of CH2 to CH3 was 0.933 and 0.94 times that of primary coal, and the aromaticity I was 1.01 and 1.34 times that of primary coal, respectively. The conclusions showed that fault tectonism influenced promoting the shedding of functional groups and fat chains, increasing the degree of polycondensation of coal, reducing the length of fat side chains in coal and increasing the content of aromatic structure. The XRD test results showed that the primary coal and the fault tectonic coal had similar mineral components, and the fault tectonism did not significantly change the types of mineral components in coal. There was a (002) characteristic band near the diffraction angle 2θ=26°, which indicated the presence of a small amount of layered graphite structure, while there was no displayed zone, which indicated the basal low growth level of graphite structure in coal. Compared with the primary coal, the value of the variation range of intergranular spacing d002 of the fault tectonic of coal in Hongqingliang and Duanwang mine was very small. The value of the microcrystalline stacking height Lc was 0.904 5 times and 0.902 7 times of the primary coal respectively, the value of the aromaticity fa-XRD was 1.143 9 times and 1.066 9 times of the primary coal respectively, and the value of the average layer Nave of the crystal stacking was 0.909 45 times and 0.923 56 times of the primary coal respectively. Because of the fault tectonism, the stacking degree of aromatic lamellae of coal reduced, the aromaticity of coal increased, and the stacking layers of coal microcrystalline structural units reduced, indicating that the fault tectonism promoted the transformation from non-aromatic compounds to aromatic compounds in coal. The aromaticity and maturity of the fault tectonic coal were higher, but the combustion reactivity of fault tectonic coal was weaker than that of primary coal.